commit 34f675953290fb70a87e54d8c0f0643b355629e6 Author: lifegpc Date: Tue Mar 11 12:47:55 2025 +0800 Add files diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..d7b4b59 --- /dev/null +++ b/.gitignore @@ -0,0 +1,3 @@ +build/ +.vscode/ +*.zip diff --git a/.gitmodules b/.gitmodules new file mode 100644 index 0000000..6576c53 --- /dev/null +++ b/.gitmodules @@ -0,0 +1,6 @@ +[submodule "libzip"] + path = libzip + url = https://github.com/nih-at/libzip +[submodule "utils"] + path = utils + url = https://github.com/lifegpc/c-utils diff --git a/CMakeLists.txt b/CMakeLists.txt new file mode 100644 index 0000000..f40fcb8 --- /dev/null +++ b/CMakeLists.txt @@ -0,0 +1,47 @@ +cmake_minimum_required(VERSION 3.20) + +project(jeweha_patch) + +if (MSVC) + add_compile_options(/utf-8) +endif() + +include_directories("${CMAKE_CURRENT_SOURCE_DIR}/include") + +set(DETOURS_LIB "${CMAKE_CURRENT_SOURCE_DIR}/lib/detours.lib") + +set(ENABLE_ICONV OFF CACHE BOOL "Libiconv is not needed.") +add_subdirectory(utils) +include_directories("${CMAKE_CURRENT_SOURCE_DIR}/utils") + +set(ENABLE_COMMONCRYPTO OFF CACHE BOOL "CommonCrypto is not needed.") +set(ENABLE_GNUTLS OFF CACHE BOOL "GnuTLS is not needed.") +set(ENABLE_MBEDTLS OFF CACHE BOOL "MbedTLS is not needed.") +set(ENABLE_OPENSSL OFF CACHE BOOL "OpenSSL is not needed.") +set(ENABLE_WINDOWS_CRYPTO OFF CACHE BOOL "Schannel is not needed.") + +set(ENABLE_BZIP2 OFF CACHE BOOL "Bzip2 is not needed.") +set(ENABLE_LZMA OFF CACHE BOOL "Lzma is not needed.") +set(ENABLE_ZSTD OFF CACHE BOOL "Zstd is not needed.") + +set(BUILD_TOOLS OFF CACHE BOOL "Tools are not needed.") +set(BUILD_REGRESS OFF CACHE BOOL "Tests are not needed.") +set(BUILD_OSSFUZZ OFF CACHE BOOL "Fuzz tests are not needed.") +set(BUILD_EXAMPLES OFF CACHE BOOL "Examples are not needed.") +set(BUILD_DOC OFF CACHE BOOL "Documentation is not needed.") + +set(BUILD_SHARED_LIBS OFF CACHE BOOL "Static library is needed.") +set(LIBZIP_DO_INSTALL OFF CACHE BOOL "Installation is not needed.") +set(ZLIB_ROOT "${CMAKE_CURRENT_SOURCE_DIR}" CACHE PATH "Zlib is needed.") + +find_package(ZLIB REQUIRED) +add_subdirectory("libzip") + +add_library(jeweha_patch SHARED dllmain.cpp config.hpp config.cpp vfs.hpp vfs.cpp string_replace_file.hpp string_replace_file.cpp dll_winres.rc) +target_link_libraries(jeweha_patch "${DETOURS_LIB}") +target_link_libraries(jeweha_patch utils) +target_link_libraries(jeweha_patch zip) +set_target_properties(jeweha_patch PROPERTIES OUTPUT_NAME "jeweha_CHS_1.0") + +add_executable(jeweha-chs WIN32 main.cpp winres.rc jeweha-chs.exe.manifest) +set_target_properties(jeweha-chs PROPERTIES OUTPUT_NAME "jeweha_CHS_1.0") diff --git a/ICON.ico b/ICON.ico new file mode 100644 index 0000000..0060689 Binary files /dev/null and b/ICON.ico differ diff --git a/README.md b/README.md new file mode 100644 index 0000000..7b55c30 --- /dev/null +++ b/README.md @@ -0,0 +1,7 @@ +# Jeweha Chinese Patch +## How to Compile +```powershell +md build && cd build +cmake -A Win32 ../ +cmake --build . --config Release +``` diff --git a/config.cpp b/config.cpp new file mode 100644 index 0000000..fc84d6b --- /dev/null +++ b/config.cpp @@ -0,0 +1,48 @@ +#include "config.hpp" +#include "fileop.h" +#include "file_reader.h" +#include "malloc.h" + +#include +#if _WIN32 +#include +#endif + +bool Config::Load(std::string path) { + if (!fileop::exists(path)) { + return false; + } + int fd = 0; + int err = fileop::open(path, fd, O_RDONLY, _SH_DENYRW, _S_IREAD); + if (err < 0) { + return false; + } + FILE* f = fileop::fdopen(fd, "rb"); + if (!f) { + fileop::close(fd); + return false; + } + auto reader = create_file_reader(f, 0); + char* line = nullptr; + size_t line_size = 0; + while (!file_reader_read_line(reader, &line, &line_size)) { + std::string l(line, line_size); + free(line); + line = nullptr; + line_size = 0; + size_t comment_pos = l.find_first_of('#'); + if (comment_pos != std::string::npos) { + l = l.substr(0, comment_pos); + } + size_t eq_pos = l.find_first_of('='); + if (eq_pos == std::string::npos) { + continue; + } + std::string key = l.substr(0, eq_pos); + std::string value = l.substr(eq_pos + 1); + configs[key] = value; + } + free_file_reader(reader); + fileop::fclose(f); + return true; +} diff --git a/config.hpp b/config.hpp new file mode 100644 index 0000000..fd11bec --- /dev/null +++ b/config.hpp @@ -0,0 +1,12 @@ +#include +#include + +class Config { + public: + std::unordered_map configs; + Config() { + configs["defaultFont"] = "微软雅黑"; + configs["stringReplaceFile"] = ""; + } + bool Load(std::string path); +}; diff --git a/dll_winres.rc b/dll_winres.rc new file mode 100644 index 0000000..c6db6ca --- /dev/null +++ b/dll_winres.rc @@ -0,0 +1 @@ +114514 RCDATA "embbed.zip" diff --git a/dllmain.cpp b/dllmain.cpp new file mode 100644 index 0000000..be996ba --- /dev/null +++ b/dllmain.cpp @@ -0,0 +1,212 @@ +#include +#include "config.hpp" +#include "detours.h" +#include +#include "wchar_util.h" +#include "vfs.hpp" +#include "str_util.h" +#include "fileop.h" +#include +#include "string_replace_file.hpp" + +static HFONT(WINAPI *TrueCreateFontW)(int nHeight, int nWidth, int nEscapement, int nOrientation, int fnWeight, DWORD dwItalic, DWORD dwUnderline, DWORD dwStrikeOut, DWORD dwCharSet, DWORD dwOutPrecision, DWORD dwClipPrecision, DWORD dwQuality, DWORD dwPitchAndFamily, LPCWSTR lpFaceName) = CreateFontW; +static HFONT(WINAPI *TrueCreateFontA)(int nHeight, int nWidth, int nEscapement, int nOrientation, int fnWeight, DWORD dwItalic, DWORD dwUnderline, DWORD dwStrikeOut, DWORD dwCharSet, DWORD dwOutPrecision, DWORD dwClipPrecision, DWORD dwQuality, DWORD dwPitchAndFamily, LPCSTR lpFaceName) = CreateFontA; +static HANDLE(WINAPI *TrueCreateFileW)(LPCWSTR lpFileName, DWORD dwDesiredAccess, DWORD dwShareMode, LPSECURITY_ATTRIBUTES lpSecurityAttributes, DWORD dwCreationDisposition, DWORD dwFlagsAndAttributes, HANDLE hTemplateFile) = CreateFileW; +static BOOL(WINAPI *TrueReadFile)(HANDLE hFile, LPVOID lpBuffer, DWORD nNumberOfBytesToRead, LPDWORD lpNumberOfBytesRead, LPOVERLAPPED lpOverlapped) = ReadFile; +static BOOL(WINAPI *TrueCloseHandle)(HANDLE hObject) = CloseHandle; +static DWORD(WINAPI *TrueGetFileSize)(HANDLE hFile, LPDWORD lpFileSizeHigh) = GetFileSize; +static decltype(GetFileSizeEx) *TrueGetFileSizeEx = GetFileSizeEx; +static DWORD(WINAPI *TrueSetFilePointer)(HANDLE hFile, LONG lDistanceToMove, PLONG lpDistanceToMoveHigh, DWORD dwMoveMethod) = SetFilePointer; + +static Config config; +static std::wstring defaultFont; +static VFS vfs; +static StringReplaceFile replaceFile; +static HMODULE hDll = NULL; + +char* to_utf8(char* target, const char* source, UINT cp) { + int count = MultiByteToWideChar(cp, MB_ERR_INVALID_CHARS, source, -1, NULL, 0); + if (!count) return nullptr; + WCHAR* ws = new WCHAR[count + 1]; + MultiByteToWideChar(cp, MB_ERR_INVALID_CHARS, source, -1, ws, count); + char* result = nullptr; + int ncount = WideCharToMultiByte(CP_UTF8, 0, ws, -1, nullptr, 0, nullptr, nullptr); + if (ncount) { + if (!target) { + target = new char[ncount + 1]; + } + result = target; + WideCharToMultiByte(CP_UTF8, 0, ws, -1, result, ncount, nullptr, nullptr); + } + delete[] ws; + return result; +} + +char* jis_to_utf8(char* target, const char* source) { + char* result = to_utf8(target, source, CP_UTF8); + if (!result) { + result = to_utf8(target, source, 932); + } + if (!replaceFile.messages.empty() && result) { + std::string str(result); + auto re = replaceFile.messages.find(str); + if (re != replaceFile.messages.end()) { + str = (*re).second; + if (target) { + strcpy(target, str.c_str()); + result = target; + } else { + delete[] result; + result = new char[str.size() + 1]; + strcpy(result, str.c_str()); + } + } + } + return result; +} + +PVOID GetHandle() { + HMODULE hModule = GetModuleHandleA(NULL); + return (char*)hModule + 0xa3f60; +} + +static PVOID h = nullptr; + +HFONT WINAPI HookedCreateFontW(int nHeight, int nWidth, int nEscapement, int nOrientation, int fnWeight, DWORD dwItalic, DWORD dwUnderline, DWORD dwStrikeOut, DWORD dwCharSet, DWORD dwOutPrecision, DWORD dwClipPrecision, DWORD dwQuality, DWORD dwPitchAndFamily, LPCWSTR lpFaceName) { + std::wstring name(lpFaceName); + if (name == L"Meiryo") { + lpFaceName = defaultFont.c_str(); + } + return TrueCreateFontW(nHeight, nWidth, nEscapement, nOrientation, fnWeight, dwItalic, dwUnderline, dwStrikeOut, dwCharSet, dwOutPrecision, dwClipPrecision, dwQuality, dwPitchAndFamily, lpFaceName); +} + +HFONT WINAPI HookedCreateFontA(int nHeight, int nWidth, int nEscapement, int nOrientation, int fnWeight, DWORD dwItalic, DWORD dwUnderline, DWORD dwStrikeOut, DWORD dwCharSet, DWORD dwOutPrecision, DWORD dwClipPrecision, DWORD dwQuality, DWORD dwPitchAndFamily, LPCSTR lpFaceName) { + UINT cp[] = { CP_UTF8, CP_OEMCP, CP_ACP, 932 }; + std::wstring font; + for (int i = 0; i < 4; i++) { + if (wchar_util::str_to_wstr(font, lpFaceName, cp[i])) { + if (font == L"Meiryo") { + font = defaultFont; + } + return TrueCreateFontW(nHeight, nWidth, nEscapement, nOrientation, fnWeight, dwItalic, dwUnderline, dwStrikeOut, dwCharSet, dwOutPrecision, dwClipPrecision, dwQuality, dwPitchAndFamily, font.c_str()); + } + } + if (!strcmp(lpFaceName, "Meiryo")) { + lpFaceName = "Microsoft YaHei"; + } + return TrueCreateFontA(nHeight, nWidth, nEscapement, nOrientation, fnWeight, dwItalic, dwUnderline, dwStrikeOut, dwCharSet, dwOutPrecision, dwClipPrecision, dwQuality, dwPitchAndFamily, lpFaceName); +} + +HANDLE WINAPI HookedCreateFileW(LPCWSTR lpFileName, DWORD dwDesiredAccess, DWORD dwShareMode, LPSECURITY_ATTRIBUTES lpSecurityAttributes, DWORD dwCreationDisposition, DWORD dwFlagsAndAttributes, HANDLE hTemplateFile) { + if (vfs.ContainsFile(lpFileName)) { + return vfs.CreateFileW(lpFileName); + } + return TrueCreateFileW(lpFileName, dwDesiredAccess, dwShareMode, lpSecurityAttributes, dwCreationDisposition, dwFlagsAndAttributes, hTemplateFile); +} + +BOOL WINAPI HookedReadFile(HANDLE hFile, LPVOID lpBuffer, DWORD nNumberOfBytesToRead, LPDWORD lpNumberOfBytesRead, LPOVERLAPPED lpOverlapped) { + if (vfs.ContainsHandle(hFile)) { + if (lpOverlapped) { + SetLastError(ERROR_INVALID_PARAMETER); + return FALSE; + } + return vfs.ReadFile(hFile, lpBuffer, nNumberOfBytesToRead, lpNumberOfBytesRead); + } + return TrueReadFile(hFile, lpBuffer, nNumberOfBytesToRead, lpNumberOfBytesRead, lpOverlapped); +} + +BOOL WINAPI HookedCloseHandle(HANDLE hObject) { + if (vfs.ContainsHandle(hObject)) { + vfs.CloseHandle(hObject); + return TRUE; + } + return TrueCloseHandle(hObject); +} + +DWORD WINAPI HookedGetFileSize(HANDLE hFile, LPDWORD lpFileSizeHigh) { + if (vfs.ContainsHandle(hFile)) { + return vfs.GetFileSize(hFile, lpFileSizeHigh); + } + return TrueGetFileSize(hFile, lpFileSizeHigh); +} + +BOOL WINAPI HookedGetFileSizeEx(HANDLE hFile, PLARGE_INTEGER lpFileSize) { + if (vfs.ContainsHandle(hFile)) { + return vfs.GetFileSizeEx(hFile, lpFileSize); + } + return TrueGetFileSizeEx(hFile, lpFileSize); +} + +DWORD WINAPI HookedSetFilePointer(HANDLE hFile, LONG lDistanceToMove, PLONG lpDistanceToMoveHigh, DWORD dwMoveMethod) { + if (vfs.ContainsHandle(hFile)) { + return vfs.SetFilePointer(hFile, lDistanceToMove, lpDistanceToMoveHigh, dwMoveMethod); + } + return TrueSetFilePointer(hFile, lDistanceToMove, lpDistanceToMoveHigh, dwMoveMethod); +} + +extern "C" __declspec(dllexport) void Attach() { + config.Load("config.txt"); + if (!wchar_util::str_to_wstr(defaultFont, config.configs["defaultFont"], CP_UTF8)) { + defaultFont = L"微软雅黑"; + } + if (defaultFont.empty()) { + defaultFont = L"微软雅黑"; + } + vfs.AddArchiveFromResourceWithErrorMsg(hDll, 114514); + vfs.AddArchiveWithErrorMsg("evimage.xp3"); + vfs.AddArchiveWithErrorMsg("scn.xp3"); + vfs.AddArchiveWithErrorMsg("uipsd.xp3"); + vfs.AddArchiveWithErrorMsg("video.xp3"); + DetourTransactionBegin(); + DetourUpdateThread(GetCurrentThread()); + h = GetHandle(); + // DetourAttach(&h, (PVOID)jis_to_utf8); + DetourAttach(&TrueCreateFontW, HookedCreateFontW); + DetourAttach(&TrueCreateFontA, HookedCreateFontA); + DetourAttach(&TrueCreateFileW, HookedCreateFileW); + DetourAttach(&TrueReadFile, HookedReadFile); + DetourAttach(&TrueCloseHandle, HookedCloseHandle); + DetourAttach(&TrueGetFileSize, HookedGetFileSize); + DetourAttach(&TrueGetFileSizeEx, HookedGetFileSizeEx); + DetourAttach(&TrueSetFilePointer, HookedSetFilePointer); + DetourTransactionCommit(); + std::string stringReplaceFile = config.configs["stringReplaceFile"]; + if (!stringReplaceFile.empty()) { + if (!replaceFile.Load(stringReplaceFile)) { + MessageBoxW(NULL, L"无法加载文本替换文件。", L"错误", MB_ICONERROR); + } + } +#if _DEBUG + while( !::IsDebuggerPresent() ) + ::Sleep( 1000 ); +#endif +} + +extern "C" __declspec(dllexport) void Detach() { + if (!h) return; + DetourTransactionBegin(); + DetourUpdateThread(GetCurrentThread()); + // DetourDetach(&h, (PVOID)jis_to_utf8); + DetourDetach(&TrueCreateFontW, HookedCreateFontW); + DetourDetach(&TrueCreateFontA, HookedCreateFontA); + DetourDetach(&TrueCreateFileW, HookedCreateFileW); + DetourDetach(&TrueReadFile, HookedReadFile); + DetourDetach(&TrueCloseHandle, HookedCloseHandle); + DetourDetach(&TrueGetFileSize, HookedGetFileSize); + DetourDetach(&TrueGetFileSizeEx, HookedGetFileSizeEx); + DetourDetach(&TrueSetFilePointer, HookedSetFilePointer); + DetourTransactionCommit(); +} + +BOOL APIENTRY DllMain(HMODULE hModule, DWORD reason, LPVOID rev) { + switch (reason) { + case DLL_PROCESS_ATTACH: + hDll = hModule; + Attach(); + break; + case DLL_PROCESS_DETACH: + Detach(); + break; + } + return TRUE; +} diff --git a/include/detours.h b/include/detours.h new file mode 100644 index 0000000..4b9e232 --- /dev/null +++ b/include/detours.h @@ -0,0 +1,1233 @@ +///////////////////////////////////////////////////////////////////////////// +// +// Core Detours Functionality (detours.h of detours.lib) +// +// Microsoft Research Detours Package, Version 4.0.1 +// +// Copyright (c) Microsoft Corporation. All rights reserved. +// + +#pragma once +#ifndef _DETOURS_H_ +#define _DETOURS_H_ + +#define DETOURS_VERSION 0x4c0c1 // 0xMAJORcMINORcPATCH + +////////////////////////////////////////////////////////////////////////////// +// + +#ifdef DETOURS_INTERNAL + +#define _CRT_STDIO_ARBITRARY_WIDE_SPECIFIERS 1 +#define _ARM_WINAPI_PARTITION_DESKTOP_SDK_AVAILABLE 1 + +#pragma warning(disable:4068) // unknown pragma (suppress) + +#if _MSC_VER >= 1900 +#pragma warning(push) +#pragma warning(disable:4091) // empty typedef +#endif + +// Suppress declspec(dllimport) for the sake of Detours +// users that provide kernel32 functionality themselves. +// This is ok in the mainstream case, it will just cost +// an extra instruction calling some functions, which +// LTCG optimizes away. +// +#define _KERNEL32_ 1 +#define _USER32_ 1 + +#include +#if (_MSC_VER < 1310) +#else +#pragma warning(push) +#if _MSC_VER > 1400 +#pragma warning(disable:6102 6103) // /analyze warnings +#endif +#include +#include +#pragma warning(pop) +#endif +#include + +// Allow Detours to cleanly compile with the MingW toolchain. +// +#ifdef __GNUC__ +#define __try +#define __except(x) if (0) +#include +#include +#endif + +// From winerror.h, as this error isn't found in some SDKs: +// +// MessageId: ERROR_DYNAMIC_CODE_BLOCKED +// +// MessageText: +// +// The operation was blocked as the process prohibits dynamic code generation. +// +#define ERROR_DYNAMIC_CODE_BLOCKED 1655L + +#endif // DETOURS_INTERNAL + +////////////////////////////////////////////////////////////////////////////// +// + +#undef DETOURS_X64 +#undef DETOURS_X86 +#undef DETOURS_IA64 +#undef DETOURS_ARM +#undef DETOURS_ARM64 +#undef DETOURS_BITS +#undef DETOURS_32BIT +#undef DETOURS_64BIT + +#if defined(_X86_) +#define DETOURS_X86 +#define DETOURS_OPTION_BITS 64 + +#elif defined(_AMD64_) +#define DETOURS_X64 +#define DETOURS_OPTION_BITS 32 + +#elif defined(_IA64_) +#define DETOURS_IA64 +#define DETOURS_OPTION_BITS 32 + +#elif defined(_ARM_) +#define DETOURS_ARM + +#elif defined(_ARM64_) +#define DETOURS_ARM64 + +#else +#error Unknown architecture (x86, amd64, ia64, arm, arm64) +#endif + +#ifdef _WIN64 +#undef DETOURS_32BIT +#define DETOURS_64BIT 1 +#define DETOURS_BITS 64 +// If all 64bit kernels can run one and only one 32bit architecture. +//#define DETOURS_OPTION_BITS 32 +#else +#define DETOURS_32BIT 1 +#undef DETOURS_64BIT +#define DETOURS_BITS 32 +// If all 64bit kernels can run one and only one 32bit architecture. +//#define DETOURS_OPTION_BITS 32 +#endif + +/////////////////////////////////////////////////////////////// Helper Macros. +// +#define DETOURS_STRINGIFY_(x) #x +#define DETOURS_STRINGIFY(x) DETOURS_STRINGIFY_(x) + +#define VER_DETOURS_BITS DETOURS_STRINGIFY(DETOURS_BITS) + +////////////////////////////////////////////////////////////////////////////// +// + +#if (_MSC_VER < 1299) && !defined(__MINGW32__) +typedef LONG LONG_PTR; +typedef ULONG ULONG_PTR; +#endif + +///////////////////////////////////////////////// SAL 2.0 Annotations w/o SAL. +// +// These definitions are include so that Detours will build even if the +// compiler doesn't have full SAL 2.0 support. +// +#ifndef DETOURS_DONT_REMOVE_SAL_20 + +#ifdef DETOURS_TEST_REMOVE_SAL_20 +#undef _Analysis_assume_ +#undef _Benign_race_begin_ +#undef _Benign_race_end_ +#undef _Field_range_ +#undef _Field_size_ +#undef _In_ +#undef _In_bytecount_ +#undef _In_count_ +#undef __in_ecount +#undef _In_opt_ +#undef _In_opt_bytecount_ +#undef _In_opt_count_ +#undef _In_opt_z_ +#undef _In_range_ +#undef _In_reads_ +#undef _In_reads_bytes_ +#undef _In_reads_opt_ +#undef _In_reads_opt_bytes_ +#undef _In_reads_or_z_ +#undef _In_z_ +#undef _Inout_ +#undef _Inout_opt_ +#undef _Inout_z_count_ +#undef _Out_ +#undef _Out_opt_ +#undef _Out_writes_ +#undef _Outptr_result_maybenull_ +#undef _Readable_bytes_ +#undef _Success_ +#undef _Writable_bytes_ +#undef _Pre_notnull_ +#endif + +#if defined(_Deref_out_opt_z_) && !defined(_Outptr_result_maybenull_) +#define _Outptr_result_maybenull_ _Deref_out_opt_z_ +#endif + +#if defined(_In_count_) && !defined(_In_reads_) +#define _In_reads_(x) _In_count_(x) +#endif + +#if defined(_In_opt_count_) && !defined(_In_reads_opt_) +#define _In_reads_opt_(x) _In_opt_count_(x) +#endif + +#if defined(_In_opt_bytecount_) && !defined(_In_reads_opt_bytes_) +#define _In_reads_opt_bytes_(x) _In_opt_bytecount_(x) +#endif + +#if defined(_In_bytecount_) && !defined(_In_reads_bytes_) +#define _In_reads_bytes_(x) _In_bytecount_(x) +#endif + +#ifndef _In_ +#define _In_ +#endif + +#ifndef _In_bytecount_ +#define _In_bytecount_(x) +#endif + +#ifndef _In_count_ +#define _In_count_(x) +#endif + +#ifndef __in_ecount +#define __in_ecount(x) +#endif + +#ifndef _In_opt_ +#define _In_opt_ +#endif + +#ifndef _In_opt_bytecount_ +#define _In_opt_bytecount_(x) +#endif + +#ifndef _In_opt_count_ +#define _In_opt_count_(x) +#endif + +#ifndef _In_opt_z_ +#define _In_opt_z_ +#endif + +#ifndef _In_range_ +#define _In_range_(x,y) +#endif + +#ifndef _In_reads_ +#define _In_reads_(x) +#endif + +#ifndef _In_reads_bytes_ +#define _In_reads_bytes_(x) +#endif + +#ifndef _In_reads_opt_ +#define _In_reads_opt_(x) +#endif + +#ifndef _In_reads_opt_bytes_ +#define _In_reads_opt_bytes_(x) +#endif + +#ifndef _In_reads_or_z_ +#define _In_reads_or_z_ +#endif + +#ifndef _In_z_ +#define _In_z_ +#endif + +#ifndef _Inout_ +#define _Inout_ +#endif + +#ifndef _Inout_opt_ +#define _Inout_opt_ +#endif + +#ifndef _Inout_z_count_ +#define _Inout_z_count_(x) +#endif + +#ifndef _Out_ +#define _Out_ +#endif + +#ifndef _Out_opt_ +#define _Out_opt_ +#endif + +#ifndef _Out_writes_ +#define _Out_writes_(x) +#endif + +#ifndef _Outptr_result_maybenull_ +#define _Outptr_result_maybenull_ +#endif + +#ifndef _Writable_bytes_ +#define _Writable_bytes_(x) +#endif + +#ifndef _Readable_bytes_ +#define _Readable_bytes_(x) +#endif + +#ifndef _Success_ +#define _Success_(x) +#endif + +#ifndef _Pre_notnull_ +#define _Pre_notnull_ +#endif + +#ifdef DETOURS_INTERNAL + +#pragma warning(disable:4615) // unknown warning type (suppress with older compilers) + +#ifndef _Benign_race_begin_ +#define _Benign_race_begin_ +#endif + +#ifndef _Benign_race_end_ +#define _Benign_race_end_ +#endif + +#ifndef _Field_size_ +#define _Field_size_(x) +#endif + +#ifndef _Field_range_ +#define _Field_range_(x,y) +#endif + +#ifndef _Analysis_assume_ +#define _Analysis_assume_(x) +#endif + +#endif // DETOURS_INTERNAL +#endif // DETOURS_DONT_REMOVE_SAL_20 + +////////////////////////////////////////////////////////////////////////////// +// +#ifndef GUID_DEFINED +#define GUID_DEFINED +typedef struct _GUID +{ + DWORD Data1; + WORD Data2; + WORD Data3; + BYTE Data4[ 8 ]; +} GUID; + +#ifdef INITGUID +#define DEFINE_GUID(name, l, w1, w2, b1, b2, b3, b4, b5, b6, b7, b8) \ + const GUID name \ + = { l, w1, w2, { b1, b2, b3, b4, b5, b6, b7, b8 } } +#else +#define DEFINE_GUID(name, l, w1, w2, b1, b2, b3, b4, b5, b6, b7, b8) \ + const GUID name +#endif // INITGUID +#endif // !GUID_DEFINED + +#if defined(__cplusplus) +#ifndef _REFGUID_DEFINED +#define _REFGUID_DEFINED +#define REFGUID const GUID & +#endif // !_REFGUID_DEFINED +#else // !__cplusplus +#ifndef _REFGUID_DEFINED +#define _REFGUID_DEFINED +#define REFGUID const GUID * const +#endif // !_REFGUID_DEFINED +#endif // !__cplusplus + +#ifndef ARRAYSIZE +#define ARRAYSIZE(x) (sizeof(x)/sizeof(x[0])) +#endif + +// +////////////////////////////////////////////////////////////////////////////// + +#ifdef __cplusplus +extern "C" { +#endif // __cplusplus + +/////////////////////////////////////////////////// Instruction Target Macros. +// +#define DETOUR_INSTRUCTION_TARGET_NONE ((PVOID)0) +#define DETOUR_INSTRUCTION_TARGET_DYNAMIC ((PVOID)(LONG_PTR)-1) +#define DETOUR_SECTION_HEADER_SIGNATURE 0x00727444 // "Dtr\0" + +extern const GUID DETOUR_EXE_RESTORE_GUID; +extern const GUID DETOUR_EXE_HELPER_GUID; + +#define DETOUR_TRAMPOLINE_SIGNATURE 0x21727444 // Dtr! +typedef struct _DETOUR_TRAMPOLINE DETOUR_TRAMPOLINE, *PDETOUR_TRAMPOLINE; + +#ifndef DETOUR_MAX_SUPPORTED_IMAGE_SECTION_HEADERS +#define DETOUR_MAX_SUPPORTED_IMAGE_SECTION_HEADERS 32 +#endif // !DETOUR_MAX_SUPPORTED_IMAGE_SECTION_HEADERS + +/////////////////////////////////////////////////////////// Binary Structures. +// +#pragma pack(push, 8) +typedef struct _DETOUR_SECTION_HEADER +{ + DWORD cbHeaderSize; + DWORD nSignature; + DWORD nDataOffset; + DWORD cbDataSize; + + DWORD nOriginalImportVirtualAddress; + DWORD nOriginalImportSize; + DWORD nOriginalBoundImportVirtualAddress; + DWORD nOriginalBoundImportSize; + + DWORD nOriginalIatVirtualAddress; + DWORD nOriginalIatSize; + DWORD nOriginalSizeOfImage; + DWORD cbPrePE; + + DWORD nOriginalClrFlags; + DWORD reserved1; + DWORD reserved2; + DWORD reserved3; + + // Followed by cbPrePE bytes of data. +} DETOUR_SECTION_HEADER, *PDETOUR_SECTION_HEADER; + +typedef struct _DETOUR_SECTION_RECORD +{ + DWORD cbBytes; + DWORD nReserved; + GUID guid; +} DETOUR_SECTION_RECORD, *PDETOUR_SECTION_RECORD; + +typedef struct _DETOUR_CLR_HEADER +{ + // Header versioning + ULONG cb; + USHORT MajorRuntimeVersion; + USHORT MinorRuntimeVersion; + + // Symbol table and startup information + IMAGE_DATA_DIRECTORY MetaData; + ULONG Flags; + + // Followed by the rest of the IMAGE_COR20_HEADER +} DETOUR_CLR_HEADER, *PDETOUR_CLR_HEADER; + +typedef struct _DETOUR_EXE_RESTORE +{ + DWORD cb; + DWORD cbidh; + DWORD cbinh; + DWORD cbclr; + + PBYTE pidh; + PBYTE pinh; + PBYTE pclr; + + IMAGE_DOS_HEADER idh; + union { + IMAGE_NT_HEADERS inh; // all environments have this +#ifdef IMAGE_NT_OPTIONAL_HDR32_MAGIC // some environments do not have this + IMAGE_NT_HEADERS32 inh32; +#endif +#ifdef IMAGE_NT_OPTIONAL_HDR64_MAGIC // some environments do not have this + IMAGE_NT_HEADERS64 inh64; +#endif +#ifdef IMAGE_NT_OPTIONAL_HDR64_MAGIC // some environments do not have this + BYTE raw[sizeof(IMAGE_NT_HEADERS64) + + sizeof(IMAGE_SECTION_HEADER) * DETOUR_MAX_SUPPORTED_IMAGE_SECTION_HEADERS]; +#else + BYTE raw[0x108 + sizeof(IMAGE_SECTION_HEADER) * DETOUR_MAX_SUPPORTED_IMAGE_SECTION_HEADERS]; +#endif + }; + DETOUR_CLR_HEADER clr; + +} DETOUR_EXE_RESTORE, *PDETOUR_EXE_RESTORE; + +#ifdef IMAGE_NT_OPTIONAL_HDR64_MAGIC +C_ASSERT(sizeof(IMAGE_NT_HEADERS64) == 0x108); +#endif + +// The size can change, but assert for clarity due to the muddying #ifdefs. +#ifdef _WIN64 +C_ASSERT(sizeof(DETOUR_EXE_RESTORE) == 0x688); +#else +C_ASSERT(sizeof(DETOUR_EXE_RESTORE) == 0x678); +#endif + +typedef struct _DETOUR_EXE_HELPER +{ + DWORD cb; + DWORD pid; + DWORD nDlls; + CHAR rDlls[4]; +} DETOUR_EXE_HELPER, *PDETOUR_EXE_HELPER; + +#pragma pack(pop) + +#define DETOUR_SECTION_HEADER_DECLARE(cbSectionSize) \ +{ \ + sizeof(DETOUR_SECTION_HEADER),\ + DETOUR_SECTION_HEADER_SIGNATURE,\ + sizeof(DETOUR_SECTION_HEADER),\ + (cbSectionSize),\ + \ + 0,\ + 0,\ + 0,\ + 0,\ + \ + 0,\ + 0,\ + 0,\ + 0,\ +} + +///////////////////////////////////////////////////////////// Binary Typedefs. +// +typedef BOOL (CALLBACK *PF_DETOUR_BINARY_BYWAY_CALLBACK)( + _In_opt_ PVOID pContext, + _In_opt_ LPCSTR pszFile, + _Outptr_result_maybenull_ LPCSTR *ppszOutFile); + +typedef BOOL (CALLBACK *PF_DETOUR_BINARY_FILE_CALLBACK)( + _In_opt_ PVOID pContext, + _In_ LPCSTR pszOrigFile, + _In_ LPCSTR pszFile, + _Outptr_result_maybenull_ LPCSTR *ppszOutFile); + +typedef BOOL (CALLBACK *PF_DETOUR_BINARY_SYMBOL_CALLBACK)( + _In_opt_ PVOID pContext, + _In_ ULONG nOrigOrdinal, + _In_ ULONG nOrdinal, + _Out_ ULONG *pnOutOrdinal, + _In_opt_ LPCSTR pszOrigSymbol, + _In_opt_ LPCSTR pszSymbol, + _Outptr_result_maybenull_ LPCSTR *ppszOutSymbol); + +typedef BOOL (CALLBACK *PF_DETOUR_BINARY_COMMIT_CALLBACK)( + _In_opt_ PVOID pContext); + +typedef BOOL (CALLBACK *PF_DETOUR_ENUMERATE_EXPORT_CALLBACK)(_In_opt_ PVOID pContext, + _In_ ULONG nOrdinal, + _In_opt_ LPCSTR pszName, + _In_opt_ PVOID pCode); + +typedef BOOL (CALLBACK *PF_DETOUR_IMPORT_FILE_CALLBACK)(_In_opt_ PVOID pContext, + _In_opt_ HMODULE hModule, + _In_opt_ LPCSTR pszFile); + +typedef BOOL (CALLBACK *PF_DETOUR_IMPORT_FUNC_CALLBACK)(_In_opt_ PVOID pContext, + _In_ DWORD nOrdinal, + _In_opt_ LPCSTR pszFunc, + _In_opt_ PVOID pvFunc); + +// Same as PF_DETOUR_IMPORT_FUNC_CALLBACK but extra indirection on last parameter. +typedef BOOL (CALLBACK *PF_DETOUR_IMPORT_FUNC_CALLBACK_EX)(_In_opt_ PVOID pContext, + _In_ DWORD nOrdinal, + _In_opt_ LPCSTR pszFunc, + _In_opt_ PVOID* ppvFunc); + +typedef VOID * PDETOUR_BINARY; +typedef VOID * PDETOUR_LOADED_BINARY; + +//////////////////////////////////////////////////////////// Transaction APIs. +// +LONG WINAPI DetourTransactionBegin(VOID); +LONG WINAPI DetourTransactionAbort(VOID); +LONG WINAPI DetourTransactionCommit(VOID); +LONG WINAPI DetourTransactionCommitEx(_Out_opt_ PVOID **pppFailedPointer); + +LONG WINAPI DetourUpdateThread(_In_ HANDLE hThread); + +LONG WINAPI DetourAttach(_Inout_ PVOID *ppPointer, + _In_ PVOID pDetour); + +LONG WINAPI DetourAttachEx(_Inout_ PVOID *ppPointer, + _In_ PVOID pDetour, + _Out_opt_ PDETOUR_TRAMPOLINE *ppRealTrampoline, + _Out_opt_ PVOID *ppRealTarget, + _Out_opt_ PVOID *ppRealDetour); + +LONG WINAPI DetourDetach(_Inout_ PVOID *ppPointer, + _In_ PVOID pDetour); + +BOOL WINAPI DetourSetIgnoreTooSmall(_In_ BOOL fIgnore); +BOOL WINAPI DetourSetRetainRegions(_In_ BOOL fRetain); +PVOID WINAPI DetourSetSystemRegionLowerBound(_In_ PVOID pSystemRegionLowerBound); +PVOID WINAPI DetourSetSystemRegionUpperBound(_In_ PVOID pSystemRegionUpperBound); + +////////////////////////////////////////////////////////////// Code Functions. +// +PVOID WINAPI DetourFindFunction(_In_ LPCSTR pszModule, + _In_ LPCSTR pszFunction); +PVOID WINAPI DetourCodeFromPointer(_In_ PVOID pPointer, + _Out_opt_ PVOID *ppGlobals); +PVOID WINAPI DetourCopyInstruction(_In_opt_ PVOID pDst, + _Inout_opt_ PVOID *ppDstPool, + _In_ PVOID pSrc, + _Out_opt_ PVOID *ppTarget, + _Out_opt_ LONG *plExtra); +BOOL WINAPI DetourSetCodeModule(_In_ HMODULE hModule, + _In_ BOOL fLimitReferencesToModule); +PVOID WINAPI DetourAllocateRegionWithinJumpBounds(_In_ LPCVOID pbTarget, + _Out_ PDWORD pcbAllocatedSize); +BOOL WINAPI DetourIsFunctionImported(_In_ PBYTE pbCode, + _In_ PBYTE pbAddress); + +///////////////////////////////////////////////////// Loaded Binary Functions. +// +HMODULE WINAPI DetourGetContainingModule(_In_ PVOID pvAddr); +HMODULE WINAPI DetourEnumerateModules(_In_opt_ HMODULE hModuleLast); +PVOID WINAPI DetourGetEntryPoint(_In_opt_ HMODULE hModule); +ULONG WINAPI DetourGetModuleSize(_In_opt_ HMODULE hModule); +BOOL WINAPI DetourEnumerateExports(_In_ HMODULE hModule, + _In_opt_ PVOID pContext, + _In_ PF_DETOUR_ENUMERATE_EXPORT_CALLBACK pfExport); +BOOL WINAPI DetourEnumerateImports(_In_opt_ HMODULE hModule, + _In_opt_ PVOID pContext, + _In_opt_ PF_DETOUR_IMPORT_FILE_CALLBACK pfImportFile, + _In_opt_ PF_DETOUR_IMPORT_FUNC_CALLBACK pfImportFunc); + +BOOL WINAPI DetourEnumerateImportsEx(_In_opt_ HMODULE hModule, + _In_opt_ PVOID pContext, + _In_opt_ PF_DETOUR_IMPORT_FILE_CALLBACK pfImportFile, + _In_opt_ PF_DETOUR_IMPORT_FUNC_CALLBACK_EX pfImportFuncEx); + +_Writable_bytes_(*pcbData) +_Readable_bytes_(*pcbData) +_Success_(return != NULL) +PVOID WINAPI DetourFindPayload(_In_opt_ HMODULE hModule, + _In_ REFGUID rguid, + _Out_opt_ DWORD *pcbData); + +_Writable_bytes_(*pcbData) +_Readable_bytes_(*pcbData) +_Success_(return != NULL) +PVOID WINAPI DetourFindPayloadEx(_In_ REFGUID rguid, + _Out_opt_ DWORD *pcbData); + +DWORD WINAPI DetourGetSizeOfPayloads(_In_opt_ HMODULE hModule); + +BOOL WINAPI DetourFreePayload(_In_ PVOID pvData); +///////////////////////////////////////////////// Persistent Binary Functions. +// + +PDETOUR_BINARY WINAPI DetourBinaryOpen(_In_ HANDLE hFile); + +_Writable_bytes_(*pcbData) +_Readable_bytes_(*pcbData) +_Success_(return != NULL) +PVOID WINAPI DetourBinaryEnumeratePayloads(_In_ PDETOUR_BINARY pBinary, + _Out_opt_ GUID *pGuid, + _Out_ DWORD *pcbData, + _Inout_ DWORD *pnIterator); + +_Writable_bytes_(*pcbData) +_Readable_bytes_(*pcbData) +_Success_(return != NULL) +PVOID WINAPI DetourBinaryFindPayload(_In_ PDETOUR_BINARY pBinary, + _In_ REFGUID rguid, + _Out_ DWORD *pcbData); + +PVOID WINAPI DetourBinarySetPayload(_In_ PDETOUR_BINARY pBinary, + _In_ REFGUID rguid, + _In_reads_opt_(cbData) PVOID pData, + _In_ DWORD cbData); +BOOL WINAPI DetourBinaryDeletePayload(_In_ PDETOUR_BINARY pBinary, _In_ REFGUID rguid); +BOOL WINAPI DetourBinaryPurgePayloads(_In_ PDETOUR_BINARY pBinary); +BOOL WINAPI DetourBinaryResetImports(_In_ PDETOUR_BINARY pBinary); +BOOL WINAPI DetourBinaryEditImports(_In_ PDETOUR_BINARY pBinary, + _In_opt_ PVOID pContext, + _In_opt_ PF_DETOUR_BINARY_BYWAY_CALLBACK pfByway, + _In_opt_ PF_DETOUR_BINARY_FILE_CALLBACK pfFile, + _In_opt_ PF_DETOUR_BINARY_SYMBOL_CALLBACK pfSymbol, + _In_opt_ PF_DETOUR_BINARY_COMMIT_CALLBACK pfCommit); +BOOL WINAPI DetourBinaryWrite(_In_ PDETOUR_BINARY pBinary, _In_ HANDLE hFile); +BOOL WINAPI DetourBinaryClose(_In_ PDETOUR_BINARY pBinary); + +/////////////////////////////////////////////////// Create Process & Load Dll. +// +_Success_(return != NULL) +PVOID WINAPI DetourFindRemotePayload(_In_ HANDLE hProcess, + _In_ REFGUID rguid, + _Out_opt_ DWORD *pcbData); + +typedef BOOL (WINAPI *PDETOUR_CREATE_PROCESS_ROUTINEA)( + _In_opt_ LPCSTR lpApplicationName, + _Inout_opt_ LPSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOA lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation); + +typedef BOOL (WINAPI *PDETOUR_CREATE_PROCESS_ROUTINEW)( + _In_opt_ LPCWSTR lpApplicationName, + _Inout_opt_ LPWSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCWSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOW lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation); + +BOOL WINAPI DetourCreateProcessWithDllA(_In_opt_ LPCSTR lpApplicationName, + _Inout_opt_ LPSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOA lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ LPCSTR lpDllName, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEA pfCreateProcessA); + +BOOL WINAPI DetourCreateProcessWithDllW(_In_opt_ LPCWSTR lpApplicationName, + _Inout_opt_ LPWSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCWSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOW lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ LPCSTR lpDllName, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEW pfCreateProcessW); + +#ifdef UNICODE +#define DetourCreateProcessWithDll DetourCreateProcessWithDllW +#define PDETOUR_CREATE_PROCESS_ROUTINE PDETOUR_CREATE_PROCESS_ROUTINEW +#else +#define DetourCreateProcessWithDll DetourCreateProcessWithDllA +#define PDETOUR_CREATE_PROCESS_ROUTINE PDETOUR_CREATE_PROCESS_ROUTINEA +#endif // !UNICODE + +BOOL WINAPI DetourCreateProcessWithDllExA(_In_opt_ LPCSTR lpApplicationName, + _Inout_opt_ LPSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOA lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ LPCSTR lpDllName, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEA pfCreateProcessA); + +BOOL WINAPI DetourCreateProcessWithDllExW(_In_opt_ LPCWSTR lpApplicationName, + _Inout_opt_ LPWSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCWSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOW lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ LPCSTR lpDllName, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEW pfCreateProcessW); + +#ifdef UNICODE +#define DetourCreateProcessWithDllEx DetourCreateProcessWithDllExW +#else +#define DetourCreateProcessWithDllEx DetourCreateProcessWithDllExA +#endif // !UNICODE + +BOOL WINAPI DetourCreateProcessWithDllsA(_In_opt_ LPCSTR lpApplicationName, + _Inout_opt_ LPSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOA lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ DWORD nDlls, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEA pfCreateProcessA); + +BOOL WINAPI DetourCreateProcessWithDllsW(_In_opt_ LPCWSTR lpApplicationName, + _Inout_opt_ LPWSTR lpCommandLine, + _In_opt_ LPSECURITY_ATTRIBUTES lpProcessAttributes, + _In_opt_ LPSECURITY_ATTRIBUTES lpThreadAttributes, + _In_ BOOL bInheritHandles, + _In_ DWORD dwCreationFlags, + _In_opt_ LPVOID lpEnvironment, + _In_opt_ LPCWSTR lpCurrentDirectory, + _In_ LPSTARTUPINFOW lpStartupInfo, + _Out_ LPPROCESS_INFORMATION lpProcessInformation, + _In_ DWORD nDlls, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_opt_ PDETOUR_CREATE_PROCESS_ROUTINEW pfCreateProcessW); + +#ifdef UNICODE +#define DetourCreateProcessWithDlls DetourCreateProcessWithDllsW +#else +#define DetourCreateProcessWithDlls DetourCreateProcessWithDllsA +#endif // !UNICODE + +BOOL WINAPI DetourProcessViaHelperA(_In_ DWORD dwTargetPid, + _In_ LPCSTR lpDllName, + _In_ PDETOUR_CREATE_PROCESS_ROUTINEA pfCreateProcessA); + +BOOL WINAPI DetourProcessViaHelperW(_In_ DWORD dwTargetPid, + _In_ LPCSTR lpDllName, + _In_ PDETOUR_CREATE_PROCESS_ROUTINEW pfCreateProcessW); + +#ifdef UNICODE +#define DetourProcessViaHelper DetourProcessViaHelperW +#else +#define DetourProcessViaHelper DetourProcessViaHelperA +#endif // !UNICODE + +BOOL WINAPI DetourProcessViaHelperDllsA(_In_ DWORD dwTargetPid, + _In_ DWORD nDlls, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_ PDETOUR_CREATE_PROCESS_ROUTINEA pfCreateProcessA); + +BOOL WINAPI DetourProcessViaHelperDllsW(_In_ DWORD dwTargetPid, + _In_ DWORD nDlls, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_ PDETOUR_CREATE_PROCESS_ROUTINEW pfCreateProcessW); + +#ifdef UNICODE +#define DetourProcessViaHelperDlls DetourProcessViaHelperDllsW +#else +#define DetourProcessViaHelperDlls DetourProcessViaHelperDllsA +#endif // !UNICODE + +BOOL WINAPI DetourUpdateProcessWithDll(_In_ HANDLE hProcess, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_ DWORD nDlls); + +BOOL WINAPI DetourUpdateProcessWithDllEx(_In_ HANDLE hProcess, + _In_ HMODULE hImage, + _In_ BOOL bIs32Bit, + _In_reads_(nDlls) LPCSTR *rlpDlls, + _In_ DWORD nDlls); + +BOOL WINAPI DetourCopyPayloadToProcess(_In_ HANDLE hProcess, + _In_ REFGUID rguid, + _In_reads_bytes_(cbData) LPCVOID pvData, + _In_ DWORD cbData); +_Success_(return != NULL) +PVOID WINAPI DetourCopyPayloadToProcessEx(_In_ HANDLE hProcess, + _In_ REFGUID rguid, + _In_reads_bytes_(cbData) LPCVOID pvData, + _In_ DWORD cbData); + +BOOL WINAPI DetourRestoreAfterWith(VOID); +BOOL WINAPI DetourRestoreAfterWithEx(_In_reads_bytes_(cbData) PVOID pvData, + _In_ DWORD cbData); +BOOL WINAPI DetourIsHelperProcess(VOID); +VOID CALLBACK DetourFinishHelperProcess(_In_ HWND, + _In_ HINSTANCE, + _In_ LPSTR, + _In_ INT); + +// +////////////////////////////////////////////////////////////////////////////// +#ifdef __cplusplus +} +#endif // __cplusplus + +/////////////////////////////////////////////////// Type-safe overloads for C++ +// +#if __cplusplus >= 201103L || _MSVC_LANG >= 201103L +#include + +template +struct DetoursIsFunctionPointer : std::false_type {}; + +template +struct DetoursIsFunctionPointer : std::is_function::type> {}; + +template< + typename T, + typename std::enable_if::value, int>::type = 0> +LONG DetourAttach(_Inout_ T *ppPointer, + _In_ T pDetour) noexcept +{ + return DetourAttach( + reinterpret_cast(ppPointer), + reinterpret_cast(pDetour)); +} + +template< + typename T, + typename std::enable_if::value, int>::type = 0> +LONG DetourAttachEx(_Inout_ T *ppPointer, + _In_ T pDetour, + _Out_opt_ PDETOUR_TRAMPOLINE *ppRealTrampoline, + _Out_opt_ T *ppRealTarget, + _Out_opt_ T *ppRealDetour) noexcept +{ + return DetourAttachEx( + reinterpret_cast(ppPointer), + reinterpret_cast(pDetour), + ppRealTrampoline, + reinterpret_cast(ppRealTarget), + reinterpret_cast(ppRealDetour)); +} + +template< + typename T, + typename std::enable_if::value, int>::type = 0> +LONG DetourDetach(_Inout_ T *ppPointer, + _In_ T pDetour) noexcept +{ + return DetourDetach( + reinterpret_cast(ppPointer), + reinterpret_cast(pDetour)); +} + +#endif // __cplusplus >= 201103L || _MSVC_LANG >= 201103L +// +////////////////////////////////////////////////////////////////////////////// + +//////////////////////////////////////////////// Detours Internal Definitions. +// +#ifdef __cplusplus +#ifdef DETOURS_INTERNAL + +#define NOTHROW +// #define NOTHROW (nothrow) + +////////////////////////////////////////////////////////////////////////////// +// +#if (_MSC_VER < 1299) && !defined(__GNUC__) +#include +typedef IMAGEHLP_MODULE IMAGEHLP_MODULE64; +typedef PIMAGEHLP_MODULE PIMAGEHLP_MODULE64; +typedef IMAGEHLP_SYMBOL SYMBOL_INFO; +typedef PIMAGEHLP_SYMBOL PSYMBOL_INFO; + +static inline +LONG InterlockedCompareExchange(_Inout_ LONG *ptr, _In_ LONG nval, _In_ LONG oval) +{ + return (LONG)::InterlockedCompareExchange((PVOID*)ptr, (PVOID)nval, (PVOID)oval); +} +#else +#pragma warning(push) +#pragma warning(disable:4091) // empty typedef +#include +#pragma warning(pop) +#endif + +#ifdef IMAGEAPI // defined by DBGHELP.H +typedef LPAPI_VERSION (NTAPI *PF_ImagehlpApiVersionEx)(_In_ LPAPI_VERSION AppVersion); + +typedef BOOL (NTAPI *PF_SymInitialize)(_In_ HANDLE hProcess, + _In_opt_ LPCSTR UserSearchPath, + _In_ BOOL fInvadeProcess); +typedef DWORD (NTAPI *PF_SymSetOptions)(_In_ DWORD SymOptions); +typedef DWORD (NTAPI *PF_SymGetOptions)(VOID); +typedef DWORD64 (NTAPI *PF_SymLoadModule64)(_In_ HANDLE hProcess, + _In_opt_ HANDLE hFile, + _In_opt_ LPSTR ImageName, + _In_opt_ LPSTR ModuleName, + _In_ DWORD64 BaseOfDll, + _In_ DWORD SizeOfDll); +typedef BOOL (NTAPI *PF_SymGetModuleInfo64)(_In_ HANDLE hProcess, + _In_ DWORD64 qwAddr, + _Out_ PIMAGEHLP_MODULE64 ModuleInfo); +typedef BOOL (NTAPI *PF_SymFromName)(_In_ HANDLE hProcess, + _In_ LPSTR Name, + _Out_ PSYMBOL_INFO Symbol); + +typedef struct _DETOUR_SYM_INFO +{ + HANDLE hProcess; + HMODULE hDbgHelp; + PF_ImagehlpApiVersionEx pfImagehlpApiVersionEx; + PF_SymInitialize pfSymInitialize; + PF_SymSetOptions pfSymSetOptions; + PF_SymGetOptions pfSymGetOptions; + PF_SymLoadModule64 pfSymLoadModule64; + PF_SymGetModuleInfo64 pfSymGetModuleInfo64; + PF_SymFromName pfSymFromName; +} DETOUR_SYM_INFO, *PDETOUR_SYM_INFO; + +PDETOUR_SYM_INFO DetourLoadImageHlp(VOID); + +#endif // IMAGEAPI + +#if defined(_INC_STDIO) && !defined(_CRT_STDIO_ARBITRARY_WIDE_SPECIFIERS) +#error detours.h must be included before stdio.h (or at least define _CRT_STDIO_ARBITRARY_WIDE_SPECIFIERS earlier) +#endif +#define _CRT_STDIO_ARBITRARY_WIDE_SPECIFIERS 1 + +#ifdef _DEBUG + +int Detour_AssertExprWithFunctionName(int reportType, const char* filename, int linenumber, const char* FunctionName, const char* msg); + +#define DETOUR_ASSERT_EXPR_WITH_FUNCTION(expr, msg) \ + (void) ((expr) || \ + (1 != Detour_AssertExprWithFunctionName(_CRT_ASSERT, __FILE__, __LINE__,__FUNCTION__, msg)) || \ + (_CrtDbgBreak(), 0)) + +#define DETOUR_ASSERT(expr) DETOUR_ASSERT_EXPR_WITH_FUNCTION((expr), #expr) + +#else// _DEBUG +#define DETOUR_ASSERT(expr) +#endif// _DEBUG + +#ifndef DETOUR_TRACE +#if DETOUR_DEBUG +#define DETOUR_TRACE(x) printf x +#define DETOUR_BREAK() __debugbreak() +#include +#include +#else +#define DETOUR_TRACE(x) +#define DETOUR_BREAK() +#endif +#endif + +#if 1 || defined(DETOURS_IA64) + +// +// IA64 instructions are 41 bits, 3 per bundle, plus 5 bit bundle template => 128 bits per bundle. +// + +#define DETOUR_IA64_INSTRUCTIONS_PER_BUNDLE (3) + +#define DETOUR_IA64_TEMPLATE_OFFSET (0) +#define DETOUR_IA64_TEMPLATE_SIZE (5) + +#define DETOUR_IA64_INSTRUCTION_SIZE (41) +#define DETOUR_IA64_INSTRUCTION0_OFFSET (DETOUR_IA64_TEMPLATE_SIZE) +#define DETOUR_IA64_INSTRUCTION1_OFFSET (DETOUR_IA64_TEMPLATE_SIZE + DETOUR_IA64_INSTRUCTION_SIZE) +#define DETOUR_IA64_INSTRUCTION2_OFFSET (DETOUR_IA64_TEMPLATE_SIZE + DETOUR_IA64_INSTRUCTION_SIZE + DETOUR_IA64_INSTRUCTION_SIZE) + +C_ASSERT(DETOUR_IA64_TEMPLATE_SIZE + DETOUR_IA64_INSTRUCTIONS_PER_BUNDLE * DETOUR_IA64_INSTRUCTION_SIZE == 128); + +__declspec(align(16)) struct DETOUR_IA64_BUNDLE +{ + public: + union + { + BYTE data[16]; + UINT64 wide[2]; + }; + + enum { + A_UNIT = 1u, + I_UNIT = 2u, + M_UNIT = 3u, + B_UNIT = 4u, + F_UNIT = 5u, + L_UNIT = 6u, + X_UNIT = 7u, + }; + struct DETOUR_IA64_METADATA + { + ULONG nTemplate : 8; // Instruction template. + ULONG nUnit0 : 4; // Unit for slot 0 + ULONG nUnit1 : 4; // Unit for slot 1 + ULONG nUnit2 : 4; // Unit for slot 2 + }; + + protected: + static const DETOUR_IA64_METADATA s_rceCopyTable[33]; + + UINT RelocateBundle(_Inout_ DETOUR_IA64_BUNDLE* pDst, _Inout_opt_ DETOUR_IA64_BUNDLE* pBundleExtra) const; + + bool RelocateInstruction(_Inout_ DETOUR_IA64_BUNDLE* pDst, + _In_ BYTE slot, + _Inout_opt_ DETOUR_IA64_BUNDLE* pBundleExtra) const; + + // 120 112 104 96 88 80 72 64 56 48 40 32 24 16 8 0 + // f. e. d. c. b. a. 9. 8. 7. 6. 5. 4. 3. 2. 1. 0. + + // 00 + // f.e. d.c. b.a. 9.8. 7.6. 5.4. 3.2. 1.0. + // 0000 0000 0000 0000 0000 0000 0000 001f : Template [4..0] + // 0000 0000 0000 0000 0000 03ff ffff ffe0 : Zero [ 41.. 5] + // 0000 0000 0000 0000 0000 3c00 0000 0000 : Zero [ 45.. 42] + // 0000 0000 0007 ffff ffff c000 0000 0000 : One [ 82.. 46] + // 0000 0000 0078 0000 0000 0000 0000 0000 : One [ 86.. 83] + // 0fff ffff ff80 0000 0000 0000 0000 0000 : Two [123.. 87] + // f000 0000 0000 0000 0000 0000 0000 0000 : Two [127..124] + BYTE GetTemplate() const; + // Get 4 bit opcodes. + BYTE GetInst0() const; + BYTE GetInst1() const; + BYTE GetInst2() const; + BYTE GetUnit(BYTE slot) const; + BYTE GetUnit0() const; + BYTE GetUnit1() const; + BYTE GetUnit2() const; + // Get 37 bit data. + UINT64 GetData0() const; + UINT64 GetData1() const; + UINT64 GetData2() const; + + // Get/set the full 41 bit instructions. + UINT64 GetInstruction(BYTE slot) const; + UINT64 GetInstruction0() const; + UINT64 GetInstruction1() const; + UINT64 GetInstruction2() const; + void SetInstruction(BYTE slot, UINT64 instruction); + void SetInstruction0(UINT64 instruction); + void SetInstruction1(UINT64 instruction); + void SetInstruction2(UINT64 instruction); + + // Get/set bitfields. + static UINT64 GetBits(UINT64 Value, UINT64 Offset, UINT64 Count); + static UINT64 SetBits(UINT64 Value, UINT64 Offset, UINT64 Count, UINT64 Field); + + // Get specific read-only fields. + static UINT64 GetOpcode(UINT64 instruction); // 4bit opcode + static UINT64 GetX(UINT64 instruction); // 1bit opcode extension + static UINT64 GetX3(UINT64 instruction); // 3bit opcode extension + static UINT64 GetX6(UINT64 instruction); // 6bit opcode extension + + // Get/set specific fields. + static UINT64 GetImm7a(UINT64 instruction); + static UINT64 SetImm7a(UINT64 instruction, UINT64 imm7a); + static UINT64 GetImm13c(UINT64 instruction); + static UINT64 SetImm13c(UINT64 instruction, UINT64 imm13c); + static UINT64 GetSignBit(UINT64 instruction); + static UINT64 SetSignBit(UINT64 instruction, UINT64 signBit); + static UINT64 GetImm20a(UINT64 instruction); + static UINT64 SetImm20a(UINT64 instruction, UINT64 imm20a); + static UINT64 GetImm20b(UINT64 instruction); + static UINT64 SetImm20b(UINT64 instruction, UINT64 imm20b); + + static UINT64 SignExtend(UINT64 Value, UINT64 Offset); + + BOOL IsMovlGp() const; + + VOID SetInst(BYTE Slot, BYTE nInst); + VOID SetInst0(BYTE nInst); + VOID SetInst1(BYTE nInst); + VOID SetInst2(BYTE nInst); + VOID SetData(BYTE Slot, UINT64 nData); + VOID SetData0(UINT64 nData); + VOID SetData1(UINT64 nData); + VOID SetData2(UINT64 nData); + BOOL SetNop(BYTE Slot); + BOOL SetNop0(); + BOOL SetNop1(); + BOOL SetNop2(); + + public: + BOOL IsBrl() const; + VOID SetBrl(); + VOID SetBrl(UINT64 target); + UINT64 GetBrlTarget() const; + VOID SetBrlTarget(UINT64 target); + VOID SetBrlImm(UINT64 imm); + UINT64 GetBrlImm() const; + + UINT64 GetMovlGp() const; + VOID SetMovlGp(UINT64 gp); + + VOID SetStop(); + + UINT Copy(_Out_ DETOUR_IA64_BUNDLE *pDst, _Inout_opt_ DETOUR_IA64_BUNDLE* pBundleExtra = NULL) const; +}; +#endif // DETOURS_IA64 + +#ifdef DETOURS_ARM + +#define DETOURS_PFUNC_TO_PBYTE(p) ((PBYTE)(((ULONG_PTR)(p)) & ~(ULONG_PTR)1)) +#define DETOURS_PBYTE_TO_PFUNC(p) ((PBYTE)(((ULONG_PTR)(p)) | (ULONG_PTR)1)) + +#endif // DETOURS_ARM + +////////////////////////////////////////////////////////////////////////////// + +#ifdef __cplusplus +extern "C" { +#endif // __cplusplus + +#define DETOUR_OFFLINE_LIBRARY(x) \ +PVOID WINAPI DetourCopyInstruction##x(_In_opt_ PVOID pDst, \ + _Inout_opt_ PVOID *ppDstPool, \ + _In_ PVOID pSrc, \ + _Out_opt_ PVOID *ppTarget, \ + _Out_opt_ LONG *plExtra); \ + \ +BOOL WINAPI DetourSetCodeModule##x(_In_ HMODULE hModule, \ + _In_ BOOL fLimitReferencesToModule); \ + +DETOUR_OFFLINE_LIBRARY(X86) +DETOUR_OFFLINE_LIBRARY(X64) +DETOUR_OFFLINE_LIBRARY(ARM) +DETOUR_OFFLINE_LIBRARY(ARM64) +DETOUR_OFFLINE_LIBRARY(IA64) + +#undef DETOUR_OFFLINE_LIBRARY + +////////////////////////////////////////////////////////////////////////////// +// +// Helpers for manipulating page protection. +// + +_Success_(return != FALSE) +BOOL WINAPI DetourVirtualProtectSameExecuteEx(_In_ HANDLE hProcess, + _In_ PVOID pAddress, + _In_ SIZE_T nSize, + _In_ DWORD dwNewProtect, + _Out_ PDWORD pdwOldProtect); + +_Success_(return != FALSE) +BOOL WINAPI DetourVirtualProtectSameExecute(_In_ PVOID pAddress, + _In_ SIZE_T nSize, + _In_ DWORD dwNewProtect, + _Out_ PDWORD pdwOldProtect); + +// Detours must depend only on kernel32.lib, so we cannot use IsEqualGUID +BOOL WINAPI DetourAreSameGuid(_In_ REFGUID left, _In_ REFGUID right); +#ifdef __cplusplus +} +#endif // __cplusplus + +////////////////////////////////////////////////////////////////////////////// + +#define MM_ALLOCATION_GRANULARITY 0x10000 + +////////////////////////////////////////////////////////////////////////////// + +#endif // DETOURS_INTERNAL +#endif // __cplusplus + +#endif // _DETOURS_H_ +// +//////////////////////////////////////////////////////////////// End of File. diff --git a/include/detver.h b/include/detver.h new file mode 100644 index 0000000..3d4f544 --- /dev/null +++ b/include/detver.h @@ -0,0 +1,27 @@ +////////////////////////////////////////////////////////////////////////////// +// +// Common version parameters. +// +// Microsoft Research Detours Package, Version 4.0.1 +// +// Copyright (c) Microsoft Corporation. All rights reserved. +// + +#define _USING_V110_SDK71_ 1 +#include "winver.h" +#if 0 +#include +#include +#else +#ifndef DETOURS_STRINGIFY +#define DETOURS_STRINGIFY_(x) #x +#define DETOURS_STRINGIFY(x) DETOURS_STRINGIFY_(x) +#endif + +#define VER_FILEFLAGSMASK 0x3fL +#define VER_FILEFLAGS 0x0L +#define VER_FILEOS 0x00040004L +#define VER_FILETYPE 0x00000002L +#define VER_FILESUBTYPE 0x00000000L +#endif +#define VER_DETOURS_BITS DETOURS_STRINGIFY(DETOURS_BITS) diff --git a/include/rapidjson/allocators.h b/include/rapidjson/allocators.h new file mode 100644 index 0000000..275417b --- /dev/null +++ b/include/rapidjson/allocators.h @@ -0,0 +1,693 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ALLOCATORS_H_ +#define RAPIDJSON_ALLOCATORS_H_ + +#include "rapidjson.h" +#include "internal/meta.h" + +#include +#include + +#if RAPIDJSON_HAS_CXX11 +#include +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// Allocator + +/*! \class rapidjson::Allocator + \brief Concept for allocating, resizing and freeing memory block. + + Note that Malloc() and Realloc() are non-static but Free() is static. + + So if an allocator need to support Free(), it needs to put its pointer in + the header of memory block. + +\code +concept Allocator { + static const bool kNeedFree; //!< Whether this allocator needs to call Free(). + + // Allocate a memory block. + // \param size of the memory block in bytes. + // \returns pointer to the memory block. + void* Malloc(size_t size); + + // Resize a memory block. + // \param originalPtr The pointer to current memory block. Null pointer is permitted. + // \param originalSize The current size in bytes. (Design issue: since some allocator may not book-keep this, explicitly pass to it can save memory.) + // \param newSize the new size in bytes. + void* Realloc(void* originalPtr, size_t originalSize, size_t newSize); + + // Free a memory block. + // \param pointer to the memory block. Null pointer is permitted. + static void Free(void *ptr); +}; +\endcode +*/ + + +/*! \def RAPIDJSON_ALLOCATOR_DEFAULT_CHUNK_CAPACITY + \ingroup RAPIDJSON_CONFIG + \brief User-defined kDefaultChunkCapacity definition. + + User can define this as any \c size that is a power of 2. +*/ + +#ifndef RAPIDJSON_ALLOCATOR_DEFAULT_CHUNK_CAPACITY +#define RAPIDJSON_ALLOCATOR_DEFAULT_CHUNK_CAPACITY (64 * 1024) +#endif + + +/////////////////////////////////////////////////////////////////////////////// +// CrtAllocator + +//! C-runtime library allocator. +/*! This class is just wrapper for standard C library memory routines. + \note implements Allocator concept +*/ +class CrtAllocator { +public: + static const bool kNeedFree = true; + void* Malloc(size_t size) { + if (size) // behavior of malloc(0) is implementation defined. + return RAPIDJSON_MALLOC(size); + else + return NULL; // standardize to returning NULL. + } + void* Realloc(void* originalPtr, size_t originalSize, size_t newSize) { + (void)originalSize; + if (newSize == 0) { + RAPIDJSON_FREE(originalPtr); + return NULL; + } + return RAPIDJSON_REALLOC(originalPtr, newSize); + } + static void Free(void *ptr) RAPIDJSON_NOEXCEPT { RAPIDJSON_FREE(ptr); } + + bool operator==(const CrtAllocator&) const RAPIDJSON_NOEXCEPT { + return true; + } + bool operator!=(const CrtAllocator&) const RAPIDJSON_NOEXCEPT { + return false; + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// MemoryPoolAllocator + +//! Default memory allocator used by the parser and DOM. +/*! This allocator allocate memory blocks from pre-allocated memory chunks. + + It does not free memory blocks. And Realloc() only allocate new memory. + + The memory chunks are allocated by BaseAllocator, which is CrtAllocator by default. + + User may also supply a buffer as the first chunk. + + If the user-buffer is full then additional chunks are allocated by BaseAllocator. + + The user-buffer is not deallocated by this allocator. + + \tparam BaseAllocator the allocator type for allocating memory chunks. Default is CrtAllocator. + \note implements Allocator concept +*/ +template +class MemoryPoolAllocator { + //! Chunk header for perpending to each chunk. + /*! Chunks are stored as a singly linked list. + */ + struct ChunkHeader { + size_t capacity; //!< Capacity of the chunk in bytes (excluding the header itself). + size_t size; //!< Current size of allocated memory in bytes. + ChunkHeader *next; //!< Next chunk in the linked list. + }; + + struct SharedData { + ChunkHeader *chunkHead; //!< Head of the chunk linked-list. Only the head chunk serves allocation. + BaseAllocator* ownBaseAllocator; //!< base allocator created by this object. + size_t refcount; + bool ownBuffer; + }; + + static const size_t SIZEOF_SHARED_DATA = RAPIDJSON_ALIGN(sizeof(SharedData)); + static const size_t SIZEOF_CHUNK_HEADER = RAPIDJSON_ALIGN(sizeof(ChunkHeader)); + + static inline ChunkHeader *GetChunkHead(SharedData *shared) + { + return reinterpret_cast(reinterpret_cast(shared) + SIZEOF_SHARED_DATA); + } + static inline uint8_t *GetChunkBuffer(SharedData *shared) + { + return reinterpret_cast(shared->chunkHead) + SIZEOF_CHUNK_HEADER; + } + + static const size_t kDefaultChunkCapacity = RAPIDJSON_ALLOCATOR_DEFAULT_CHUNK_CAPACITY; //!< Default chunk capacity. + +public: + static const bool kNeedFree = false; //!< Tell users that no need to call Free() with this allocator. (concept Allocator) + static const bool kRefCounted = true; //!< Tell users that this allocator is reference counted on copy + + //! Constructor with chunkSize. + /*! \param chunkSize The size of memory chunk. The default is kDefaultChunkSize. + \param baseAllocator The allocator for allocating memory chunks. + */ + explicit + MemoryPoolAllocator(size_t chunkSize = kDefaultChunkCapacity, BaseAllocator* baseAllocator = 0) : + chunk_capacity_(chunkSize), + baseAllocator_(baseAllocator ? baseAllocator : RAPIDJSON_NEW(BaseAllocator)()), + shared_(static_cast(baseAllocator_ ? baseAllocator_->Malloc(SIZEOF_SHARED_DATA + SIZEOF_CHUNK_HEADER) : 0)) + { + RAPIDJSON_ASSERT(baseAllocator_ != 0); + RAPIDJSON_ASSERT(shared_ != 0); + if (baseAllocator) { + shared_->ownBaseAllocator = 0; + } + else { + shared_->ownBaseAllocator = baseAllocator_; + } + shared_->chunkHead = GetChunkHead(shared_); + shared_->chunkHead->capacity = 0; + shared_->chunkHead->size = 0; + shared_->chunkHead->next = 0; + shared_->ownBuffer = true; + shared_->refcount = 1; + } + + //! Constructor with user-supplied buffer. + /*! The user buffer will be used firstly. When it is full, memory pool allocates new chunk with chunk size. + + The user buffer will not be deallocated when this allocator is destructed. + + \param buffer User supplied buffer. + \param size Size of the buffer in bytes. It must at least larger than sizeof(ChunkHeader). + \param chunkSize The size of memory chunk. The default is kDefaultChunkSize. + \param baseAllocator The allocator for allocating memory chunks. + */ + MemoryPoolAllocator(void *buffer, size_t size, size_t chunkSize = kDefaultChunkCapacity, BaseAllocator* baseAllocator = 0) : + chunk_capacity_(chunkSize), + baseAllocator_(baseAllocator), + shared_(static_cast(AlignBuffer(buffer, size))) + { + RAPIDJSON_ASSERT(size >= SIZEOF_SHARED_DATA + SIZEOF_CHUNK_HEADER); + shared_->chunkHead = GetChunkHead(shared_); + shared_->chunkHead->capacity = size - SIZEOF_SHARED_DATA - SIZEOF_CHUNK_HEADER; + shared_->chunkHead->size = 0; + shared_->chunkHead->next = 0; + shared_->ownBaseAllocator = 0; + shared_->ownBuffer = false; + shared_->refcount = 1; + } + + MemoryPoolAllocator(const MemoryPoolAllocator& rhs) RAPIDJSON_NOEXCEPT : + chunk_capacity_(rhs.chunk_capacity_), + baseAllocator_(rhs.baseAllocator_), + shared_(rhs.shared_) + { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + ++shared_->refcount; + } + MemoryPoolAllocator& operator=(const MemoryPoolAllocator& rhs) RAPIDJSON_NOEXCEPT + { + RAPIDJSON_NOEXCEPT_ASSERT(rhs.shared_->refcount > 0); + ++rhs.shared_->refcount; + this->~MemoryPoolAllocator(); + baseAllocator_ = rhs.baseAllocator_; + chunk_capacity_ = rhs.chunk_capacity_; + shared_ = rhs.shared_; + return *this; + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + MemoryPoolAllocator(MemoryPoolAllocator&& rhs) RAPIDJSON_NOEXCEPT : + chunk_capacity_(rhs.chunk_capacity_), + baseAllocator_(rhs.baseAllocator_), + shared_(rhs.shared_) + { + RAPIDJSON_NOEXCEPT_ASSERT(rhs.shared_->refcount > 0); + rhs.shared_ = 0; + } + MemoryPoolAllocator& operator=(MemoryPoolAllocator&& rhs) RAPIDJSON_NOEXCEPT + { + RAPIDJSON_NOEXCEPT_ASSERT(rhs.shared_->refcount > 0); + this->~MemoryPoolAllocator(); + baseAllocator_ = rhs.baseAllocator_; + chunk_capacity_ = rhs.chunk_capacity_; + shared_ = rhs.shared_; + rhs.shared_ = 0; + return *this; + } +#endif + + //! Destructor. + /*! This deallocates all memory chunks, excluding the user-supplied buffer. + */ + ~MemoryPoolAllocator() RAPIDJSON_NOEXCEPT { + if (!shared_) { + // do nothing if moved + return; + } + if (shared_->refcount > 1) { + --shared_->refcount; + return; + } + Clear(); + BaseAllocator *a = shared_->ownBaseAllocator; + if (shared_->ownBuffer) { + baseAllocator_->Free(shared_); + } + RAPIDJSON_DELETE(a); + } + + //! Deallocates all memory chunks, excluding the first/user one. + void Clear() RAPIDJSON_NOEXCEPT { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + for (;;) { + ChunkHeader* c = shared_->chunkHead; + if (!c->next) { + break; + } + shared_->chunkHead = c->next; + baseAllocator_->Free(c); + } + shared_->chunkHead->size = 0; + } + + //! Computes the total capacity of allocated memory chunks. + /*! \return total capacity in bytes. + */ + size_t Capacity() const RAPIDJSON_NOEXCEPT { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + size_t capacity = 0; + for (ChunkHeader* c = shared_->chunkHead; c != 0; c = c->next) + capacity += c->capacity; + return capacity; + } + + //! Computes the memory blocks allocated. + /*! \return total used bytes. + */ + size_t Size() const RAPIDJSON_NOEXCEPT { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + size_t size = 0; + for (ChunkHeader* c = shared_->chunkHead; c != 0; c = c->next) + size += c->size; + return size; + } + + //! Whether the allocator is shared. + /*! \return true or false. + */ + bool Shared() const RAPIDJSON_NOEXCEPT { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + return shared_->refcount > 1; + } + + //! Allocates a memory block. (concept Allocator) + void* Malloc(size_t size) { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + if (!size) + return NULL; + + size = RAPIDJSON_ALIGN(size); + if (RAPIDJSON_UNLIKELY(shared_->chunkHead->size + size > shared_->chunkHead->capacity)) + if (!AddChunk(chunk_capacity_ > size ? chunk_capacity_ : size)) + return NULL; + + void *buffer = GetChunkBuffer(shared_) + shared_->chunkHead->size; + shared_->chunkHead->size += size; + return buffer; + } + + //! Resizes a memory block (concept Allocator) + void* Realloc(void* originalPtr, size_t originalSize, size_t newSize) { + if (originalPtr == 0) + return Malloc(newSize); + + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + if (newSize == 0) + return NULL; + + originalSize = RAPIDJSON_ALIGN(originalSize); + newSize = RAPIDJSON_ALIGN(newSize); + + // Do not shrink if new size is smaller than original + if (originalSize >= newSize) + return originalPtr; + + // Simply expand it if it is the last allocation and there is sufficient space + if (originalPtr == GetChunkBuffer(shared_) + shared_->chunkHead->size - originalSize) { + size_t increment = static_cast(newSize - originalSize); + if (shared_->chunkHead->size + increment <= shared_->chunkHead->capacity) { + shared_->chunkHead->size += increment; + return originalPtr; + } + } + + // Realloc process: allocate and copy memory, do not free original buffer. + if (void* newBuffer = Malloc(newSize)) { + if (originalSize) + std::memcpy(newBuffer, originalPtr, originalSize); + return newBuffer; + } + else + return NULL; + } + + //! Frees a memory block (concept Allocator) + static void Free(void *ptr) RAPIDJSON_NOEXCEPT { (void)ptr; } // Do nothing + + //! Compare (equality) with another MemoryPoolAllocator + bool operator==(const MemoryPoolAllocator& rhs) const RAPIDJSON_NOEXCEPT { + RAPIDJSON_NOEXCEPT_ASSERT(shared_->refcount > 0); + RAPIDJSON_NOEXCEPT_ASSERT(rhs.shared_->refcount > 0); + return shared_ == rhs.shared_; + } + //! Compare (inequality) with another MemoryPoolAllocator + bool operator!=(const MemoryPoolAllocator& rhs) const RAPIDJSON_NOEXCEPT { + return !operator==(rhs); + } + +private: + //! Creates a new chunk. + /*! \param capacity Capacity of the chunk in bytes. + \return true if success. + */ + bool AddChunk(size_t capacity) { + if (!baseAllocator_) + shared_->ownBaseAllocator = baseAllocator_ = RAPIDJSON_NEW(BaseAllocator)(); + if (ChunkHeader* chunk = static_cast(baseAllocator_->Malloc(SIZEOF_CHUNK_HEADER + capacity))) { + chunk->capacity = capacity; + chunk->size = 0; + chunk->next = shared_->chunkHead; + shared_->chunkHead = chunk; + return true; + } + else + return false; + } + + static inline void* AlignBuffer(void* buf, size_t &size) + { + RAPIDJSON_NOEXCEPT_ASSERT(buf != 0); + const uintptr_t mask = sizeof(void*) - 1; + const uintptr_t ubuf = reinterpret_cast(buf); + if (RAPIDJSON_UNLIKELY(ubuf & mask)) { + const uintptr_t abuf = (ubuf + mask) & ~mask; + RAPIDJSON_ASSERT(size >= abuf - ubuf); + buf = reinterpret_cast(abuf); + size -= abuf - ubuf; + } + return buf; + } + + size_t chunk_capacity_; //!< The minimum capacity of chunk when they are allocated. + BaseAllocator* baseAllocator_; //!< base allocator for allocating memory chunks. + SharedData *shared_; //!< The shared data of the allocator +}; + +namespace internal { + template + struct IsRefCounted : + public FalseType + { }; + template + struct IsRefCounted::Type> : + public TrueType + { }; +} + +template +inline T* Realloc(A& a, T* old_p, size_t old_n, size_t new_n) +{ + RAPIDJSON_NOEXCEPT_ASSERT(old_n <= (std::numeric_limits::max)() / sizeof(T) && new_n <= (std::numeric_limits::max)() / sizeof(T)); + return static_cast(a.Realloc(old_p, old_n * sizeof(T), new_n * sizeof(T))); +} + +template +inline T *Malloc(A& a, size_t n = 1) +{ + return Realloc(a, NULL, 0, n); +} + +template +inline void Free(A& a, T *p, size_t n = 1) +{ + static_cast(Realloc(a, p, n, 0)); +} + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) // std::allocator can safely be inherited +#endif + +template +class StdAllocator : + public std::allocator +{ + typedef std::allocator allocator_type; +#if RAPIDJSON_HAS_CXX11 + typedef std::allocator_traits traits_type; +#else + typedef allocator_type traits_type; +#endif + +public: + typedef BaseAllocator BaseAllocatorType; + + StdAllocator() RAPIDJSON_NOEXCEPT : + allocator_type(), + baseAllocator_() + { } + + StdAllocator(const StdAllocator& rhs) RAPIDJSON_NOEXCEPT : + allocator_type(rhs), + baseAllocator_(rhs.baseAllocator_) + { } + + template + StdAllocator(const StdAllocator& rhs) RAPIDJSON_NOEXCEPT : + allocator_type(rhs), + baseAllocator_(rhs.baseAllocator_) + { } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + StdAllocator(StdAllocator&& rhs) RAPIDJSON_NOEXCEPT : + allocator_type(std::move(rhs)), + baseAllocator_(std::move(rhs.baseAllocator_)) + { } +#endif +#if RAPIDJSON_HAS_CXX11 + using propagate_on_container_move_assignment = std::true_type; + using propagate_on_container_swap = std::true_type; +#endif + + /* implicit */ + StdAllocator(const BaseAllocator& baseAllocator) RAPIDJSON_NOEXCEPT : + allocator_type(), + baseAllocator_(baseAllocator) + { } + + ~StdAllocator() RAPIDJSON_NOEXCEPT + { } + + template + struct rebind { + typedef StdAllocator other; + }; + + typedef typename traits_type::size_type size_type; + typedef typename traits_type::difference_type difference_type; + + typedef typename traits_type::value_type value_type; + typedef typename traits_type::pointer pointer; + typedef typename traits_type::const_pointer const_pointer; + +#if RAPIDJSON_HAS_CXX11 + + typedef typename std::add_lvalue_reference::type &reference; + typedef typename std::add_lvalue_reference::type>::type &const_reference; + + pointer address(reference r) const RAPIDJSON_NOEXCEPT + { + return std::addressof(r); + } + const_pointer address(const_reference r) const RAPIDJSON_NOEXCEPT + { + return std::addressof(r); + } + + size_type max_size() const RAPIDJSON_NOEXCEPT + { + return traits_type::max_size(*this); + } + + template + void construct(pointer p, Args&&... args) + { + traits_type::construct(*this, p, std::forward(args)...); + } + void destroy(pointer p) + { + traits_type::destroy(*this, p); + } + +#else // !RAPIDJSON_HAS_CXX11 + + typedef typename allocator_type::reference reference; + typedef typename allocator_type::const_reference const_reference; + + pointer address(reference r) const RAPIDJSON_NOEXCEPT + { + return allocator_type::address(r); + } + const_pointer address(const_reference r) const RAPIDJSON_NOEXCEPT + { + return allocator_type::address(r); + } + + size_type max_size() const RAPIDJSON_NOEXCEPT + { + return allocator_type::max_size(); + } + + void construct(pointer p, const_reference r) + { + allocator_type::construct(p, r); + } + void destroy(pointer p) + { + allocator_type::destroy(p); + } + +#endif // !RAPIDJSON_HAS_CXX11 + + template + U* allocate(size_type n = 1, const void* = 0) + { + return RAPIDJSON_NAMESPACE::Malloc(baseAllocator_, n); + } + template + void deallocate(U* p, size_type n = 1) + { + RAPIDJSON_NAMESPACE::Free(baseAllocator_, p, n); + } + + pointer allocate(size_type n = 1, const void* = 0) + { + return allocate(n); + } + void deallocate(pointer p, size_type n = 1) + { + deallocate(p, n); + } + +#if RAPIDJSON_HAS_CXX11 + using is_always_equal = std::is_empty; +#endif + + template + bool operator==(const StdAllocator& rhs) const RAPIDJSON_NOEXCEPT + { + return baseAllocator_ == rhs.baseAllocator_; + } + template + bool operator!=(const StdAllocator& rhs) const RAPIDJSON_NOEXCEPT + { + return !operator==(rhs); + } + + //! rapidjson Allocator concept + static const bool kNeedFree = BaseAllocator::kNeedFree; + static const bool kRefCounted = internal::IsRefCounted::Value; + void* Malloc(size_t size) + { + return baseAllocator_.Malloc(size); + } + void* Realloc(void* originalPtr, size_t originalSize, size_t newSize) + { + return baseAllocator_.Realloc(originalPtr, originalSize, newSize); + } + static void Free(void *ptr) RAPIDJSON_NOEXCEPT + { + BaseAllocator::Free(ptr); + } + +private: + template + friend class StdAllocator; // access to StdAllocator.* + + BaseAllocator baseAllocator_; +}; + +#if !RAPIDJSON_HAS_CXX17 // std::allocator deprecated in C++17 +template +class StdAllocator : + public std::allocator +{ + typedef std::allocator allocator_type; + +public: + typedef BaseAllocator BaseAllocatorType; + + StdAllocator() RAPIDJSON_NOEXCEPT : + allocator_type(), + baseAllocator_() + { } + + StdAllocator(const StdAllocator& rhs) RAPIDJSON_NOEXCEPT : + allocator_type(rhs), + baseAllocator_(rhs.baseAllocator_) + { } + + template + StdAllocator(const StdAllocator& rhs) RAPIDJSON_NOEXCEPT : + allocator_type(rhs), + baseAllocator_(rhs.baseAllocator_) + { } + + /* implicit */ + StdAllocator(const BaseAllocator& baseAllocator) RAPIDJSON_NOEXCEPT : + allocator_type(), + baseAllocator_(baseAllocator) + { } + + ~StdAllocator() RAPIDJSON_NOEXCEPT + { } + + template + struct rebind { + typedef StdAllocator other; + }; + + typedef typename allocator_type::value_type value_type; + +private: + template + friend class StdAllocator; // access to StdAllocator.* + + BaseAllocator baseAllocator_; +}; +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_ENCODINGS_H_ diff --git a/include/rapidjson/cursorstreamwrapper.h b/include/rapidjson/cursorstreamwrapper.h new file mode 100644 index 0000000..fd6513d --- /dev/null +++ b/include/rapidjson/cursorstreamwrapper.h @@ -0,0 +1,78 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_CURSORSTREAMWRAPPER_H_ +#define RAPIDJSON_CURSORSTREAMWRAPPER_H_ + +#include "stream.h" + +#if defined(__GNUC__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#if defined(_MSC_VER) && _MSC_VER <= 1800 +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4702) // unreachable code +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +RAPIDJSON_NAMESPACE_BEGIN + + +//! Cursor stream wrapper for counting line and column number if error exists. +/*! + \tparam InputStream Any stream that implements Stream Concept +*/ +template > +class CursorStreamWrapper : public GenericStreamWrapper { +public: + typedef typename Encoding::Ch Ch; + + CursorStreamWrapper(InputStream& is): + GenericStreamWrapper(is), line_(1), col_(0) {} + + // counting line and column number + Ch Take() { + Ch ch = this->is_.Take(); + if(ch == '\n') { + line_ ++; + col_ = 0; + } else { + col_ ++; + } + return ch; + } + + //! Get the error line number, if error exists. + size_t GetLine() const { return line_; } + //! Get the error column number, if error exists. + size_t GetColumn() const { return col_; } + +private: + size_t line_; //!< Current Line + size_t col_; //!< Current Column +}; + +#if defined(_MSC_VER) && _MSC_VER <= 1800 +RAPIDJSON_DIAG_POP +#endif + +#if defined(__GNUC__) +RAPIDJSON_DIAG_POP +#endif + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_CURSORSTREAMWRAPPER_H_ diff --git a/include/rapidjson/document.h b/include/rapidjson/document.h new file mode 100644 index 0000000..4b2d723 --- /dev/null +++ b/include/rapidjson/document.h @@ -0,0 +1,3044 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_DOCUMENT_H_ +#define RAPIDJSON_DOCUMENT_H_ + +/*! \file document.h */ + +#include "reader.h" +#include "internal/meta.h" +#include "internal/strfunc.h" +#include "memorystream.h" +#include "encodedstream.h" +#include // placement new +#include +#ifdef __cpp_lib_three_way_comparison +#include +#endif + +RAPIDJSON_DIAG_PUSH +#ifdef __clang__ +RAPIDJSON_DIAG_OFF(padded) +RAPIDJSON_DIAG_OFF(switch-enum) +RAPIDJSON_DIAG_OFF(c++98-compat) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_OFF(4127) // conditional expression is constant +RAPIDJSON_DIAG_OFF(4244) // conversion from kXxxFlags to 'uint16_t', possible loss of data +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_OFF(effc++) +#endif // __GNUC__ + +#ifdef GetObject +// see https://github.com/Tencent/rapidjson/issues/1448 +// a former included windows.h might have defined a macro called GetObject, which affects +// GetObject defined here. This ensures the macro does not get applied +#pragma push_macro("GetObject") +#define RAPIDJSON_WINDOWS_GETOBJECT_WORKAROUND_APPLIED +#undef GetObject +#endif + +#ifndef RAPIDJSON_NOMEMBERITERATORCLASS +#include // std::random_access_iterator_tag +#endif + +#if RAPIDJSON_USE_MEMBERSMAP +#include // std::multimap +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +// Forward declaration. +template +class GenericValue; + +template +class GenericDocument; + +/*! \def RAPIDJSON_DEFAULT_ALLOCATOR + \ingroup RAPIDJSON_CONFIG + \brief Allows to choose default allocator. + + User can define this to use CrtAllocator or MemoryPoolAllocator. +*/ +#ifndef RAPIDJSON_DEFAULT_ALLOCATOR +#define RAPIDJSON_DEFAULT_ALLOCATOR ::RAPIDJSON_NAMESPACE::MemoryPoolAllocator<::RAPIDJSON_NAMESPACE::CrtAllocator> +#endif + +/*! \def RAPIDJSON_DEFAULT_STACK_ALLOCATOR + \ingroup RAPIDJSON_CONFIG + \brief Allows to choose default stack allocator for Document. + + User can define this to use CrtAllocator or MemoryPoolAllocator. +*/ +#ifndef RAPIDJSON_DEFAULT_STACK_ALLOCATOR +#define RAPIDJSON_DEFAULT_STACK_ALLOCATOR ::RAPIDJSON_NAMESPACE::CrtAllocator +#endif + +/*! \def RAPIDJSON_VALUE_DEFAULT_OBJECT_CAPACITY + \ingroup RAPIDJSON_CONFIG + \brief User defined kDefaultObjectCapacity value. + + User can define this as any natural number. +*/ +#ifndef RAPIDJSON_VALUE_DEFAULT_OBJECT_CAPACITY +// number of objects that rapidjson::Value allocates memory for by default +#define RAPIDJSON_VALUE_DEFAULT_OBJECT_CAPACITY 16 +#endif + +/*! \def RAPIDJSON_VALUE_DEFAULT_ARRAY_CAPACITY + \ingroup RAPIDJSON_CONFIG + \brief User defined kDefaultArrayCapacity value. + + User can define this as any natural number. +*/ +#ifndef RAPIDJSON_VALUE_DEFAULT_ARRAY_CAPACITY +// number of array elements that rapidjson::Value allocates memory for by default +#define RAPIDJSON_VALUE_DEFAULT_ARRAY_CAPACITY 16 +#endif + +//! Name-value pair in a JSON object value. +/*! + This class was internal to GenericValue. It used to be a inner struct. + But a compiler (IBM XL C/C++ for AIX) have reported to have problem with that so it moved as a namespace scope struct. + https://code.google.com/p/rapidjson/issues/detail?id=64 +*/ +template +class GenericMember { +public: + GenericValue name; //!< name of member (must be a string) + GenericValue value; //!< value of member. + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move constructor in C++11 + GenericMember(GenericMember&& rhs) RAPIDJSON_NOEXCEPT + : name(std::move(rhs.name)), + value(std::move(rhs.value)) + { + } + + //! Move assignment in C++11 + GenericMember& operator=(GenericMember&& rhs) RAPIDJSON_NOEXCEPT { + return *this = static_cast(rhs); + } +#endif + + //! Assignment with move semantics. + /*! \param rhs Source of the assignment. Its name and value will become a null value after assignment. + */ + GenericMember& operator=(GenericMember& rhs) RAPIDJSON_NOEXCEPT { + if (RAPIDJSON_LIKELY(this != &rhs)) { + name = rhs.name; + value = rhs.value; + } + return *this; + } + + // swap() for std::sort() and other potential use in STL. + friend inline void swap(GenericMember& a, GenericMember& b) RAPIDJSON_NOEXCEPT { + a.name.Swap(b.name); + a.value.Swap(b.value); + } + +private: + //! Copy constructor is not permitted. + GenericMember(const GenericMember& rhs); +}; + +/////////////////////////////////////////////////////////////////////////////// +// GenericMemberIterator + +#ifndef RAPIDJSON_NOMEMBERITERATORCLASS + +//! (Constant) member iterator for a JSON object value +/*! + \tparam Const Is this a constant iterator? + \tparam Encoding Encoding of the value. (Even non-string values need to have the same encoding in a document) + \tparam Allocator Allocator type for allocating memory of object, array and string. + + This class implements a Random Access Iterator for GenericMember elements + of a GenericValue, see ISO/IEC 14882:2003(E) C++ standard, 24.1 [lib.iterator.requirements]. + + \note This iterator implementation is mainly intended to avoid implicit + conversions from iterator values to \c NULL, + e.g. from GenericValue::FindMember. + + \note Define \c RAPIDJSON_NOMEMBERITERATORCLASS to fall back to a + pointer-based implementation, if your platform doesn't provide + the C++ header. + + \see GenericMember, GenericValue::MemberIterator, GenericValue::ConstMemberIterator + */ +template +class GenericMemberIterator { + + friend class GenericValue; + template friend class GenericMemberIterator; + + typedef GenericMember PlainType; + typedef typename internal::MaybeAddConst::Type ValueType; + +public: + //! Iterator type itself + typedef GenericMemberIterator Iterator; + //! Constant iterator type + typedef GenericMemberIterator ConstIterator; + //! Non-constant iterator type + typedef GenericMemberIterator NonConstIterator; + + /** \name std::iterator_traits support */ + //@{ + typedef ValueType value_type; + typedef ValueType * pointer; + typedef ValueType & reference; + typedef std::ptrdiff_t difference_type; + typedef std::random_access_iterator_tag iterator_category; + //@} + + //! Pointer to (const) GenericMember + typedef pointer Pointer; + //! Reference to (const) GenericMember + typedef reference Reference; + //! Signed integer type (e.g. \c ptrdiff_t) + typedef difference_type DifferenceType; + + //! Default constructor (singular value) + /*! Creates an iterator pointing to no element. + \note All operations, except for comparisons, are undefined on such values. + */ + GenericMemberIterator() : ptr_() {} + + //! Iterator conversions to more const + /*! + \param it (Non-const) iterator to copy from + + Allows the creation of an iterator from another GenericMemberIterator + that is "less const". Especially, creating a non-constant iterator + from a constant iterator are disabled: + \li const -> non-const (not ok) + \li const -> const (ok) + \li non-const -> const (ok) + \li non-const -> non-const (ok) + + \note If the \c Const template parameter is already \c false, this + constructor effectively defines a regular copy-constructor. + Otherwise, the copy constructor is implicitly defined. + */ + GenericMemberIterator(const NonConstIterator & it) : ptr_(it.ptr_) {} + Iterator& operator=(const NonConstIterator & it) { ptr_ = it.ptr_; return *this; } + + //! @name stepping + //@{ + Iterator& operator++(){ ++ptr_; return *this; } + Iterator& operator--(){ --ptr_; return *this; } + Iterator operator++(int){ Iterator old(*this); ++ptr_; return old; } + Iterator operator--(int){ Iterator old(*this); --ptr_; return old; } + //@} + + //! @name increment/decrement + //@{ + Iterator operator+(DifferenceType n) const { return Iterator(ptr_+n); } + Iterator operator-(DifferenceType n) const { return Iterator(ptr_-n); } + + Iterator& operator+=(DifferenceType n) { ptr_+=n; return *this; } + Iterator& operator-=(DifferenceType n) { ptr_-=n; return *this; } + //@} + + //! @name relations + //@{ + template bool operator==(const GenericMemberIterator& that) const { return ptr_ == that.ptr_; } + template bool operator!=(const GenericMemberIterator& that) const { return ptr_ != that.ptr_; } + template bool operator<=(const GenericMemberIterator& that) const { return ptr_ <= that.ptr_; } + template bool operator>=(const GenericMemberIterator& that) const { return ptr_ >= that.ptr_; } + template bool operator< (const GenericMemberIterator& that) const { return ptr_ < that.ptr_; } + template bool operator> (const GenericMemberIterator& that) const { return ptr_ > that.ptr_; } + +#ifdef __cpp_lib_three_way_comparison + template std::strong_ordering operator<=>(const GenericMemberIterator& that) const { return ptr_ <=> that.ptr_; } +#endif + //@} + + //! @name dereference + //@{ + Reference operator*() const { return *ptr_; } + Pointer operator->() const { return ptr_; } + Reference operator[](DifferenceType n) const { return ptr_[n]; } + //@} + + //! Distance + DifferenceType operator-(ConstIterator that) const { return ptr_-that.ptr_; } + +private: + //! Internal constructor from plain pointer + explicit GenericMemberIterator(Pointer p) : ptr_(p) {} + + Pointer ptr_; //!< raw pointer +}; + +#else // RAPIDJSON_NOMEMBERITERATORCLASS + +// class-based member iterator implementation disabled, use plain pointers + +template +class GenericMemberIterator; + +//! non-const GenericMemberIterator +template +class GenericMemberIterator { +public: + //! use plain pointer as iterator type + typedef GenericMember* Iterator; +}; +//! const GenericMemberIterator +template +class GenericMemberIterator { +public: + //! use plain const pointer as iterator type + typedef const GenericMember* Iterator; +}; + +#endif // RAPIDJSON_NOMEMBERITERATORCLASS + +/////////////////////////////////////////////////////////////////////////////// +// GenericStringRef + +//! Reference to a constant string (not taking a copy) +/*! + \tparam CharType character type of the string + + This helper class is used to automatically infer constant string + references for string literals, especially from \c const \b (!) + character arrays. + + The main use is for creating JSON string values without copying the + source string via an \ref Allocator. This requires that the referenced + string pointers have a sufficient lifetime, which exceeds the lifetime + of the associated GenericValue. + + \b Example + \code + Value v("foo"); // ok, no need to copy & calculate length + const char foo[] = "foo"; + v.SetString(foo); // ok + + const char* bar = foo; + // Value x(bar); // not ok, can't rely on bar's lifetime + Value x(StringRef(bar)); // lifetime explicitly guaranteed by user + Value y(StringRef(bar, 3)); // ok, explicitly pass length + \endcode + + \see StringRef, GenericValue::SetString +*/ +template +struct GenericStringRef { + typedef CharType Ch; //!< character type of the string + + //! Create string reference from \c const character array +#ifndef __clang__ // -Wdocumentation + /*! + This constructor implicitly creates a constant string reference from + a \c const character array. It has better performance than + \ref StringRef(const CharType*) by inferring the string \ref length + from the array length, and also supports strings containing null + characters. + + \tparam N length of the string, automatically inferred + + \param str Constant character array, lifetime assumed to be longer + than the use of the string in e.g. a GenericValue + + \post \ref s == str + + \note Constant complexity. + \note There is a hidden, private overload to disallow references to + non-const character arrays to be created via this constructor. + By this, e.g. function-scope arrays used to be filled via + \c snprintf are excluded from consideration. + In such cases, the referenced string should be \b copied to the + GenericValue instead. + */ +#endif + template + GenericStringRef(const CharType (&str)[N]) RAPIDJSON_NOEXCEPT + : s(str), length(N-1) {} + + //! Explicitly create string reference from \c const character pointer +#ifndef __clang__ // -Wdocumentation + /*! + This constructor can be used to \b explicitly create a reference to + a constant string pointer. + + \see StringRef(const CharType*) + + \param str Constant character pointer, lifetime assumed to be longer + than the use of the string in e.g. a GenericValue + + \post \ref s == str + + \note There is a hidden, private overload to disallow references to + non-const character arrays to be created via this constructor. + By this, e.g. function-scope arrays used to be filled via + \c snprintf are excluded from consideration. + In such cases, the referenced string should be \b copied to the + GenericValue instead. + */ +#endif + explicit GenericStringRef(const CharType* str) + : s(str), length(NotNullStrLen(str)) {} + + //! Create constant string reference from pointer and length +#ifndef __clang__ // -Wdocumentation + /*! \param str constant string, lifetime assumed to be longer than the use of the string in e.g. a GenericValue + \param len length of the string, excluding the trailing NULL terminator + + \post \ref s == str && \ref length == len + \note Constant complexity. + */ +#endif + GenericStringRef(const CharType* str, SizeType len) + : s(RAPIDJSON_LIKELY(str) ? str : emptyString), length(len) { RAPIDJSON_ASSERT(str != 0 || len == 0u); } + + GenericStringRef(const GenericStringRef& rhs) : s(rhs.s), length(rhs.length) {} + + //! implicit conversion to plain CharType pointer + operator const Ch *() const { return s; } + + const Ch* const s; //!< plain CharType pointer + const SizeType length; //!< length of the string (excluding the trailing NULL terminator) + +private: + SizeType NotNullStrLen(const CharType* str) { + RAPIDJSON_ASSERT(str != 0); + return internal::StrLen(str); + } + + /// Empty string - used when passing in a NULL pointer + static const Ch emptyString[]; + + //! Disallow construction from non-const array + template + GenericStringRef(CharType (&str)[N]) /* = delete */; + //! Copy assignment operator not permitted - immutable type + GenericStringRef& operator=(const GenericStringRef& rhs) /* = delete */; +}; + +template +const CharType GenericStringRef::emptyString[] = { CharType() }; + +//! Mark a character pointer as constant string +/*! Mark a plain character pointer as a "string literal". This function + can be used to avoid copying a character string to be referenced as a + value in a JSON GenericValue object, if the string's lifetime is known + to be valid long enough. + \tparam CharType Character type of the string + \param str Constant string, lifetime assumed to be longer than the use of the string in e.g. a GenericValue + \return GenericStringRef string reference object + \relatesalso GenericStringRef + + \see GenericValue::GenericValue(StringRefType), GenericValue::operator=(StringRefType), GenericValue::SetString(StringRefType), GenericValue::PushBack(StringRefType, Allocator&), GenericValue::AddMember +*/ +template +inline GenericStringRef StringRef(const CharType* str) { + return GenericStringRef(str); +} + +//! Mark a character pointer as constant string +/*! Mark a plain character pointer as a "string literal". This function + can be used to avoid copying a character string to be referenced as a + value in a JSON GenericValue object, if the string's lifetime is known + to be valid long enough. + + This version has better performance with supplied length, and also + supports string containing null characters. + + \tparam CharType character type of the string + \param str Constant string, lifetime assumed to be longer than the use of the string in e.g. a GenericValue + \param length The length of source string. + \return GenericStringRef string reference object + \relatesalso GenericStringRef +*/ +template +inline GenericStringRef StringRef(const CharType* str, size_t length) { + return GenericStringRef(str, SizeType(length)); +} + +#if RAPIDJSON_HAS_STDSTRING +//! Mark a string object as constant string +/*! Mark a string object (e.g. \c std::string) as a "string literal". + This function can be used to avoid copying a string to be referenced as a + value in a JSON GenericValue object, if the string's lifetime is known + to be valid long enough. + + \tparam CharType character type of the string + \param str Constant string, lifetime assumed to be longer than the use of the string in e.g. a GenericValue + \return GenericStringRef string reference object + \relatesalso GenericStringRef + \note Requires the definition of the preprocessor symbol \ref RAPIDJSON_HAS_STDSTRING. +*/ +template +inline GenericStringRef StringRef(const std::basic_string& str) { + return GenericStringRef(str.data(), SizeType(str.size())); +} +#endif + +/////////////////////////////////////////////////////////////////////////////// +// GenericValue type traits +namespace internal { + +template +struct IsGenericValueImpl : FalseType {}; + +// select candidates according to nested encoding and allocator types +template struct IsGenericValueImpl::Type, typename Void::Type> + : IsBaseOf, T>::Type {}; + +// helper to match arbitrary GenericValue instantiations, including derived classes +template struct IsGenericValue : IsGenericValueImpl::Type {}; + +} // namespace internal + +/////////////////////////////////////////////////////////////////////////////// +// TypeHelper + +namespace internal { + +template +struct TypeHelper {}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsBool(); } + static bool Get(const ValueType& v) { return v.GetBool(); } + static ValueType& Set(ValueType& v, bool data) { return v.SetBool(data); } + static ValueType& Set(ValueType& v, bool data, typename ValueType::AllocatorType&) { return v.SetBool(data); } +}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsInt(); } + static int Get(const ValueType& v) { return v.GetInt(); } + static ValueType& Set(ValueType& v, int data) { return v.SetInt(data); } + static ValueType& Set(ValueType& v, int data, typename ValueType::AllocatorType&) { return v.SetInt(data); } +}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsUint(); } + static unsigned Get(const ValueType& v) { return v.GetUint(); } + static ValueType& Set(ValueType& v, unsigned data) { return v.SetUint(data); } + static ValueType& Set(ValueType& v, unsigned data, typename ValueType::AllocatorType&) { return v.SetUint(data); } +}; + +#ifdef _MSC_VER +RAPIDJSON_STATIC_ASSERT(sizeof(long) == sizeof(int)); +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsInt(); } + static long Get(const ValueType& v) { return v.GetInt(); } + static ValueType& Set(ValueType& v, long data) { return v.SetInt(data); } + static ValueType& Set(ValueType& v, long data, typename ValueType::AllocatorType&) { return v.SetInt(data); } +}; + +RAPIDJSON_STATIC_ASSERT(sizeof(unsigned long) == sizeof(unsigned)); +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsUint(); } + static unsigned long Get(const ValueType& v) { return v.GetUint(); } + static ValueType& Set(ValueType& v, unsigned long data) { return v.SetUint(data); } + static ValueType& Set(ValueType& v, unsigned long data, typename ValueType::AllocatorType&) { return v.SetUint(data); } +}; +#endif + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsInt64(); } + static int64_t Get(const ValueType& v) { return v.GetInt64(); } + static ValueType& Set(ValueType& v, int64_t data) { return v.SetInt64(data); } + static ValueType& Set(ValueType& v, int64_t data, typename ValueType::AllocatorType&) { return v.SetInt64(data); } +}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsUint64(); } + static uint64_t Get(const ValueType& v) { return v.GetUint64(); } + static ValueType& Set(ValueType& v, uint64_t data) { return v.SetUint64(data); } + static ValueType& Set(ValueType& v, uint64_t data, typename ValueType::AllocatorType&) { return v.SetUint64(data); } +}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsDouble(); } + static double Get(const ValueType& v) { return v.GetDouble(); } + static ValueType& Set(ValueType& v, double data) { return v.SetDouble(data); } + static ValueType& Set(ValueType& v, double data, typename ValueType::AllocatorType&) { return v.SetDouble(data); } +}; + +template +struct TypeHelper { + static bool Is(const ValueType& v) { return v.IsFloat(); } + static float Get(const ValueType& v) { return v.GetFloat(); } + static ValueType& Set(ValueType& v, float data) { return v.SetFloat(data); } + static ValueType& Set(ValueType& v, float data, typename ValueType::AllocatorType&) { return v.SetFloat(data); } +}; + +template +struct TypeHelper { + typedef const typename ValueType::Ch* StringType; + static bool Is(const ValueType& v) { return v.IsString(); } + static StringType Get(const ValueType& v) { return v.GetString(); } + static ValueType& Set(ValueType& v, const StringType data) { return v.SetString(typename ValueType::StringRefType(data)); } + static ValueType& Set(ValueType& v, const StringType data, typename ValueType::AllocatorType& a) { return v.SetString(data, a); } +}; + +#if RAPIDJSON_HAS_STDSTRING +template +struct TypeHelper > { + typedef std::basic_string StringType; + static bool Is(const ValueType& v) { return v.IsString(); } + static StringType Get(const ValueType& v) { return StringType(v.GetString(), v.GetStringLength()); } + static ValueType& Set(ValueType& v, const StringType& data, typename ValueType::AllocatorType& a) { return v.SetString(data, a); } +}; +#endif + +template +struct TypeHelper { + typedef typename ValueType::Array ArrayType; + static bool Is(const ValueType& v) { return v.IsArray(); } + static ArrayType Get(ValueType& v) { return v.GetArray(); } + static ValueType& Set(ValueType& v, ArrayType data) { return v = data; } + static ValueType& Set(ValueType& v, ArrayType data, typename ValueType::AllocatorType&) { return v = data; } +}; + +template +struct TypeHelper { + typedef typename ValueType::ConstArray ArrayType; + static bool Is(const ValueType& v) { return v.IsArray(); } + static ArrayType Get(const ValueType& v) { return v.GetArray(); } +}; + +template +struct TypeHelper { + typedef typename ValueType::Object ObjectType; + static bool Is(const ValueType& v) { return v.IsObject(); } + static ObjectType Get(ValueType& v) { return v.GetObject(); } + static ValueType& Set(ValueType& v, ObjectType data) { return v = data; } + static ValueType& Set(ValueType& v, ObjectType data, typename ValueType::AllocatorType&) { return v = data; } +}; + +template +struct TypeHelper { + typedef typename ValueType::ConstObject ObjectType; + static bool Is(const ValueType& v) { return v.IsObject(); } + static ObjectType Get(const ValueType& v) { return v.GetObject(); } +}; + +} // namespace internal + +// Forward declarations +template class GenericArray; +template class GenericObject; + +/////////////////////////////////////////////////////////////////////////////// +// GenericValue + +//! Represents a JSON value. Use Value for UTF8 encoding and default allocator. +/*! + A JSON value can be one of 7 types. This class is a variant type supporting + these types. + + Use the Value if UTF8 and default allocator + + \tparam Encoding Encoding of the value. (Even non-string values need to have the same encoding in a document) + \tparam Allocator Allocator type for allocating memory of object, array and string. +*/ +template +class GenericValue { +public: + //! Name-value pair in an object. + typedef GenericMember Member; + typedef Encoding EncodingType; //!< Encoding type from template parameter. + typedef Allocator AllocatorType; //!< Allocator type from template parameter. + typedef typename Encoding::Ch Ch; //!< Character type derived from Encoding. + typedef GenericStringRef StringRefType; //!< Reference to a constant string + typedef typename GenericMemberIterator::Iterator MemberIterator; //!< Member iterator for iterating in object. + typedef typename GenericMemberIterator::Iterator ConstMemberIterator; //!< Constant member iterator for iterating in object. + typedef GenericValue* ValueIterator; //!< Value iterator for iterating in array. + typedef const GenericValue* ConstValueIterator; //!< Constant value iterator for iterating in array. + typedef GenericValue ValueType; //!< Value type of itself. + typedef GenericArray Array; + typedef GenericArray ConstArray; + typedef GenericObject Object; + typedef GenericObject ConstObject; + + //!@name Constructors and destructor. + //@{ + + //! Default constructor creates a null value. + GenericValue() RAPIDJSON_NOEXCEPT : data_() { data_.f.flags = kNullFlag; } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move constructor in C++11 + GenericValue(GenericValue&& rhs) RAPIDJSON_NOEXCEPT : data_(rhs.data_) { + rhs.data_.f.flags = kNullFlag; // give up contents + } +#endif + +private: + //! Copy constructor is not permitted. + GenericValue(const GenericValue& rhs); + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Moving from a GenericDocument is not permitted. + template + GenericValue(GenericDocument&& rhs); + + //! Move assignment from a GenericDocument is not permitted. + template + GenericValue& operator=(GenericDocument&& rhs); +#endif + +public: + + //! Constructor with JSON value type. + /*! This creates a Value of specified type with default content. + \param type Type of the value. + \note Default content for number is zero. + */ + explicit GenericValue(Type type) RAPIDJSON_NOEXCEPT : data_() { + static const uint16_t defaultFlags[] = { + kNullFlag, kFalseFlag, kTrueFlag, kObjectFlag, kArrayFlag, kShortStringFlag, + kNumberAnyFlag + }; + RAPIDJSON_NOEXCEPT_ASSERT(type >= kNullType && type <= kNumberType); + data_.f.flags = defaultFlags[type]; + + // Use ShortString to store empty string. + if (type == kStringType) + data_.ss.SetLength(0); + } + + //! Explicit copy constructor (with allocator) + /*! Creates a copy of a Value by using the given Allocator + \tparam SourceAllocator allocator of \c rhs + \param rhs Value to copy from (read-only) + \param allocator Allocator for allocating copied elements and buffers. Commonly use GenericDocument::GetAllocator(). + \param copyConstStrings Force copying of constant strings (e.g. referencing an in-situ buffer) + \see CopyFrom() + */ + template + GenericValue(const GenericValue& rhs, Allocator& allocator, bool copyConstStrings = false) { + switch (rhs.GetType()) { + case kObjectType: + DoCopyMembers(rhs, allocator, copyConstStrings); + break; + case kArrayType: { + SizeType count = rhs.data_.a.size; + GenericValue* le = reinterpret_cast(allocator.Malloc(count * sizeof(GenericValue))); + const GenericValue* re = rhs.GetElementsPointer(); + for (SizeType i = 0; i < count; i++) + new (&le[i]) GenericValue(re[i], allocator, copyConstStrings); + data_.f.flags = kArrayFlag; + data_.a.size = data_.a.capacity = count; + SetElementsPointer(le); + } + break; + case kStringType: + if (rhs.data_.f.flags == kConstStringFlag && !copyConstStrings) { + data_.f.flags = rhs.data_.f.flags; + data_ = *reinterpret_cast(&rhs.data_); + } + else + SetStringRaw(StringRef(rhs.GetString(), rhs.GetStringLength()), allocator); + break; + default: + data_.f.flags = rhs.data_.f.flags; + data_ = *reinterpret_cast(&rhs.data_); + break; + } + } + + //! Constructor for boolean value. + /*! \param b Boolean value + \note This constructor is limited to \em real boolean values and rejects + implicitly converted types like arbitrary pointers. Use an explicit cast + to \c bool, if you want to construct a boolean JSON value in such cases. + */ +#ifndef RAPIDJSON_DOXYGEN_RUNNING // hide SFINAE from Doxygen + template + explicit GenericValue(T b, RAPIDJSON_ENABLEIF((internal::IsSame))) RAPIDJSON_NOEXCEPT // See #472 +#else + explicit GenericValue(bool b) RAPIDJSON_NOEXCEPT +#endif + : data_() { + // safe-guard against failing SFINAE + RAPIDJSON_STATIC_ASSERT((internal::IsSame::Value)); + data_.f.flags = b ? kTrueFlag : kFalseFlag; + } + + //! Constructor for int value. + explicit GenericValue(int i) RAPIDJSON_NOEXCEPT : data_() { + data_.n.i64 = i; + data_.f.flags = (i >= 0) ? (kNumberIntFlag | kUintFlag | kUint64Flag) : kNumberIntFlag; + } + + //! Constructor for unsigned value. + explicit GenericValue(unsigned u) RAPIDJSON_NOEXCEPT : data_() { + data_.n.u64 = u; + data_.f.flags = (u & 0x80000000) ? kNumberUintFlag : (kNumberUintFlag | kIntFlag | kInt64Flag); + } + + //! Constructor for int64_t value. + explicit GenericValue(int64_t i64) RAPIDJSON_NOEXCEPT : data_() { + data_.n.i64 = i64; + data_.f.flags = kNumberInt64Flag; + if (i64 >= 0) { + data_.f.flags |= kNumberUint64Flag; + if (!(static_cast(i64) & RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x00000000))) + data_.f.flags |= kUintFlag; + if (!(static_cast(i64) & RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x80000000))) + data_.f.flags |= kIntFlag; + } + else if (i64 >= static_cast(RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x80000000))) + data_.f.flags |= kIntFlag; + } + + //! Constructor for uint64_t value. + explicit GenericValue(uint64_t u64) RAPIDJSON_NOEXCEPT : data_() { + data_.n.u64 = u64; + data_.f.flags = kNumberUint64Flag; + if (!(u64 & RAPIDJSON_UINT64_C2(0x80000000, 0x00000000))) + data_.f.flags |= kInt64Flag; + if (!(u64 & RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x00000000))) + data_.f.flags |= kUintFlag; + if (!(u64 & RAPIDJSON_UINT64_C2(0xFFFFFFFF, 0x80000000))) + data_.f.flags |= kIntFlag; + } + + //! Constructor for double value. + explicit GenericValue(double d) RAPIDJSON_NOEXCEPT : data_() { data_.n.d = d; data_.f.flags = kNumberDoubleFlag; } + + //! Constructor for float value. + explicit GenericValue(float f) RAPIDJSON_NOEXCEPT : data_() { data_.n.d = static_cast(f); data_.f.flags = kNumberDoubleFlag; } + + //! Constructor for constant string (i.e. do not make a copy of string) + GenericValue(const Ch* s, SizeType length) RAPIDJSON_NOEXCEPT : data_() { SetStringRaw(StringRef(s, length)); } + + //! Constructor for constant string (i.e. do not make a copy of string) + explicit GenericValue(StringRefType s) RAPIDJSON_NOEXCEPT : data_() { SetStringRaw(s); } + + //! Constructor for copy-string (i.e. do make a copy of string) + GenericValue(const Ch* s, SizeType length, Allocator& allocator) : data_() { SetStringRaw(StringRef(s, length), allocator); } + + //! Constructor for copy-string (i.e. do make a copy of string) + GenericValue(const Ch*s, Allocator& allocator) : data_() { SetStringRaw(StringRef(s), allocator); } + +#if RAPIDJSON_HAS_STDSTRING + //! Constructor for copy-string from a string object (i.e. do make a copy of string) + /*! \note Requires the definition of the preprocessor symbol \ref RAPIDJSON_HAS_STDSTRING. + */ + GenericValue(const std::basic_string& s, Allocator& allocator) : data_() { SetStringRaw(StringRef(s), allocator); } +#endif + + //! Constructor for Array. + /*! + \param a An array obtained by \c GetArray(). + \note \c Array is always pass-by-value. + \note the source array is moved into this value and the sourec array becomes empty. + */ + GenericValue(Array a) RAPIDJSON_NOEXCEPT : data_(a.value_.data_) { + a.value_.data_ = Data(); + a.value_.data_.f.flags = kArrayFlag; + } + + //! Constructor for Object. + /*! + \param o An object obtained by \c GetObject(). + \note \c Object is always pass-by-value. + \note the source object is moved into this value and the sourec object becomes empty. + */ + GenericValue(Object o) RAPIDJSON_NOEXCEPT : data_(o.value_.data_) { + o.value_.data_ = Data(); + o.value_.data_.f.flags = kObjectFlag; + } + + //! Destructor. + /*! Need to destruct elements of array, members of object, or copy-string. + */ + ~GenericValue() { + // With RAPIDJSON_USE_MEMBERSMAP, the maps need to be destroyed to release + // their Allocator if it's refcounted (e.g. MemoryPoolAllocator). + if (Allocator::kNeedFree || (RAPIDJSON_USE_MEMBERSMAP+0 && + internal::IsRefCounted::Value)) { + switch(data_.f.flags) { + case kArrayFlag: + { + GenericValue* e = GetElementsPointer(); + for (GenericValue* v = e; v != e + data_.a.size; ++v) + v->~GenericValue(); + if (Allocator::kNeedFree) { // Shortcut by Allocator's trait + Allocator::Free(e); + } + } + break; + + case kObjectFlag: + DoFreeMembers(); + break; + + case kCopyStringFlag: + if (Allocator::kNeedFree) { // Shortcut by Allocator's trait + Allocator::Free(const_cast(GetStringPointer())); + } + break; + + default: + break; // Do nothing for other types. + } + } + } + + //@} + + //!@name Assignment operators + //@{ + + //! Assignment with move semantics. + /*! \param rhs Source of the assignment. It will become a null value after assignment. + */ + GenericValue& operator=(GenericValue& rhs) RAPIDJSON_NOEXCEPT { + if (RAPIDJSON_LIKELY(this != &rhs)) { + // Can't destroy "this" before assigning "rhs", otherwise "rhs" + // could be used after free if it's an sub-Value of "this", + // hence the temporary danse. + GenericValue temp; + temp.RawAssign(rhs); + this->~GenericValue(); + RawAssign(temp); + } + return *this; + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move assignment in C++11 + GenericValue& operator=(GenericValue&& rhs) RAPIDJSON_NOEXCEPT { + return *this = rhs.Move(); + } +#endif + + //! Assignment of constant string reference (no copy) + /*! \param str Constant string reference to be assigned + \note This overload is needed to avoid clashes with the generic primitive type assignment overload below. + \see GenericStringRef, operator=(T) + */ + GenericValue& operator=(StringRefType str) RAPIDJSON_NOEXCEPT { + GenericValue s(str); + return *this = s; + } + + //! Assignment with primitive types. + /*! \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t + \param value The value to be assigned. + + \note The source type \c T explicitly disallows all pointer types, + especially (\c const) \ref Ch*. This helps avoiding implicitly + referencing character strings with insufficient lifetime, use + \ref SetString(const Ch*, Allocator&) (for copying) or + \ref StringRef() (to explicitly mark the pointer as constant) instead. + All other pointer types would implicitly convert to \c bool, + use \ref SetBool() instead. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::IsPointer), (GenericValue&)) + operator=(T value) { + GenericValue v(value); + return *this = v; + } + + //! Deep-copy assignment from Value + /*! Assigns a \b copy of the Value to the current Value object + \tparam SourceAllocator Allocator type of \c rhs + \param rhs Value to copy from (read-only) + \param allocator Allocator to use for copying + \param copyConstStrings Force copying of constant strings (e.g. referencing an in-situ buffer) + */ + template + GenericValue& CopyFrom(const GenericValue& rhs, Allocator& allocator, bool copyConstStrings = false) { + RAPIDJSON_ASSERT(static_cast(this) != static_cast(&rhs)); + this->~GenericValue(); + new (this) GenericValue(rhs, allocator, copyConstStrings); + return *this; + } + + //! Exchange the contents of this value with those of other. + /*! + \param other Another value. + \note Constant complexity. + */ + GenericValue& Swap(GenericValue& other) RAPIDJSON_NOEXCEPT { + GenericValue temp; + temp.RawAssign(*this); + RawAssign(other); + other.RawAssign(temp); + return *this; + } + + //! free-standing swap function helper + /*! + Helper function to enable support for common swap implementation pattern based on \c std::swap: + \code + void swap(MyClass& a, MyClass& b) { + using std::swap; + swap(a.value, b.value); + // ... + } + \endcode + \see Swap() + */ + friend inline void swap(GenericValue& a, GenericValue& b) RAPIDJSON_NOEXCEPT { a.Swap(b); } + + //! Prepare Value for move semantics + /*! \return *this */ + GenericValue& Move() RAPIDJSON_NOEXCEPT { return *this; } + //@} + + //!@name Equal-to and not-equal-to operators + //@{ + //! Equal-to operator + /*! + \note If an object contains duplicated named member, comparing equality with any object is always \c false. + \note Complexity is quadratic in Object's member number and linear for the rest (number of all values in the subtree and total lengths of all strings). + */ + template + bool operator==(const GenericValue& rhs) const { + typedef GenericValue RhsType; + if (GetType() != rhs.GetType()) + return false; + + switch (GetType()) { + case kObjectType: // Warning: O(n^2) inner-loop + if (data_.o.size != rhs.data_.o.size) + return false; + for (ConstMemberIterator lhsMemberItr = MemberBegin(); lhsMemberItr != MemberEnd(); ++lhsMemberItr) { + typename RhsType::ConstMemberIterator rhsMemberItr = rhs.FindMember(lhsMemberItr->name); + if (rhsMemberItr == rhs.MemberEnd() || (!(lhsMemberItr->value == rhsMemberItr->value))) + return false; + } + return true; + + case kArrayType: + if (data_.a.size != rhs.data_.a.size) + return false; + for (SizeType i = 0; i < data_.a.size; i++) + if (!((*this)[i] == rhs[i])) + return false; + return true; + + case kStringType: + return StringEqual(rhs); + + case kNumberType: + if (IsDouble() || rhs.IsDouble()) { + double a = GetDouble(); // May convert from integer to double. + double b = rhs.GetDouble(); // Ditto + return a >= b && a <= b; // Prevent -Wfloat-equal + } + else + return data_.n.u64 == rhs.data_.n.u64; + + default: + return true; + } + } + + //! Equal-to operator with const C-string pointer + bool operator==(const Ch* rhs) const { return *this == GenericValue(StringRef(rhs)); } + +#if RAPIDJSON_HAS_STDSTRING + //! Equal-to operator with string object + /*! \note Requires the definition of the preprocessor symbol \ref RAPIDJSON_HAS_STDSTRING. + */ + bool operator==(const std::basic_string& rhs) const { return *this == GenericValue(StringRef(rhs)); } +#endif + + //! Equal-to operator with primitive types + /*! \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t, \c double, \c true, \c false + */ + template RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr,internal::IsGenericValue >), (bool)) operator==(const T& rhs) const { return *this == GenericValue(rhs); } + +#ifndef __cpp_impl_three_way_comparison + //! Not-equal-to operator + /*! \return !(*this == rhs) + */ + template + bool operator!=(const GenericValue& rhs) const { return !(*this == rhs); } + + //! Not-equal-to operator with const C-string pointer + bool operator!=(const Ch* rhs) const { return !(*this == rhs); } + + //! Not-equal-to operator with arbitrary types + /*! \return !(*this == rhs) + */ + template RAPIDJSON_DISABLEIF_RETURN((internal::IsGenericValue), (bool)) operator!=(const T& rhs) const { return !(*this == rhs); } + + //! Equal-to operator with arbitrary types (symmetric version) + /*! \return (rhs == lhs) + */ + template friend RAPIDJSON_DISABLEIF_RETURN((internal::IsGenericValue), (bool)) operator==(const T& lhs, const GenericValue& rhs) { return rhs == lhs; } + + //! Not-Equal-to operator with arbitrary types (symmetric version) + /*! \return !(rhs == lhs) + */ + template friend RAPIDJSON_DISABLEIF_RETURN((internal::IsGenericValue), (bool)) operator!=(const T& lhs, const GenericValue& rhs) { return !(rhs == lhs); } + //@} +#endif + + //!@name Type + //@{ + + Type GetType() const { return static_cast(data_.f.flags & kTypeMask); } + bool IsNull() const { return data_.f.flags == kNullFlag; } + bool IsFalse() const { return data_.f.flags == kFalseFlag; } + bool IsTrue() const { return data_.f.flags == kTrueFlag; } + bool IsBool() const { return (data_.f.flags & kBoolFlag) != 0; } + bool IsObject() const { return data_.f.flags == kObjectFlag; } + bool IsArray() const { return data_.f.flags == kArrayFlag; } + bool IsNumber() const { return (data_.f.flags & kNumberFlag) != 0; } + bool IsInt() const { return (data_.f.flags & kIntFlag) != 0; } + bool IsUint() const { return (data_.f.flags & kUintFlag) != 0; } + bool IsInt64() const { return (data_.f.flags & kInt64Flag) != 0; } + bool IsUint64() const { return (data_.f.flags & kUint64Flag) != 0; } + bool IsDouble() const { return (data_.f.flags & kDoubleFlag) != 0; } + bool IsString() const { return (data_.f.flags & kStringFlag) != 0; } + + // Checks whether a number can be losslessly converted to a double. + bool IsLosslessDouble() const { + if (!IsNumber()) return false; + if (IsUint64()) { + uint64_t u = GetUint64(); + volatile double d = static_cast(u); + return (d >= 0.0) + && (d < static_cast((std::numeric_limits::max)())) + && (u == static_cast(d)); + } + if (IsInt64()) { + int64_t i = GetInt64(); + volatile double d = static_cast(i); + return (d >= static_cast((std::numeric_limits::min)())) + && (d < static_cast((std::numeric_limits::max)())) + && (i == static_cast(d)); + } + return true; // double, int, uint are always lossless + } + + // Checks whether a number is a float (possible lossy). + bool IsFloat() const { + if ((data_.f.flags & kDoubleFlag) == 0) + return false; + double d = GetDouble(); + return d >= -3.4028234e38 && d <= 3.4028234e38; + } + // Checks whether a number can be losslessly converted to a float. + bool IsLosslessFloat() const { + if (!IsNumber()) return false; + double a = GetDouble(); + if (a < static_cast(-(std::numeric_limits::max)()) + || a > static_cast((std::numeric_limits::max)())) + return false; + double b = static_cast(static_cast(a)); + return a >= b && a <= b; // Prevent -Wfloat-equal + } + + //@} + + //!@name Null + //@{ + + GenericValue& SetNull() { this->~GenericValue(); new (this) GenericValue(); return *this; } + + //@} + + //!@name Bool + //@{ + + bool GetBool() const { RAPIDJSON_ASSERT(IsBool()); return data_.f.flags == kTrueFlag; } + //!< Set boolean value + /*! \post IsBool() == true */ + GenericValue& SetBool(bool b) { this->~GenericValue(); new (this) GenericValue(b); return *this; } + + //@} + + //!@name Object + //@{ + + //! Set this value as an empty object. + /*! \post IsObject() == true */ + GenericValue& SetObject() { this->~GenericValue(); new (this) GenericValue(kObjectType); return *this; } + + //! Get the number of members in the object. + SizeType MemberCount() const { RAPIDJSON_ASSERT(IsObject()); return data_.o.size; } + + //! Get the capacity of object. + SizeType MemberCapacity() const { RAPIDJSON_ASSERT(IsObject()); return data_.o.capacity; } + + //! Check whether the object is empty. + bool ObjectEmpty() const { RAPIDJSON_ASSERT(IsObject()); return data_.o.size == 0; } + + //! Get a value from an object associated with the name. + /*! \pre IsObject() == true + \tparam T Either \c Ch or \c const \c Ch (template used for disambiguation with \ref operator[](SizeType)) + \note In version 0.1x, if the member is not found, this function returns a null value. This makes issue 7. + Since 0.2, if the name is not correct, it will assert. + If user is unsure whether a member exists, user should use HasMember() first. + A better approach is to use FindMember(). + \note Linear time complexity. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::NotExpr::Type, Ch> >),(GenericValue&)) operator[](T* name) { + GenericValue n(StringRef(name)); + return (*this)[n]; + } + template + RAPIDJSON_DISABLEIF_RETURN((internal::NotExpr::Type, Ch> >),(const GenericValue&)) operator[](T* name) const { return const_cast(*this)[name]; } + + //! Get a value from an object associated with the name. + /*! \pre IsObject() == true + \tparam SourceAllocator Allocator of the \c name value + + \note Compared to \ref operator[](T*), this version is faster because it does not need a StrLen(). + And it can also handle strings with embedded null characters. + + \note Linear time complexity. + */ + template + GenericValue& operator[](const GenericValue& name) { + MemberIterator member = FindMember(name); + if (member != MemberEnd()) + return member->value; + else { + RAPIDJSON_ASSERT(false); // see above note + +#if RAPIDJSON_HAS_CXX11 + // Use thread-local storage to prevent races between threads. + // Use static buffer and placement-new to prevent destruction, with + // alignas() to ensure proper alignment. + alignas(GenericValue) thread_local static char buffer[sizeof(GenericValue)]; + return *new (buffer) GenericValue(); +#elif defined(_MSC_VER) && _MSC_VER < 1900 + // There's no way to solve both thread locality and proper alignment + // simultaneously. + __declspec(thread) static char buffer[sizeof(GenericValue)]; + return *new (buffer) GenericValue(); +#elif defined(__GNUC__) || defined(__clang__) + // This will generate -Wexit-time-destructors in clang, but that's + // better than having under-alignment. + __thread static GenericValue buffer; + return buffer; +#else + // Don't know what compiler this is, so don't know how to ensure + // thread-locality. + static GenericValue buffer; + return buffer; +#endif + } + } + template + const GenericValue& operator[](const GenericValue& name) const { return const_cast(*this)[name]; } + +#if RAPIDJSON_HAS_STDSTRING + //! Get a value from an object associated with name (string object). + GenericValue& operator[](const std::basic_string& name) { return (*this)[GenericValue(StringRef(name))]; } + const GenericValue& operator[](const std::basic_string& name) const { return (*this)[GenericValue(StringRef(name))]; } +#endif + + //! Const member iterator + /*! \pre IsObject() == true */ + ConstMemberIterator MemberBegin() const { RAPIDJSON_ASSERT(IsObject()); return ConstMemberIterator(GetMembersPointer()); } + //! Const \em past-the-end member iterator + /*! \pre IsObject() == true */ + ConstMemberIterator MemberEnd() const { RAPIDJSON_ASSERT(IsObject()); return ConstMemberIterator(GetMembersPointer() + data_.o.size); } + //! Member iterator + /*! \pre IsObject() == true */ + MemberIterator MemberBegin() { RAPIDJSON_ASSERT(IsObject()); return MemberIterator(GetMembersPointer()); } + //! \em Past-the-end member iterator + /*! \pre IsObject() == true */ + MemberIterator MemberEnd() { RAPIDJSON_ASSERT(IsObject()); return MemberIterator(GetMembersPointer() + data_.o.size); } + + //! Request the object to have enough capacity to store members. + /*! \param newCapacity The capacity that the object at least need to have. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \note Linear time complexity. + */ + GenericValue& MemberReserve(SizeType newCapacity, Allocator &allocator) { + RAPIDJSON_ASSERT(IsObject()); + DoReserveMembers(newCapacity, allocator); + return *this; + } + + //! Check whether a member exists in the object. + /*! + \param name Member name to be searched. + \pre IsObject() == true + \return Whether a member with that name exists. + \note It is better to use FindMember() directly if you need the obtain the value as well. + \note Linear time complexity. + */ + bool HasMember(const Ch* name) const { return FindMember(name) != MemberEnd(); } + +#if RAPIDJSON_HAS_STDSTRING + //! Check whether a member exists in the object with string object. + /*! + \param name Member name to be searched. + \pre IsObject() == true + \return Whether a member with that name exists. + \note It is better to use FindMember() directly if you need the obtain the value as well. + \note Linear time complexity. + */ + bool HasMember(const std::basic_string& name) const { return FindMember(name) != MemberEnd(); } +#endif + + //! Check whether a member exists in the object with GenericValue name. + /*! + This version is faster because it does not need a StrLen(). It can also handle string with null character. + \param name Member name to be searched. + \pre IsObject() == true + \return Whether a member with that name exists. + \note It is better to use FindMember() directly if you need the obtain the value as well. + \note Linear time complexity. + */ + template + bool HasMember(const GenericValue& name) const { return FindMember(name) != MemberEnd(); } + + //! Find member by name. + /*! + \param name Member name to be searched. + \pre IsObject() == true + \return Iterator to member, if it exists. + Otherwise returns \ref MemberEnd(). + + \note Earlier versions of Rapidjson returned a \c NULL pointer, in case + the requested member doesn't exist. For consistency with e.g. + \c std::map, this has been changed to MemberEnd() now. + \note Linear time complexity. + */ + MemberIterator FindMember(const Ch* name) { + GenericValue n(StringRef(name)); + return FindMember(n); + } + + ConstMemberIterator FindMember(const Ch* name) const { return const_cast(*this).FindMember(name); } + + //! Find member by name. + /*! + This version is faster because it does not need a StrLen(). It can also handle string with null character. + \param name Member name to be searched. + \pre IsObject() == true + \return Iterator to member, if it exists. + Otherwise returns \ref MemberEnd(). + + \note Earlier versions of Rapidjson returned a \c NULL pointer, in case + the requested member doesn't exist. For consistency with e.g. + \c std::map, this has been changed to MemberEnd() now. + \note Linear time complexity. + */ + template + MemberIterator FindMember(const GenericValue& name) { + RAPIDJSON_ASSERT(IsObject()); + RAPIDJSON_ASSERT(name.IsString()); + return DoFindMember(name); + } + template ConstMemberIterator FindMember(const GenericValue& name) const { return const_cast(*this).FindMember(name); } + +#if RAPIDJSON_HAS_STDSTRING + //! Find member by string object name. + /*! + \param name Member name to be searched. + \pre IsObject() == true + \return Iterator to member, if it exists. + Otherwise returns \ref MemberEnd(). + */ + MemberIterator FindMember(const std::basic_string& name) { return FindMember(GenericValue(StringRef(name))); } + ConstMemberIterator FindMember(const std::basic_string& name) const { return FindMember(GenericValue(StringRef(name))); } +#endif + + //! Add a member (name-value pair) to the object. + /*! \param name A string value as name of member. + \param value Value of any type. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \note The ownership of \c name and \c value will be transferred to this object on success. + \pre IsObject() && name.IsString() + \post name.IsNull() && value.IsNull() + \note Amortized Constant time complexity. + */ + GenericValue& AddMember(GenericValue& name, GenericValue& value, Allocator& allocator) { + RAPIDJSON_ASSERT(IsObject()); + RAPIDJSON_ASSERT(name.IsString()); + DoAddMember(name, value, allocator); + return *this; + } + + //! Add a constant string value as member (name-value pair) to the object. + /*! \param name A string value as name of member. + \param value constant string reference as value of member. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \pre IsObject() + \note This overload is needed to avoid clashes with the generic primitive type AddMember(GenericValue&,T,Allocator&) overload below. + \note Amortized Constant time complexity. + */ + GenericValue& AddMember(GenericValue& name, StringRefType value, Allocator& allocator) { + GenericValue v(value); + return AddMember(name, v, allocator); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Add a string object as member (name-value pair) to the object. + /*! \param name A string value as name of member. + \param value constant string reference as value of member. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \pre IsObject() + \note This overload is needed to avoid clashes with the generic primitive type AddMember(GenericValue&,T,Allocator&) overload below. + \note Amortized Constant time complexity. + */ + GenericValue& AddMember(GenericValue& name, std::basic_string& value, Allocator& allocator) { + GenericValue v(value, allocator); + return AddMember(name, v, allocator); + } +#endif + + //! Add any primitive value as member (name-value pair) to the object. + /*! \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t + \param name A string value as name of member. + \param value Value of primitive type \c T as value of member + \param allocator Allocator for reallocating memory. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \pre IsObject() + + \note The source type \c T explicitly disallows all pointer types, + especially (\c const) \ref Ch*. This helps avoiding implicitly + referencing character strings with insufficient lifetime, use + \ref AddMember(StringRefType, GenericValue&, Allocator&) or \ref + AddMember(StringRefType, StringRefType, Allocator&). + All other pointer types would implicitly convert to \c bool, + use an explicit cast instead, if needed. + \note Amortized Constant time complexity. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (GenericValue&)) + AddMember(GenericValue& name, T value, Allocator& allocator) { + GenericValue v(value); + return AddMember(name, v, allocator); + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericValue& AddMember(GenericValue&& name, GenericValue&& value, Allocator& allocator) { + return AddMember(name, value, allocator); + } + GenericValue& AddMember(GenericValue&& name, GenericValue& value, Allocator& allocator) { + return AddMember(name, value, allocator); + } + GenericValue& AddMember(GenericValue& name, GenericValue&& value, Allocator& allocator) { + return AddMember(name, value, allocator); + } + GenericValue& AddMember(StringRefType name, GenericValue&& value, Allocator& allocator) { + GenericValue n(name); + return AddMember(n, value, allocator); + } +#endif // RAPIDJSON_HAS_CXX11_RVALUE_REFS + + + //! Add a member (name-value pair) to the object. + /*! \param name A constant string reference as name of member. + \param value Value of any type. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \note The ownership of \c value will be transferred to this object on success. + \pre IsObject() + \post value.IsNull() + \note Amortized Constant time complexity. + */ + GenericValue& AddMember(StringRefType name, GenericValue& value, Allocator& allocator) { + GenericValue n(name); + return AddMember(n, value, allocator); + } + + //! Add a constant string value as member (name-value pair) to the object. + /*! \param name A constant string reference as name of member. + \param value constant string reference as value of member. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \pre IsObject() + \note This overload is needed to avoid clashes with the generic primitive type AddMember(StringRefType,T,Allocator&) overload below. + \note Amortized Constant time complexity. + */ + GenericValue& AddMember(StringRefType name, StringRefType value, Allocator& allocator) { + GenericValue v(value); + return AddMember(name, v, allocator); + } + + //! Add any primitive value as member (name-value pair) to the object. + /*! \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t + \param name A constant string reference as name of member. + \param value Value of primitive type \c T as value of member + \param allocator Allocator for reallocating memory. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \pre IsObject() + + \note The source type \c T explicitly disallows all pointer types, + especially (\c const) \ref Ch*. This helps avoiding implicitly + referencing character strings with insufficient lifetime, use + \ref AddMember(StringRefType, GenericValue&, Allocator&) or \ref + AddMember(StringRefType, StringRefType, Allocator&). + All other pointer types would implicitly convert to \c bool, + use an explicit cast instead, if needed. + \note Amortized Constant time complexity. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (GenericValue&)) + AddMember(StringRefType name, T value, Allocator& allocator) { + GenericValue n(name); + return AddMember(n, value, allocator); + } + + //! Remove all members in the object. + /*! This function do not deallocate memory in the object, i.e. the capacity is unchanged. + \note Linear time complexity. + */ + void RemoveAllMembers() { + RAPIDJSON_ASSERT(IsObject()); + DoClearMembers(); + } + + //! Remove a member in object by its name. + /*! \param name Name of member to be removed. + \return Whether the member existed. + \note This function may reorder the object members. Use \ref + EraseMember(ConstMemberIterator) if you need to preserve the + relative order of the remaining members. + \note Linear time complexity. + */ + bool RemoveMember(const Ch* name) { + GenericValue n(StringRef(name)); + return RemoveMember(n); + } + +#if RAPIDJSON_HAS_STDSTRING + bool RemoveMember(const std::basic_string& name) { return RemoveMember(GenericValue(StringRef(name))); } +#endif + + template + bool RemoveMember(const GenericValue& name) { + MemberIterator m = FindMember(name); + if (m != MemberEnd()) { + RemoveMember(m); + return true; + } + else + return false; + } + + //! Remove a member in object by iterator. + /*! \param m member iterator (obtained by FindMember() or MemberBegin()). + \return the new iterator after removal. + \note This function may reorder the object members. Use \ref + EraseMember(ConstMemberIterator) if you need to preserve the + relative order of the remaining members. + \note Constant time complexity. + */ + MemberIterator RemoveMember(MemberIterator m) { + RAPIDJSON_ASSERT(IsObject()); + RAPIDJSON_ASSERT(data_.o.size > 0); + RAPIDJSON_ASSERT(GetMembersPointer() != 0); + RAPIDJSON_ASSERT(m >= MemberBegin() && m < MemberEnd()); + return DoRemoveMember(m); + } + + //! Remove a member from an object by iterator. + /*! \param pos iterator to the member to remove + \pre IsObject() == true && \ref MemberBegin() <= \c pos < \ref MemberEnd() + \return Iterator following the removed element. + If the iterator \c pos refers to the last element, the \ref MemberEnd() iterator is returned. + \note This function preserves the relative order of the remaining object + members. If you do not need this, use the more efficient \ref RemoveMember(MemberIterator). + \note Linear time complexity. + */ + MemberIterator EraseMember(ConstMemberIterator pos) { + return EraseMember(pos, pos +1); + } + + //! Remove members in the range [first, last) from an object. + /*! \param first iterator to the first member to remove + \param last iterator following the last member to remove + \pre IsObject() == true && \ref MemberBegin() <= \c first <= \c last <= \ref MemberEnd() + \return Iterator following the last removed element. + \note This function preserves the relative order of the remaining object + members. + \note Linear time complexity. + */ + MemberIterator EraseMember(ConstMemberIterator first, ConstMemberIterator last) { + RAPIDJSON_ASSERT(IsObject()); + RAPIDJSON_ASSERT(data_.o.size > 0); + RAPIDJSON_ASSERT(GetMembersPointer() != 0); + RAPIDJSON_ASSERT(first >= MemberBegin()); + RAPIDJSON_ASSERT(first <= last); + RAPIDJSON_ASSERT(last <= MemberEnd()); + return DoEraseMembers(first, last); + } + + //! Erase a member in object by its name. + /*! \param name Name of member to be removed. + \return Whether the member existed. + \note Linear time complexity. + */ + bool EraseMember(const Ch* name) { + GenericValue n(StringRef(name)); + return EraseMember(n); + } + +#if RAPIDJSON_HAS_STDSTRING + bool EraseMember(const std::basic_string& name) { return EraseMember(GenericValue(StringRef(name))); } +#endif + + template + bool EraseMember(const GenericValue& name) { + MemberIterator m = FindMember(name); + if (m != MemberEnd()) { + EraseMember(m); + return true; + } + else + return false; + } + + Object GetObject() { RAPIDJSON_ASSERT(IsObject()); return Object(*this); } + Object GetObj() { RAPIDJSON_ASSERT(IsObject()); return Object(*this); } + ConstObject GetObject() const { RAPIDJSON_ASSERT(IsObject()); return ConstObject(*this); } + ConstObject GetObj() const { RAPIDJSON_ASSERT(IsObject()); return ConstObject(*this); } + + //@} + + //!@name Array + //@{ + + //! Set this value as an empty array. + /*! \post IsArray == true */ + GenericValue& SetArray() { this->~GenericValue(); new (this) GenericValue(kArrayType); return *this; } + + //! Get the number of elements in array. + SizeType Size() const { RAPIDJSON_ASSERT(IsArray()); return data_.a.size; } + + //! Get the capacity of array. + SizeType Capacity() const { RAPIDJSON_ASSERT(IsArray()); return data_.a.capacity; } + + //! Check whether the array is empty. + bool Empty() const { RAPIDJSON_ASSERT(IsArray()); return data_.a.size == 0; } + + //! Remove all elements in the array. + /*! This function do not deallocate memory in the array, i.e. the capacity is unchanged. + \note Linear time complexity. + */ + void Clear() { + RAPIDJSON_ASSERT(IsArray()); + GenericValue* e = GetElementsPointer(); + for (GenericValue* v = e; v != e + data_.a.size; ++v) + v->~GenericValue(); + data_.a.size = 0; + } + + //! Get an element from array by index. + /*! \pre IsArray() == true + \param index Zero-based index of element. + \see operator[](T*) + */ + GenericValue& operator[](SizeType index) { + RAPIDJSON_ASSERT(IsArray()); + RAPIDJSON_ASSERT(index < data_.a.size); + return GetElementsPointer()[index]; + } + const GenericValue& operator[](SizeType index) const { return const_cast(*this)[index]; } + + //! Element iterator + /*! \pre IsArray() == true */ + ValueIterator Begin() { RAPIDJSON_ASSERT(IsArray()); return GetElementsPointer(); } + //! \em Past-the-end element iterator + /*! \pre IsArray() == true */ + ValueIterator End() { RAPIDJSON_ASSERT(IsArray()); return GetElementsPointer() + data_.a.size; } + //! Constant element iterator + /*! \pre IsArray() == true */ + ConstValueIterator Begin() const { return const_cast(*this).Begin(); } + //! Constant \em past-the-end element iterator + /*! \pre IsArray() == true */ + ConstValueIterator End() const { return const_cast(*this).End(); } + + //! Request the array to have enough capacity to store elements. + /*! \param newCapacity The capacity that the array at least need to have. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \note Linear time complexity. + */ + GenericValue& Reserve(SizeType newCapacity, Allocator &allocator) { + RAPIDJSON_ASSERT(IsArray()); + if (newCapacity > data_.a.capacity) { + SetElementsPointer(reinterpret_cast(allocator.Realloc(GetElementsPointer(), data_.a.capacity * sizeof(GenericValue), newCapacity * sizeof(GenericValue)))); + data_.a.capacity = newCapacity; + } + return *this; + } + + //! Append a GenericValue at the end of the array. + /*! \param value Value to be appended. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \pre IsArray() == true + \post value.IsNull() == true + \return The value itself for fluent API. + \note The ownership of \c value will be transferred to this array on success. + \note If the number of elements to be appended is known, calls Reserve() once first may be more efficient. + \note Amortized constant time complexity. + */ + GenericValue& PushBack(GenericValue& value, Allocator& allocator) { + RAPIDJSON_ASSERT(IsArray()); + if (data_.a.size >= data_.a.capacity) + Reserve(data_.a.capacity == 0 ? kDefaultArrayCapacity : (data_.a.capacity + (data_.a.capacity + 1) / 2), allocator); + GetElementsPointer()[data_.a.size++].RawAssign(value); + return *this; + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericValue& PushBack(GenericValue&& value, Allocator& allocator) { + return PushBack(value, allocator); + } +#endif // RAPIDJSON_HAS_CXX11_RVALUE_REFS + + //! Append a constant string reference at the end of the array. + /*! \param value Constant string reference to be appended. + \param allocator Allocator for reallocating memory. It must be the same one used previously. Commonly use GenericDocument::GetAllocator(). + \pre IsArray() == true + \return The value itself for fluent API. + \note If the number of elements to be appended is known, calls Reserve() once first may be more efficient. + \note Amortized constant time complexity. + \see GenericStringRef + */ + GenericValue& PushBack(StringRefType value, Allocator& allocator) { + return (*this).template PushBack(value, allocator); + } + + //! Append a primitive value at the end of the array. + /*! \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t + \param value Value of primitive type T to be appended. + \param allocator Allocator for reallocating memory. It must be the same one as used before. Commonly use GenericDocument::GetAllocator(). + \pre IsArray() == true + \return The value itself for fluent API. + \note If the number of elements to be appended is known, calls Reserve() once first may be more efficient. + + \note The source type \c T explicitly disallows all pointer types, + especially (\c const) \ref Ch*. This helps avoiding implicitly + referencing character strings with insufficient lifetime, use + \ref PushBack(GenericValue&, Allocator&) or \ref + PushBack(StringRefType, Allocator&). + All other pointer types would implicitly convert to \c bool, + use an explicit cast instead, if needed. + \note Amortized constant time complexity. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (GenericValue&)) + PushBack(T value, Allocator& allocator) { + GenericValue v(value); + return PushBack(v, allocator); + } + + //! Remove the last element in the array. + /*! + \note Constant time complexity. + */ + GenericValue& PopBack() { + RAPIDJSON_ASSERT(IsArray()); + RAPIDJSON_ASSERT(!Empty()); + GetElementsPointer()[--data_.a.size].~GenericValue(); + return *this; + } + + //! Remove an element of array by iterator. + /*! + \param pos iterator to the element to remove + \pre IsArray() == true && \ref Begin() <= \c pos < \ref End() + \return Iterator following the removed element. If the iterator pos refers to the last element, the End() iterator is returned. + \note Linear time complexity. + */ + ValueIterator Erase(ConstValueIterator pos) { + return Erase(pos, pos + 1); + } + + //! Remove elements in the range [first, last) of the array. + /*! + \param first iterator to the first element to remove + \param last iterator following the last element to remove + \pre IsArray() == true && \ref Begin() <= \c first <= \c last <= \ref End() + \return Iterator following the last removed element. + \note Linear time complexity. + */ + ValueIterator Erase(ConstValueIterator first, ConstValueIterator last) { + RAPIDJSON_ASSERT(IsArray()); + RAPIDJSON_ASSERT(data_.a.size > 0); + RAPIDJSON_ASSERT(GetElementsPointer() != 0); + RAPIDJSON_ASSERT(first >= Begin()); + RAPIDJSON_ASSERT(first <= last); + RAPIDJSON_ASSERT(last <= End()); + ValueIterator pos = Begin() + (first - Begin()); + for (ValueIterator itr = pos; itr != last; ++itr) + itr->~GenericValue(); + std::memmove(static_cast(pos), last, static_cast(End() - last) * sizeof(GenericValue)); + data_.a.size -= static_cast(last - first); + return pos; + } + + Array GetArray() { RAPIDJSON_ASSERT(IsArray()); return Array(*this); } + ConstArray GetArray() const { RAPIDJSON_ASSERT(IsArray()); return ConstArray(*this); } + + //@} + + //!@name Number + //@{ + + int GetInt() const { RAPIDJSON_ASSERT(data_.f.flags & kIntFlag); return data_.n.i.i; } + unsigned GetUint() const { RAPIDJSON_ASSERT(data_.f.flags & kUintFlag); return data_.n.u.u; } + int64_t GetInt64() const { RAPIDJSON_ASSERT(data_.f.flags & kInt64Flag); return data_.n.i64; } + uint64_t GetUint64() const { RAPIDJSON_ASSERT(data_.f.flags & kUint64Flag); return data_.n.u64; } + + //! Get the value as double type. + /*! \note If the value is 64-bit integer type, it may lose precision. Use \c IsLosslessDouble() to check whether the converison is lossless. + */ + double GetDouble() const { + RAPIDJSON_ASSERT(IsNumber()); + if ((data_.f.flags & kDoubleFlag) != 0) return data_.n.d; // exact type, no conversion. + if ((data_.f.flags & kIntFlag) != 0) return data_.n.i.i; // int -> double + if ((data_.f.flags & kUintFlag) != 0) return data_.n.u.u; // unsigned -> double + if ((data_.f.flags & kInt64Flag) != 0) return static_cast(data_.n.i64); // int64_t -> double (may lose precision) + RAPIDJSON_ASSERT((data_.f.flags & kUint64Flag) != 0); return static_cast(data_.n.u64); // uint64_t -> double (may lose precision) + } + + //! Get the value as float type. + /*! \note If the value is 64-bit integer type, it may lose precision. Use \c IsLosslessFloat() to check whether the converison is lossless. + */ + float GetFloat() const { + return static_cast(GetDouble()); + } + + GenericValue& SetInt(int i) { this->~GenericValue(); new (this) GenericValue(i); return *this; } + GenericValue& SetUint(unsigned u) { this->~GenericValue(); new (this) GenericValue(u); return *this; } + GenericValue& SetInt64(int64_t i64) { this->~GenericValue(); new (this) GenericValue(i64); return *this; } + GenericValue& SetUint64(uint64_t u64) { this->~GenericValue(); new (this) GenericValue(u64); return *this; } + GenericValue& SetDouble(double d) { this->~GenericValue(); new (this) GenericValue(d); return *this; } + GenericValue& SetFloat(float f) { this->~GenericValue(); new (this) GenericValue(static_cast(f)); return *this; } + + //@} + + //!@name String + //@{ + + const Ch* GetString() const { RAPIDJSON_ASSERT(IsString()); return DataString(data_); } + + //! Get the length of string. + /*! Since rapidjson permits "\\u0000" in the json string, strlen(v.GetString()) may not equal to v.GetStringLength(). + */ + SizeType GetStringLength() const { RAPIDJSON_ASSERT(IsString()); return DataStringLength(data_); } + + //! Set this value as a string without copying source string. + /*! This version has better performance with supplied length, and also support string containing null character. + \param s source string pointer. + \param length The length of source string, excluding the trailing null terminator. + \return The value itself for fluent API. + \post IsString() == true && GetString() == s && GetStringLength() == length + \see SetString(StringRefType) + */ + GenericValue& SetString(const Ch* s, SizeType length) { return SetString(StringRef(s, length)); } + + //! Set this value as a string without copying source string. + /*! \param s source string reference + \return The value itself for fluent API. + \post IsString() == true && GetString() == s && GetStringLength() == s.length + */ + GenericValue& SetString(StringRefType s) { this->~GenericValue(); SetStringRaw(s); return *this; } + + //! Set this value as a string by copying from source string. + /*! This version has better performance with supplied length, and also support string containing null character. + \param s source string. + \param length The length of source string, excluding the trailing null terminator. + \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \post IsString() == true && GetString() != s && strcmp(GetString(),s) == 0 && GetStringLength() == length + */ + GenericValue& SetString(const Ch* s, SizeType length, Allocator& allocator) { return SetString(StringRef(s, length), allocator); } + + //! Set this value as a string by copying from source string. + /*! \param s source string. + \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \post IsString() == true && GetString() != s && strcmp(GetString(),s) == 0 && GetStringLength() == length + */ + GenericValue& SetString(const Ch* s, Allocator& allocator) { return SetString(StringRef(s), allocator); } + + //! Set this value as a string by copying from source string. + /*! \param s source string reference + \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \post IsString() == true && GetString() != s.s && strcmp(GetString(),s) == 0 && GetStringLength() == length + */ + GenericValue& SetString(StringRefType s, Allocator& allocator) { this->~GenericValue(); SetStringRaw(s, allocator); return *this; } + +#if RAPIDJSON_HAS_STDSTRING + //! Set this value as a string by copying from source string. + /*! \param s source string. + \param allocator Allocator for allocating copied buffer. Commonly use GenericDocument::GetAllocator(). + \return The value itself for fluent API. + \post IsString() == true && GetString() != s.data() && strcmp(GetString(),s.data() == 0 && GetStringLength() == s.size() + \note Requires the definition of the preprocessor symbol \ref RAPIDJSON_HAS_STDSTRING. + */ + GenericValue& SetString(const std::basic_string& s, Allocator& allocator) { return SetString(StringRef(s), allocator); } +#endif + + //@} + + //!@name Array + //@{ + + //! Templated version for checking whether this value is type T. + /*! + \tparam T Either \c bool, \c int, \c unsigned, \c int64_t, \c uint64_t, \c double, \c float, \c const \c char*, \c std::basic_string + */ + template + bool Is() const { return internal::TypeHelper::Is(*this); } + + template + T Get() const { return internal::TypeHelper::Get(*this); } + + template + T Get() { return internal::TypeHelper::Get(*this); } + + template + ValueType& Set(const T& data) { return internal::TypeHelper::Set(*this, data); } + + template + ValueType& Set(const T& data, AllocatorType& allocator) { return internal::TypeHelper::Set(*this, data, allocator); } + + //@} + + //! Generate events of this value to a Handler. + /*! This function adopts the GoF visitor pattern. + Typical usage is to output this JSON value as JSON text via Writer, which is a Handler. + It can also be used to deep clone this value via GenericDocument, which is also a Handler. + \tparam Handler type of handler. + \param handler An object implementing concept Handler. + */ + template + bool Accept(Handler& handler) const { + switch(GetType()) { + case kNullType: return handler.Null(); + case kFalseType: return handler.Bool(false); + case kTrueType: return handler.Bool(true); + + case kObjectType: + if (RAPIDJSON_UNLIKELY(!handler.StartObject())) + return false; + for (ConstMemberIterator m = MemberBegin(); m != MemberEnd(); ++m) { + RAPIDJSON_ASSERT(m->name.IsString()); // User may change the type of name by MemberIterator. + if (RAPIDJSON_UNLIKELY(!handler.Key(m->name.GetString(), m->name.GetStringLength(), (m->name.data_.f.flags & kCopyFlag) != 0))) + return false; + if (RAPIDJSON_UNLIKELY(!m->value.Accept(handler))) + return false; + } + return handler.EndObject(data_.o.size); + + case kArrayType: + if (RAPIDJSON_UNLIKELY(!handler.StartArray())) + return false; + for (ConstValueIterator v = Begin(); v != End(); ++v) + if (RAPIDJSON_UNLIKELY(!v->Accept(handler))) + return false; + return handler.EndArray(data_.a.size); + + case kStringType: + return handler.String(GetString(), GetStringLength(), (data_.f.flags & kCopyFlag) != 0); + + default: + RAPIDJSON_ASSERT(GetType() == kNumberType); + if (IsDouble()) return handler.Double(data_.n.d); + else if (IsInt()) return handler.Int(data_.n.i.i); + else if (IsUint()) return handler.Uint(data_.n.u.u); + else if (IsInt64()) return handler.Int64(data_.n.i64); + else return handler.Uint64(data_.n.u64); + } + } + +private: + template friend class GenericValue; + template friend class GenericDocument; + + enum { + kBoolFlag = 0x0008, + kNumberFlag = 0x0010, + kIntFlag = 0x0020, + kUintFlag = 0x0040, + kInt64Flag = 0x0080, + kUint64Flag = 0x0100, + kDoubleFlag = 0x0200, + kStringFlag = 0x0400, + kCopyFlag = 0x0800, + kInlineStrFlag = 0x1000, + + // Initial flags of different types. + kNullFlag = kNullType, + // These casts are added to suppress the warning on MSVC about bitwise operations between enums of different types. + kTrueFlag = static_cast(kTrueType) | static_cast(kBoolFlag), + kFalseFlag = static_cast(kFalseType) | static_cast(kBoolFlag), + kNumberIntFlag = static_cast(kNumberType) | static_cast(kNumberFlag | kIntFlag | kInt64Flag), + kNumberUintFlag = static_cast(kNumberType) | static_cast(kNumberFlag | kUintFlag | kUint64Flag | kInt64Flag), + kNumberInt64Flag = static_cast(kNumberType) | static_cast(kNumberFlag | kInt64Flag), + kNumberUint64Flag = static_cast(kNumberType) | static_cast(kNumberFlag | kUint64Flag), + kNumberDoubleFlag = static_cast(kNumberType) | static_cast(kNumberFlag | kDoubleFlag), + kNumberAnyFlag = static_cast(kNumberType) | static_cast(kNumberFlag | kIntFlag | kInt64Flag | kUintFlag | kUint64Flag | kDoubleFlag), + kConstStringFlag = static_cast(kStringType) | static_cast(kStringFlag), + kCopyStringFlag = static_cast(kStringType) | static_cast(kStringFlag | kCopyFlag), + kShortStringFlag = static_cast(kStringType) | static_cast(kStringFlag | kCopyFlag | kInlineStrFlag), + kObjectFlag = kObjectType, + kArrayFlag = kArrayType, + + kTypeMask = 0x07 + }; + + static const SizeType kDefaultArrayCapacity = RAPIDJSON_VALUE_DEFAULT_ARRAY_CAPACITY; + static const SizeType kDefaultObjectCapacity = RAPIDJSON_VALUE_DEFAULT_OBJECT_CAPACITY; + + struct Flag { +#if RAPIDJSON_48BITPOINTER_OPTIMIZATION + char payload[sizeof(SizeType) * 2 + 6]; // 2 x SizeType + lower 48-bit pointer +#elif RAPIDJSON_64BIT + char payload[sizeof(SizeType) * 2 + sizeof(void*) + 6]; // 6 padding bytes +#else + char payload[sizeof(SizeType) * 2 + sizeof(void*) + 2]; // 2 padding bytes +#endif + uint16_t flags; + }; + + struct String { + SizeType length; + SizeType hashcode; //!< reserved + const Ch* str; + }; // 12 bytes in 32-bit mode, 16 bytes in 64-bit mode + + // implementation detail: ShortString can represent zero-terminated strings up to MaxSize chars + // (excluding the terminating zero) and store a value to determine the length of the contained + // string in the last character str[LenPos] by storing "MaxSize - length" there. If the string + // to store has the maximal length of MaxSize then str[LenPos] will be 0 and therefore act as + // the string terminator as well. For getting the string length back from that value just use + // "MaxSize - str[LenPos]". + // This allows to store 13-chars strings in 32-bit mode, 21-chars strings in 64-bit mode, + // 13-chars strings for RAPIDJSON_48BITPOINTER_OPTIMIZATION=1 inline (for `UTF8`-encoded strings). + struct ShortString { + enum { MaxChars = sizeof(static_cast(0)->payload) / sizeof(Ch), MaxSize = MaxChars - 1, LenPos = MaxSize }; + Ch str[MaxChars]; + + inline static bool Usable(SizeType len) { return (MaxSize >= len); } + inline void SetLength(SizeType len) { str[LenPos] = static_cast(MaxSize - len); } + inline SizeType GetLength() const { return static_cast(MaxSize - str[LenPos]); } + }; // at most as many bytes as "String" above => 12 bytes in 32-bit mode, 16 bytes in 64-bit mode + + // By using proper binary layout, retrieval of different integer types do not need conversions. + union Number { +#if RAPIDJSON_ENDIAN == RAPIDJSON_LITTLEENDIAN + struct I { + int i; + char padding[4]; + }i; + struct U { + unsigned u; + char padding2[4]; + }u; +#else + struct I { + char padding[4]; + int i; + }i; + struct U { + char padding2[4]; + unsigned u; + }u; +#endif + int64_t i64; + uint64_t u64; + double d; + }; // 8 bytes + + struct ObjectData { + SizeType size; + SizeType capacity; + Member* members; + }; // 12 bytes in 32-bit mode, 16 bytes in 64-bit mode + + struct ArrayData { + SizeType size; + SizeType capacity; + GenericValue* elements; + }; // 12 bytes in 32-bit mode, 16 bytes in 64-bit mode + + union Data { + String s; + ShortString ss; + Number n; + ObjectData o; + ArrayData a; + Flag f; + }; // 16 bytes in 32-bit mode, 24 bytes in 64-bit mode, 16 bytes in 64-bit with RAPIDJSON_48BITPOINTER_OPTIMIZATION + + static RAPIDJSON_FORCEINLINE const Ch* DataString(const Data& data) { + return (data.f.flags & kInlineStrFlag) ? data.ss.str : RAPIDJSON_GETPOINTER(Ch, data.s.str); + } + static RAPIDJSON_FORCEINLINE SizeType DataStringLength(const Data& data) { + return (data.f.flags & kInlineStrFlag) ? data.ss.GetLength() : data.s.length; + } + + RAPIDJSON_FORCEINLINE const Ch* GetStringPointer() const { return RAPIDJSON_GETPOINTER(Ch, data_.s.str); } + RAPIDJSON_FORCEINLINE const Ch* SetStringPointer(const Ch* str) { return RAPIDJSON_SETPOINTER(Ch, data_.s.str, str); } + RAPIDJSON_FORCEINLINE GenericValue* GetElementsPointer() const { return RAPIDJSON_GETPOINTER(GenericValue, data_.a.elements); } + RAPIDJSON_FORCEINLINE GenericValue* SetElementsPointer(GenericValue* elements) { return RAPIDJSON_SETPOINTER(GenericValue, data_.a.elements, elements); } + RAPIDJSON_FORCEINLINE Member* GetMembersPointer() const { return RAPIDJSON_GETPOINTER(Member, data_.o.members); } + RAPIDJSON_FORCEINLINE Member* SetMembersPointer(Member* members) { return RAPIDJSON_SETPOINTER(Member, data_.o.members, members); } + +#if RAPIDJSON_USE_MEMBERSMAP + + struct MapTraits { + struct Less { + bool operator()(const Data& s1, const Data& s2) const { + SizeType n1 = DataStringLength(s1), n2 = DataStringLength(s2); + int cmp = std::memcmp(DataString(s1), DataString(s2), sizeof(Ch) * (n1 < n2 ? n1 : n2)); + return cmp < 0 || (cmp == 0 && n1 < n2); + } + }; + typedef std::pair Pair; + typedef std::multimap > Map; + typedef typename Map::iterator Iterator; + }; + typedef typename MapTraits::Map Map; + typedef typename MapTraits::Less MapLess; + typedef typename MapTraits::Pair MapPair; + typedef typename MapTraits::Iterator MapIterator; + + // + // Layout of the members' map/array, re(al)located according to the needed capacity: + // + // {Map*}<>{capacity}<>{Member[capacity]}<>{MapIterator[capacity]} + // + // (where <> stands for the RAPIDJSON_ALIGN-ment, if needed) + // + + static RAPIDJSON_FORCEINLINE size_t GetMapLayoutSize(SizeType capacity) { + return RAPIDJSON_ALIGN(sizeof(Map*)) + + RAPIDJSON_ALIGN(sizeof(SizeType)) + + RAPIDJSON_ALIGN(capacity * sizeof(Member)) + + capacity * sizeof(MapIterator); + } + + static RAPIDJSON_FORCEINLINE SizeType &GetMapCapacity(Map* &map) { + return *reinterpret_cast(reinterpret_cast(&map) + + RAPIDJSON_ALIGN(sizeof(Map*))); + } + + static RAPIDJSON_FORCEINLINE Member* GetMapMembers(Map* &map) { + return reinterpret_cast(reinterpret_cast(&map) + + RAPIDJSON_ALIGN(sizeof(Map*)) + + RAPIDJSON_ALIGN(sizeof(SizeType))); + } + + static RAPIDJSON_FORCEINLINE MapIterator* GetMapIterators(Map* &map) { + return reinterpret_cast(reinterpret_cast(&map) + + RAPIDJSON_ALIGN(sizeof(Map*)) + + RAPIDJSON_ALIGN(sizeof(SizeType)) + + RAPIDJSON_ALIGN(GetMapCapacity(map) * sizeof(Member))); + } + + static RAPIDJSON_FORCEINLINE Map* &GetMap(Member* members) { + RAPIDJSON_ASSERT(members != 0); + return *reinterpret_cast(reinterpret_cast(members) - + RAPIDJSON_ALIGN(sizeof(SizeType)) - + RAPIDJSON_ALIGN(sizeof(Map*))); + } + + // Some compilers' debug mechanisms want all iterators to be destroyed, for their accounting.. + RAPIDJSON_FORCEINLINE MapIterator DropMapIterator(MapIterator& rhs) { +#if RAPIDJSON_HAS_CXX11 + MapIterator ret = std::move(rhs); +#else + MapIterator ret = rhs; +#endif + rhs.~MapIterator(); + return ret; + } + + Map* &DoReallocMap(Map** oldMap, SizeType newCapacity, Allocator& allocator) { + Map **newMap = static_cast(allocator.Malloc(GetMapLayoutSize(newCapacity))); + GetMapCapacity(*newMap) = newCapacity; + if (!oldMap) { + *newMap = new (allocator.Malloc(sizeof(Map))) Map(MapLess(), allocator); + } + else { + *newMap = *oldMap; + size_t count = (*oldMap)->size(); + std::memcpy(static_cast(GetMapMembers(*newMap)), + static_cast(GetMapMembers(*oldMap)), + count * sizeof(Member)); + MapIterator *oldIt = GetMapIterators(*oldMap), + *newIt = GetMapIterators(*newMap); + while (count--) { + new (&newIt[count]) MapIterator(DropMapIterator(oldIt[count])); + } + Allocator::Free(oldMap); + } + return *newMap; + } + + RAPIDJSON_FORCEINLINE Member* DoAllocMembers(SizeType capacity, Allocator& allocator) { + return GetMapMembers(DoReallocMap(0, capacity, allocator)); + } + + void DoReserveMembers(SizeType newCapacity, Allocator& allocator) { + ObjectData& o = data_.o; + if (newCapacity > o.capacity) { + Member* oldMembers = GetMembersPointer(); + Map **oldMap = oldMembers ? &GetMap(oldMembers) : 0, + *&newMap = DoReallocMap(oldMap, newCapacity, allocator); + RAPIDJSON_SETPOINTER(Member, o.members, GetMapMembers(newMap)); + o.capacity = newCapacity; + } + } + + template + MemberIterator DoFindMember(const GenericValue& name) { + if (Member* members = GetMembersPointer()) { + Map* &map = GetMap(members); + MapIterator mit = map->find(reinterpret_cast(name.data_)); + if (mit != map->end()) { + return MemberIterator(&members[mit->second]); + } + } + return MemberEnd(); + } + + void DoClearMembers() { + if (Member* members = GetMembersPointer()) { + Map* &map = GetMap(members); + MapIterator* mit = GetMapIterators(map); + for (SizeType i = 0; i < data_.o.size; i++) { + map->erase(DropMapIterator(mit[i])); + members[i].~Member(); + } + data_.o.size = 0; + } + } + + void DoFreeMembers() { + if (Member* members = GetMembersPointer()) { + GetMap(members)->~Map(); + for (SizeType i = 0; i < data_.o.size; i++) { + members[i].~Member(); + } + if (Allocator::kNeedFree) { // Shortcut by Allocator's trait + Map** map = &GetMap(members); + Allocator::Free(*map); + Allocator::Free(map); + } + } + } + +#else // !RAPIDJSON_USE_MEMBERSMAP + + RAPIDJSON_FORCEINLINE Member* DoAllocMembers(SizeType capacity, Allocator& allocator) { + return Malloc(allocator, capacity); + } + + void DoReserveMembers(SizeType newCapacity, Allocator& allocator) { + ObjectData& o = data_.o; + if (newCapacity > o.capacity) { + Member* newMembers = Realloc(allocator, GetMembersPointer(), o.capacity, newCapacity); + RAPIDJSON_SETPOINTER(Member, o.members, newMembers); + o.capacity = newCapacity; + } + } + + template + MemberIterator DoFindMember(const GenericValue& name) { + MemberIterator member = MemberBegin(); + for ( ; member != MemberEnd(); ++member) + if (name.StringEqual(member->name)) + break; + return member; + } + + void DoClearMembers() { + for (MemberIterator m = MemberBegin(); m != MemberEnd(); ++m) + m->~Member(); + data_.o.size = 0; + } + + void DoFreeMembers() { + for (MemberIterator m = MemberBegin(); m != MemberEnd(); ++m) + m->~Member(); + Allocator::Free(GetMembersPointer()); + } + +#endif // !RAPIDJSON_USE_MEMBERSMAP + + void DoAddMember(GenericValue& name, GenericValue& value, Allocator& allocator) { + ObjectData& o = data_.o; + if (o.size >= o.capacity) + DoReserveMembers(o.capacity ? (o.capacity + (o.capacity + 1) / 2) : kDefaultObjectCapacity, allocator); + Member* members = GetMembersPointer(); + Member* m = members + o.size; + m->name.RawAssign(name); + m->value.RawAssign(value); +#if RAPIDJSON_USE_MEMBERSMAP + Map* &map = GetMap(members); + MapIterator* mit = GetMapIterators(map); + new (&mit[o.size]) MapIterator(map->insert(MapPair(m->name.data_, o.size))); +#endif + ++o.size; + } + + MemberIterator DoRemoveMember(MemberIterator m) { + ObjectData& o = data_.o; + Member* members = GetMembersPointer(); +#if RAPIDJSON_USE_MEMBERSMAP + Map* &map = GetMap(members); + MapIterator* mit = GetMapIterators(map); + SizeType mpos = static_cast(&*m - members); + map->erase(DropMapIterator(mit[mpos])); +#endif + MemberIterator last(members + (o.size - 1)); + if (o.size > 1 && m != last) { +#if RAPIDJSON_USE_MEMBERSMAP + new (&mit[mpos]) MapIterator(DropMapIterator(mit[&*last - members])); + mit[mpos]->second = mpos; +#endif + *m = *last; // Move the last one to this place + } + else { + m->~Member(); // Only one left, just destroy + } + --o.size; + return m; + } + + MemberIterator DoEraseMembers(ConstMemberIterator first, ConstMemberIterator last) { + ObjectData& o = data_.o; + MemberIterator beg = MemberBegin(), + pos = beg + (first - beg), + end = MemberEnd(); +#if RAPIDJSON_USE_MEMBERSMAP + Map* &map = GetMap(GetMembersPointer()); + MapIterator* mit = GetMapIterators(map); +#endif + for (MemberIterator itr = pos; itr != last; ++itr) { +#if RAPIDJSON_USE_MEMBERSMAP + map->erase(DropMapIterator(mit[itr - beg])); +#endif + itr->~Member(); + } +#if RAPIDJSON_USE_MEMBERSMAP + if (first != last) { + // Move remaining members/iterators + MemberIterator next = pos + (last - first); + for (MemberIterator itr = pos; next != end; ++itr, ++next) { + std::memcpy(static_cast(&*itr), &*next, sizeof(Member)); + SizeType mpos = static_cast(itr - beg); + new (&mit[mpos]) MapIterator(DropMapIterator(mit[next - beg])); + mit[mpos]->second = mpos; + } + } +#else + std::memmove(static_cast(&*pos), &*last, + static_cast(end - last) * sizeof(Member)); +#endif + o.size -= static_cast(last - first); + return pos; + } + + template + void DoCopyMembers(const GenericValue& rhs, Allocator& allocator, bool copyConstStrings) { + RAPIDJSON_ASSERT(rhs.GetType() == kObjectType); + + data_.f.flags = kObjectFlag; + SizeType count = rhs.data_.o.size; + Member* lm = DoAllocMembers(count, allocator); + const typename GenericValue::Member* rm = rhs.GetMembersPointer(); +#if RAPIDJSON_USE_MEMBERSMAP + Map* &map = GetMap(lm); + MapIterator* mit = GetMapIterators(map); +#endif + for (SizeType i = 0; i < count; i++) { + new (&lm[i].name) GenericValue(rm[i].name, allocator, copyConstStrings); + new (&lm[i].value) GenericValue(rm[i].value, allocator, copyConstStrings); +#if RAPIDJSON_USE_MEMBERSMAP + new (&mit[i]) MapIterator(map->insert(MapPair(lm[i].name.data_, i))); +#endif + } + data_.o.size = data_.o.capacity = count; + SetMembersPointer(lm); + } + + // Initialize this value as array with initial data, without calling destructor. + void SetArrayRaw(GenericValue* values, SizeType count, Allocator& allocator) { + data_.f.flags = kArrayFlag; + if (count) { + GenericValue* e = static_cast(allocator.Malloc(count * sizeof(GenericValue))); + SetElementsPointer(e); + std::memcpy(static_cast(e), values, count * sizeof(GenericValue)); + } + else + SetElementsPointer(0); + data_.a.size = data_.a.capacity = count; + } + + //! Initialize this value as object with initial data, without calling destructor. + void SetObjectRaw(Member* members, SizeType count, Allocator& allocator) { + data_.f.flags = kObjectFlag; + if (count) { + Member* m = DoAllocMembers(count, allocator); + SetMembersPointer(m); + std::memcpy(static_cast(m), members, count * sizeof(Member)); +#if RAPIDJSON_USE_MEMBERSMAP + Map* &map = GetMap(m); + MapIterator* mit = GetMapIterators(map); + for (SizeType i = 0; i < count; i++) { + new (&mit[i]) MapIterator(map->insert(MapPair(m[i].name.data_, i))); + } +#endif + } + else + SetMembersPointer(0); + data_.o.size = data_.o.capacity = count; + } + + //! Initialize this value as constant string, without calling destructor. + void SetStringRaw(StringRefType s) RAPIDJSON_NOEXCEPT { + data_.f.flags = kConstStringFlag; + SetStringPointer(s); + data_.s.length = s.length; + } + + //! Initialize this value as copy string with initial data, without calling destructor. + void SetStringRaw(StringRefType s, Allocator& allocator) { + Ch* str = 0; + if (ShortString::Usable(s.length)) { + data_.f.flags = kShortStringFlag; + data_.ss.SetLength(s.length); + str = data_.ss.str; + std::memmove(str, s, s.length * sizeof(Ch)); + } else { + data_.f.flags = kCopyStringFlag; + data_.s.length = s.length; + str = static_cast(allocator.Malloc((s.length + 1) * sizeof(Ch))); + SetStringPointer(str); + std::memcpy(str, s, s.length * sizeof(Ch)); + } + str[s.length] = '\0'; + } + + //! Assignment without calling destructor + void RawAssign(GenericValue& rhs) RAPIDJSON_NOEXCEPT { + data_ = rhs.data_; + // data_.f.flags = rhs.data_.f.flags; + rhs.data_.f.flags = kNullFlag; + } + + template + bool StringEqual(const GenericValue& rhs) const { + RAPIDJSON_ASSERT(IsString()); + RAPIDJSON_ASSERT(rhs.IsString()); + + const SizeType len1 = GetStringLength(); + const SizeType len2 = rhs.GetStringLength(); + if(len1 != len2) { return false; } + + const Ch* const str1 = GetString(); + const Ch* const str2 = rhs.GetString(); + if(str1 == str2) { return true; } // fast path for constant string + + return (std::memcmp(str1, str2, sizeof(Ch) * len1) == 0); + } + + Data data_; +}; + +//! GenericValue with UTF8 encoding +typedef GenericValue > Value; + +/////////////////////////////////////////////////////////////////////////////// +// GenericDocument + +//! A document for parsing JSON text as DOM. +/*! + \note implements Handler concept + \tparam Encoding Encoding for both parsing and string storage. + \tparam Allocator Allocator for allocating memory for the DOM + \tparam StackAllocator Allocator for allocating memory for stack during parsing. + \warning Although GenericDocument inherits from GenericValue, the API does \b not provide any virtual functions, especially no virtual destructor. To avoid memory leaks, do not \c delete a GenericDocument object via a pointer to a GenericValue. +*/ +template +class GenericDocument : public GenericValue { +public: + typedef typename Encoding::Ch Ch; //!< Character type derived from Encoding. + typedef GenericValue ValueType; //!< Value type of the document. + typedef Allocator AllocatorType; //!< Allocator type from template parameter. + typedef StackAllocator StackAllocatorType; //!< StackAllocator type from template parameter. + + //! Constructor + /*! Creates an empty document of specified type. + \param type Mandatory type of object to create. + \param allocator Optional allocator for allocating memory. + \param stackCapacity Optional initial capacity of stack in bytes. + \param stackAllocator Optional allocator for allocating memory for stack. + */ + explicit GenericDocument(Type type, Allocator* allocator = 0, size_t stackCapacity = kDefaultStackCapacity, StackAllocator* stackAllocator = 0) : + GenericValue(type), allocator_(allocator), ownAllocator_(0), stack_(stackAllocator, stackCapacity), parseResult_() + { + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + } + + //! Constructor + /*! Creates an empty document which type is Null. + \param allocator Optional allocator for allocating memory. + \param stackCapacity Optional initial capacity of stack in bytes. + \param stackAllocator Optional allocator for allocating memory for stack. + */ + GenericDocument(Allocator* allocator = 0, size_t stackCapacity = kDefaultStackCapacity, StackAllocator* stackAllocator = 0) : + allocator_(allocator), ownAllocator_(0), stack_(stackAllocator, stackCapacity), parseResult_() + { + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move constructor in C++11 + GenericDocument(GenericDocument&& rhs) RAPIDJSON_NOEXCEPT + : ValueType(std::forward(rhs)), // explicit cast to avoid prohibited move from Document + allocator_(rhs.allocator_), + ownAllocator_(rhs.ownAllocator_), + stack_(std::move(rhs.stack_)), + parseResult_(rhs.parseResult_) + { + rhs.allocator_ = 0; + rhs.ownAllocator_ = 0; + rhs.parseResult_ = ParseResult(); + } +#endif + + ~GenericDocument() { + // Clear the ::ValueType before ownAllocator is destroyed, ~ValueType() + // runs last and may access its elements or members which would be freed + // with an allocator like MemoryPoolAllocator (CrtAllocator does not + // free its data when destroyed, but MemoryPoolAllocator does). + if (ownAllocator_) { + ValueType::SetNull(); + } + Destroy(); + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move assignment in C++11 + GenericDocument& operator=(GenericDocument&& rhs) RAPIDJSON_NOEXCEPT + { + // The cast to ValueType is necessary here, because otherwise it would + // attempt to call GenericValue's templated assignment operator. + ValueType::operator=(std::forward(rhs)); + + // Calling the destructor here would prematurely call stack_'s destructor + Destroy(); + + allocator_ = rhs.allocator_; + ownAllocator_ = rhs.ownAllocator_; + stack_ = std::move(rhs.stack_); + parseResult_ = rhs.parseResult_; + + rhs.allocator_ = 0; + rhs.ownAllocator_ = 0; + rhs.parseResult_ = ParseResult(); + + return *this; + } +#endif + + //! Exchange the contents of this document with those of another. + /*! + \param rhs Another document. + \note Constant complexity. + \see GenericValue::Swap + */ + GenericDocument& Swap(GenericDocument& rhs) RAPIDJSON_NOEXCEPT { + ValueType::Swap(rhs); + stack_.Swap(rhs.stack_); + internal::Swap(allocator_, rhs.allocator_); + internal::Swap(ownAllocator_, rhs.ownAllocator_); + internal::Swap(parseResult_, rhs.parseResult_); + return *this; + } + + // Allow Swap with ValueType. + // Refer to Effective C++ 3rd Edition/Item 33: Avoid hiding inherited names. + using ValueType::Swap; + + //! free-standing swap function helper + /*! + Helper function to enable support for common swap implementation pattern based on \c std::swap: + \code + void swap(MyClass& a, MyClass& b) { + using std::swap; + swap(a.doc, b.doc); + // ... + } + \endcode + \see Swap() + */ + friend inline void swap(GenericDocument& a, GenericDocument& b) RAPIDJSON_NOEXCEPT { a.Swap(b); } + + //! Populate this document by a generator which produces SAX events. + /*! \tparam Generator A functor with bool f(Handler) prototype. + \param g Generator functor which sends SAX events to the parameter. + \return The document itself for fluent API. + */ + template + GenericDocument& Populate(Generator& g) { + ClearStackOnExit scope(*this); + if (g(*this)) { + RAPIDJSON_ASSERT(stack_.GetSize() == sizeof(ValueType)); // Got one and only one root object + ValueType::operator=(*stack_.template Pop(1));// Move value from stack to document + } + return *this; + } + + //!@name Parse from stream + //!@{ + + //! Parse JSON text from an input stream (with Encoding conversion) + /*! \tparam parseFlags Combination of \ref ParseFlag. + \tparam SourceEncoding Encoding of input stream + \tparam InputStream Type of input stream, implementing Stream concept + \param is Input stream to be parsed. + \return The document itself for fluent API. + */ + template + GenericDocument& ParseStream(InputStream& is) { + GenericReader reader( + stack_.HasAllocator() ? &stack_.GetAllocator() : 0); + ClearStackOnExit scope(*this); + parseResult_ = reader.template Parse(is, *this); + if (parseResult_) { + RAPIDJSON_ASSERT(stack_.GetSize() == sizeof(ValueType)); // Got one and only one root object + ValueType::operator=(*stack_.template Pop(1));// Move value from stack to document + } + return *this; + } + + //! Parse JSON text from an input stream + /*! \tparam parseFlags Combination of \ref ParseFlag. + \tparam InputStream Type of input stream, implementing Stream concept + \param is Input stream to be parsed. + \return The document itself for fluent API. + */ + template + GenericDocument& ParseStream(InputStream& is) { + return ParseStream(is); + } + + //! Parse JSON text from an input stream (with \ref kParseDefaultFlags) + /*! \tparam InputStream Type of input stream, implementing Stream concept + \param is Input stream to be parsed. + \return The document itself for fluent API. + */ + template + GenericDocument& ParseStream(InputStream& is) { + return ParseStream(is); + } + //!@} + + //!@name Parse in-place from mutable string + //!@{ + + //! Parse JSON text from a mutable string + /*! \tparam parseFlags Combination of \ref ParseFlag. + \param str Mutable zero-terminated string to be parsed. + \return The document itself for fluent API. + */ + template + GenericDocument& ParseInsitu(Ch* str) { + GenericInsituStringStream s(str); + return ParseStream(s); + } + + //! Parse JSON text from a mutable string (with \ref kParseDefaultFlags) + /*! \param str Mutable zero-terminated string to be parsed. + \return The document itself for fluent API. + */ + GenericDocument& ParseInsitu(Ch* str) { + return ParseInsitu(str); + } + //!@} + + //!@name Parse from read-only string + //!@{ + + //! Parse JSON text from a read-only string (with Encoding conversion) + /*! \tparam parseFlags Combination of \ref ParseFlag (must not contain \ref kParseInsituFlag). + \tparam SourceEncoding Transcoding from input Encoding + \param str Read-only zero-terminated string to be parsed. + */ + template + GenericDocument& Parse(const typename SourceEncoding::Ch* str) { + RAPIDJSON_ASSERT(!(parseFlags & kParseInsituFlag)); + GenericStringStream s(str); + return ParseStream(s); + } + + //! Parse JSON text from a read-only string + /*! \tparam parseFlags Combination of \ref ParseFlag (must not contain \ref kParseInsituFlag). + \param str Read-only zero-terminated string to be parsed. + */ + template + GenericDocument& Parse(const Ch* str) { + return Parse(str); + } + + //! Parse JSON text from a read-only string (with \ref kParseDefaultFlags) + /*! \param str Read-only zero-terminated string to be parsed. + */ + GenericDocument& Parse(const Ch* str) { + return Parse(str); + } + + template + GenericDocument& Parse(const typename SourceEncoding::Ch* str, size_t length) { + RAPIDJSON_ASSERT(!(parseFlags & kParseInsituFlag)); + MemoryStream ms(reinterpret_cast(str), length * sizeof(typename SourceEncoding::Ch)); + EncodedInputStream is(ms); + ParseStream(is); + return *this; + } + + template + GenericDocument& Parse(const Ch* str, size_t length) { + return Parse(str, length); + } + + GenericDocument& Parse(const Ch* str, size_t length) { + return Parse(str, length); + } + +#if RAPIDJSON_HAS_STDSTRING + template + GenericDocument& Parse(const std::basic_string& str) { + // c_str() is constant complexity according to standard. Should be faster than Parse(const char*, size_t) + return Parse(str.c_str()); + } + + template + GenericDocument& Parse(const std::basic_string& str) { + return Parse(str.c_str()); + } + + GenericDocument& Parse(const std::basic_string& str) { + return Parse(str); + } +#endif // RAPIDJSON_HAS_STDSTRING + + //!@} + + //!@name Handling parse errors + //!@{ + + //! Whether a parse error has occurred in the last parsing. + bool HasParseError() const { return parseResult_.IsError(); } + + //! Get the \ref ParseErrorCode of last parsing. + ParseErrorCode GetParseError() const { return parseResult_.Code(); } + + //! Get the position of last parsing error in input, 0 otherwise. + size_t GetErrorOffset() const { return parseResult_.Offset(); } + + //! Implicit conversion to get the last parse result +#ifndef __clang // -Wdocumentation + /*! \return \ref ParseResult of the last parse operation + + \code + Document doc; + ParseResult ok = doc.Parse(json); + if (!ok) + printf( "JSON parse error: %s (%u)\n", GetParseError_En(ok.Code()), ok.Offset()); + \endcode + */ +#endif + operator ParseResult() const { return parseResult_; } + //!@} + + //! Get the allocator of this document. + Allocator& GetAllocator() { + RAPIDJSON_ASSERT(allocator_); + return *allocator_; + } + + //! Get the capacity of stack in bytes. + size_t GetStackCapacity() const { return stack_.GetCapacity(); } + +private: + // clear stack on any exit from ParseStream, e.g. due to exception + struct ClearStackOnExit { + explicit ClearStackOnExit(GenericDocument& d) : d_(d) {} + ~ClearStackOnExit() { d_.ClearStack(); } + private: + ClearStackOnExit(const ClearStackOnExit&); + ClearStackOnExit& operator=(const ClearStackOnExit&); + GenericDocument& d_; + }; + + // callers of the following private Handler functions + // template friend class GenericReader; // for parsing + template friend class GenericValue; // for deep copying + +public: + // Implementation of Handler + bool Null() { new (stack_.template Push()) ValueType(); return true; } + bool Bool(bool b) { new (stack_.template Push()) ValueType(b); return true; } + bool Int(int i) { new (stack_.template Push()) ValueType(i); return true; } + bool Uint(unsigned i) { new (stack_.template Push()) ValueType(i); return true; } + bool Int64(int64_t i) { new (stack_.template Push()) ValueType(i); return true; } + bool Uint64(uint64_t i) { new (stack_.template Push()) ValueType(i); return true; } + bool Double(double d) { new (stack_.template Push()) ValueType(d); return true; } + + bool RawNumber(const Ch* str, SizeType length, bool copy) { + if (copy) + new (stack_.template Push()) ValueType(str, length, GetAllocator()); + else + new (stack_.template Push()) ValueType(str, length); + return true; + } + + bool String(const Ch* str, SizeType length, bool copy) { + if (copy) + new (stack_.template Push()) ValueType(str, length, GetAllocator()); + else + new (stack_.template Push()) ValueType(str, length); + return true; + } + + bool StartObject() { new (stack_.template Push()) ValueType(kObjectType); return true; } + + bool Key(const Ch* str, SizeType length, bool copy) { return String(str, length, copy); } + + bool EndObject(SizeType memberCount) { + typename ValueType::Member* members = stack_.template Pop(memberCount); + stack_.template Top()->SetObjectRaw(members, memberCount, GetAllocator()); + return true; + } + + bool StartArray() { new (stack_.template Push()) ValueType(kArrayType); return true; } + + bool EndArray(SizeType elementCount) { + ValueType* elements = stack_.template Pop(elementCount); + stack_.template Top()->SetArrayRaw(elements, elementCount, GetAllocator()); + return true; + } + +private: + //! Prohibit copying + GenericDocument(const GenericDocument&); + //! Prohibit assignment + GenericDocument& operator=(const GenericDocument&); + + void ClearStack() { + if (Allocator::kNeedFree) + while (stack_.GetSize() > 0) // Here assumes all elements in stack array are GenericValue (Member is actually 2 GenericValue objects) + (stack_.template Pop(1))->~ValueType(); + else + stack_.Clear(); + stack_.ShrinkToFit(); + } + + void Destroy() { + RAPIDJSON_DELETE(ownAllocator_); + } + + static const size_t kDefaultStackCapacity = 1024; + Allocator* allocator_; + Allocator* ownAllocator_; + internal::Stack stack_; + ParseResult parseResult_; +}; + +//! GenericDocument with UTF8 encoding +typedef GenericDocument > Document; + + +//! Helper class for accessing Value of array type. +/*! + Instance of this helper class is obtained by \c GenericValue::GetArray(). + In addition to all APIs for array type, it provides range-based for loop if \c RAPIDJSON_HAS_CXX11_RANGE_FOR=1. +*/ +template +class GenericArray { +public: + typedef GenericArray ConstArray; + typedef GenericArray Array; + typedef ValueT PlainType; + typedef typename internal::MaybeAddConst::Type ValueType; + typedef ValueType* ValueIterator; // This may be const or non-const iterator + typedef const ValueT* ConstValueIterator; + typedef typename ValueType::AllocatorType AllocatorType; + typedef typename ValueType::StringRefType StringRefType; + + template + friend class GenericValue; + + GenericArray(const GenericArray& rhs) : value_(rhs.value_) {} + GenericArray& operator=(const GenericArray& rhs) { value_ = rhs.value_; return *this; } + ~GenericArray() {} + + operator ValueType&() const { return value_; } + SizeType Size() const { return value_.Size(); } + SizeType Capacity() const { return value_.Capacity(); } + bool Empty() const { return value_.Empty(); } + void Clear() const { value_.Clear(); } + ValueType& operator[](SizeType index) const { return value_[index]; } + ValueIterator Begin() const { return value_.Begin(); } + ValueIterator End() const { return value_.End(); } + GenericArray Reserve(SizeType newCapacity, AllocatorType &allocator) const { value_.Reserve(newCapacity, allocator); return *this; } + GenericArray PushBack(ValueType& value, AllocatorType& allocator) const { value_.PushBack(value, allocator); return *this; } +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericArray PushBack(ValueType&& value, AllocatorType& allocator) const { value_.PushBack(value, allocator); return *this; } +#endif // RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericArray PushBack(StringRefType value, AllocatorType& allocator) const { value_.PushBack(value, allocator); return *this; } + template RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (const GenericArray&)) PushBack(T value, AllocatorType& allocator) const { value_.PushBack(value, allocator); return *this; } + GenericArray PopBack() const { value_.PopBack(); return *this; } + ValueIterator Erase(ConstValueIterator pos) const { return value_.Erase(pos); } + ValueIterator Erase(ConstValueIterator first, ConstValueIterator last) const { return value_.Erase(first, last); } + +#if RAPIDJSON_HAS_CXX11_RANGE_FOR + ValueIterator begin() const { return value_.Begin(); } + ValueIterator end() const { return value_.End(); } +#endif + +private: + GenericArray(); + GenericArray(ValueType& value) : value_(value) {} + ValueType& value_; +}; + +//! Helper class for accessing Value of object type. +/*! + Instance of this helper class is obtained by \c GenericValue::GetObject(). + In addition to all APIs for array type, it provides range-based for loop if \c RAPIDJSON_HAS_CXX11_RANGE_FOR=1. +*/ +template +class GenericObject { +public: + typedef GenericObject ConstObject; + typedef GenericObject Object; + typedef ValueT PlainType; + typedef typename internal::MaybeAddConst::Type ValueType; + typedef GenericMemberIterator MemberIterator; // This may be const or non-const iterator + typedef GenericMemberIterator ConstMemberIterator; + typedef typename ValueType::AllocatorType AllocatorType; + typedef typename ValueType::StringRefType StringRefType; + typedef typename ValueType::EncodingType EncodingType; + typedef typename ValueType::Ch Ch; + + template + friend class GenericValue; + + GenericObject(const GenericObject& rhs) : value_(rhs.value_) {} + GenericObject& operator=(const GenericObject& rhs) { value_ = rhs.value_; return *this; } + ~GenericObject() {} + + operator ValueType&() const { return value_; } + SizeType MemberCount() const { return value_.MemberCount(); } + SizeType MemberCapacity() const { return value_.MemberCapacity(); } + bool ObjectEmpty() const { return value_.ObjectEmpty(); } + template ValueType& operator[](T* name) const { return value_[name]; } + template ValueType& operator[](const GenericValue& name) const { return value_[name]; } +#if RAPIDJSON_HAS_STDSTRING + ValueType& operator[](const std::basic_string& name) const { return value_[name]; } +#endif + MemberIterator MemberBegin() const { return value_.MemberBegin(); } + MemberIterator MemberEnd() const { return value_.MemberEnd(); } + GenericObject MemberReserve(SizeType newCapacity, AllocatorType &allocator) const { value_.MemberReserve(newCapacity, allocator); return *this; } + bool HasMember(const Ch* name) const { return value_.HasMember(name); } +#if RAPIDJSON_HAS_STDSTRING + bool HasMember(const std::basic_string& name) const { return value_.HasMember(name); } +#endif + template bool HasMember(const GenericValue& name) const { return value_.HasMember(name); } + MemberIterator FindMember(const Ch* name) const { return value_.FindMember(name); } + template MemberIterator FindMember(const GenericValue& name) const { return value_.FindMember(name); } +#if RAPIDJSON_HAS_STDSTRING + MemberIterator FindMember(const std::basic_string& name) const { return value_.FindMember(name); } +#endif + GenericObject AddMember(ValueType& name, ValueType& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + GenericObject AddMember(ValueType& name, StringRefType value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } +#if RAPIDJSON_HAS_STDSTRING + GenericObject AddMember(ValueType& name, std::basic_string& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } +#endif + template RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (ValueType&)) AddMember(ValueType& name, T value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericObject AddMember(ValueType&& name, ValueType&& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + GenericObject AddMember(ValueType&& name, ValueType& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + GenericObject AddMember(ValueType& name, ValueType&& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + GenericObject AddMember(StringRefType name, ValueType&& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } +#endif // RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericObject AddMember(StringRefType name, ValueType& value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + GenericObject AddMember(StringRefType name, StringRefType value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + template RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (GenericObject)) AddMember(StringRefType name, T value, AllocatorType& allocator) const { value_.AddMember(name, value, allocator); return *this; } + void RemoveAllMembers() { value_.RemoveAllMembers(); } + bool RemoveMember(const Ch* name) const { return value_.RemoveMember(name); } +#if RAPIDJSON_HAS_STDSTRING + bool RemoveMember(const std::basic_string& name) const { return value_.RemoveMember(name); } +#endif + template bool RemoveMember(const GenericValue& name) const { return value_.RemoveMember(name); } + MemberIterator RemoveMember(MemberIterator m) const { return value_.RemoveMember(m); } + MemberIterator EraseMember(ConstMemberIterator pos) const { return value_.EraseMember(pos); } + MemberIterator EraseMember(ConstMemberIterator first, ConstMemberIterator last) const { return value_.EraseMember(first, last); } + bool EraseMember(const Ch* name) const { return value_.EraseMember(name); } +#if RAPIDJSON_HAS_STDSTRING + bool EraseMember(const std::basic_string& name) const { return EraseMember(ValueType(StringRef(name))); } +#endif + template bool EraseMember(const GenericValue& name) const { return value_.EraseMember(name); } + +#if RAPIDJSON_HAS_CXX11_RANGE_FOR + MemberIterator begin() const { return value_.MemberBegin(); } + MemberIterator end() const { return value_.MemberEnd(); } +#endif + +private: + GenericObject(); + GenericObject(ValueType& value) : value_(value) {} + ValueType& value_; +}; + +RAPIDJSON_NAMESPACE_END +RAPIDJSON_DIAG_POP + +#ifdef RAPIDJSON_WINDOWS_GETOBJECT_WORKAROUND_APPLIED +#pragma pop_macro("GetObject") +#undef RAPIDJSON_WINDOWS_GETOBJECT_WORKAROUND_APPLIED +#endif + +#endif // RAPIDJSON_DOCUMENT_H_ diff --git a/include/rapidjson/encodedstream.h b/include/rapidjson/encodedstream.h new file mode 100644 index 0000000..cf046b8 --- /dev/null +++ b/include/rapidjson/encodedstream.h @@ -0,0 +1,299 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ENCODEDSTREAM_H_ +#define RAPIDJSON_ENCODEDSTREAM_H_ + +#include "stream.h" +#include "memorystream.h" + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Input byte stream wrapper with a statically bound encoding. +/*! + \tparam Encoding The interpretation of encoding of the stream. Either UTF8, UTF16LE, UTF16BE, UTF32LE, UTF32BE. + \tparam InputByteStream Type of input byte stream. For example, FileReadStream. +*/ +template +class EncodedInputStream { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); +public: + typedef typename Encoding::Ch Ch; + + EncodedInputStream(InputByteStream& is) : is_(is) { + current_ = Encoding::TakeBOM(is_); + } + + Ch Peek() const { return current_; } + Ch Take() { Ch c = current_; current_ = Encoding::Take(is_); return c; } + size_t Tell() const { return is_.Tell(); } + + // Not implemented + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + EncodedInputStream(const EncodedInputStream&); + EncodedInputStream& operator=(const EncodedInputStream&); + + InputByteStream& is_; + Ch current_; +}; + +//! Specialized for UTF8 MemoryStream. +template <> +class EncodedInputStream, MemoryStream> { +public: + typedef UTF8<>::Ch Ch; + + EncodedInputStream(MemoryStream& is) : is_(is) { + if (static_cast(is_.Peek()) == 0xEFu) is_.Take(); + if (static_cast(is_.Peek()) == 0xBBu) is_.Take(); + if (static_cast(is_.Peek()) == 0xBFu) is_.Take(); + } + Ch Peek() const { return is_.Peek(); } + Ch Take() { return is_.Take(); } + size_t Tell() const { return is_.Tell(); } + + // Not implemented + void Put(Ch) {} + void Flush() {} + Ch* PutBegin() { return 0; } + size_t PutEnd(Ch*) { return 0; } + + MemoryStream& is_; + +private: + EncodedInputStream(const EncodedInputStream&); + EncodedInputStream& operator=(const EncodedInputStream&); +}; + +//! Output byte stream wrapper with statically bound encoding. +/*! + \tparam Encoding The interpretation of encoding of the stream. Either UTF8, UTF16LE, UTF16BE, UTF32LE, UTF32BE. + \tparam OutputByteStream Type of input byte stream. For example, FileWriteStream. +*/ +template +class EncodedOutputStream { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); +public: + typedef typename Encoding::Ch Ch; + + EncodedOutputStream(OutputByteStream& os, bool putBOM = true) : os_(os) { + if (putBOM) + Encoding::PutBOM(os_); + } + + void Put(Ch c) { Encoding::Put(os_, c); } + void Flush() { os_.Flush(); } + + // Not implemented + Ch Peek() const { RAPIDJSON_ASSERT(false); return 0;} + Ch Take() { RAPIDJSON_ASSERT(false); return 0;} + size_t Tell() const { RAPIDJSON_ASSERT(false); return 0; } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + EncodedOutputStream(const EncodedOutputStream&); + EncodedOutputStream& operator=(const EncodedOutputStream&); + + OutputByteStream& os_; +}; + +#define RAPIDJSON_ENCODINGS_FUNC(x) UTF8::x, UTF16LE::x, UTF16BE::x, UTF32LE::x, UTF32BE::x + +//! Input stream wrapper with dynamically bound encoding and automatic encoding detection. +/*! + \tparam CharType Type of character for reading. + \tparam InputByteStream type of input byte stream to be wrapped. +*/ +template +class AutoUTFInputStream { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); +public: + typedef CharType Ch; + + //! Constructor. + /*! + \param is input stream to be wrapped. + \param type UTF encoding type if it is not detected from the stream. + */ + AutoUTFInputStream(InputByteStream& is, UTFType type = kUTF8) : is_(&is), type_(type), hasBOM_(false) { + RAPIDJSON_ASSERT(type >= kUTF8 && type <= kUTF32BE); + DetectType(); + static const TakeFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Take) }; + takeFunc_ = f[type_]; + current_ = takeFunc_(*is_); + } + + UTFType GetType() const { return type_; } + bool HasBOM() const { return hasBOM_; } + + Ch Peek() const { return current_; } + Ch Take() { Ch c = current_; current_ = takeFunc_(*is_); return c; } + size_t Tell() const { return is_->Tell(); } + + // Not implemented + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + AutoUTFInputStream(const AutoUTFInputStream&); + AutoUTFInputStream& operator=(const AutoUTFInputStream&); + + // Detect encoding type with BOM or RFC 4627 + void DetectType() { + // BOM (Byte Order Mark): + // 00 00 FE FF UTF-32BE + // FF FE 00 00 UTF-32LE + // FE FF UTF-16BE + // FF FE UTF-16LE + // EF BB BF UTF-8 + + const unsigned char* c = reinterpret_cast(is_->Peek4()); + if (!c) + return; + + unsigned bom = static_cast(c[0] | (c[1] << 8) | (c[2] << 16) | (c[3] << 24)); + hasBOM_ = false; + if (bom == 0xFFFE0000) { type_ = kUTF32BE; hasBOM_ = true; is_->Take(); is_->Take(); is_->Take(); is_->Take(); } + else if (bom == 0x0000FEFF) { type_ = kUTF32LE; hasBOM_ = true; is_->Take(); is_->Take(); is_->Take(); is_->Take(); } + else if ((bom & 0xFFFF) == 0xFFFE) { type_ = kUTF16BE; hasBOM_ = true; is_->Take(); is_->Take(); } + else if ((bom & 0xFFFF) == 0xFEFF) { type_ = kUTF16LE; hasBOM_ = true; is_->Take(); is_->Take(); } + else if ((bom & 0xFFFFFF) == 0xBFBBEF) { type_ = kUTF8; hasBOM_ = true; is_->Take(); is_->Take(); is_->Take(); } + + // RFC 4627: Section 3 + // "Since the first two characters of a JSON text will always be ASCII + // characters [RFC0020], it is possible to determine whether an octet + // stream is UTF-8, UTF-16 (BE or LE), or UTF-32 (BE or LE) by looking + // at the pattern of nulls in the first four octets." + // 00 00 00 xx UTF-32BE + // 00 xx 00 xx UTF-16BE + // xx 00 00 00 UTF-32LE + // xx 00 xx 00 UTF-16LE + // xx xx xx xx UTF-8 + + if (!hasBOM_) { + int pattern = (c[0] ? 1 : 0) | (c[1] ? 2 : 0) | (c[2] ? 4 : 0) | (c[3] ? 8 : 0); + switch (pattern) { + case 0x08: type_ = kUTF32BE; break; + case 0x0A: type_ = kUTF16BE; break; + case 0x01: type_ = kUTF32LE; break; + case 0x05: type_ = kUTF16LE; break; + case 0x0F: type_ = kUTF8; break; + default: break; // Use type defined by user. + } + } + + // Runtime check whether the size of character type is sufficient. It only perform checks with assertion. + if (type_ == kUTF16LE || type_ == kUTF16BE) RAPIDJSON_ASSERT(sizeof(Ch) >= 2); + if (type_ == kUTF32LE || type_ == kUTF32BE) RAPIDJSON_ASSERT(sizeof(Ch) >= 4); + } + + typedef Ch (*TakeFunc)(InputByteStream& is); + InputByteStream* is_; + UTFType type_; + Ch current_; + TakeFunc takeFunc_; + bool hasBOM_; +}; + +//! Output stream wrapper with dynamically bound encoding and automatic encoding detection. +/*! + \tparam CharType Type of character for writing. + \tparam OutputByteStream type of output byte stream to be wrapped. +*/ +template +class AutoUTFOutputStream { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); +public: + typedef CharType Ch; + + //! Constructor. + /*! + \param os output stream to be wrapped. + \param type UTF encoding type. + \param putBOM Whether to write BOM at the beginning of the stream. + */ + AutoUTFOutputStream(OutputByteStream& os, UTFType type, bool putBOM) : os_(&os), type_(type) { + RAPIDJSON_ASSERT(type >= kUTF8 && type <= kUTF32BE); + + // Runtime check whether the size of character type is sufficient. It only perform checks with assertion. + if (type_ == kUTF16LE || type_ == kUTF16BE) RAPIDJSON_ASSERT(sizeof(Ch) >= 2); + if (type_ == kUTF32LE || type_ == kUTF32BE) RAPIDJSON_ASSERT(sizeof(Ch) >= 4); + + static const PutFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Put) }; + putFunc_ = f[type_]; + + if (putBOM) + PutBOM(); + } + + UTFType GetType() const { return type_; } + + void Put(Ch c) { putFunc_(*os_, c); } + void Flush() { os_->Flush(); } + + // Not implemented + Ch Peek() const { RAPIDJSON_ASSERT(false); return 0;} + Ch Take() { RAPIDJSON_ASSERT(false); return 0;} + size_t Tell() const { RAPIDJSON_ASSERT(false); return 0; } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + AutoUTFOutputStream(const AutoUTFOutputStream&); + AutoUTFOutputStream& operator=(const AutoUTFOutputStream&); + + void PutBOM() { + typedef void (*PutBOMFunc)(OutputByteStream&); + static const PutBOMFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(PutBOM) }; + f[type_](*os_); + } + + typedef void (*PutFunc)(OutputByteStream&, Ch); + + OutputByteStream* os_; + UTFType type_; + PutFunc putFunc_; +}; + +#undef RAPIDJSON_ENCODINGS_FUNC + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_FILESTREAM_H_ diff --git a/include/rapidjson/encodings.h b/include/rapidjson/encodings.h new file mode 100644 index 0000000..c453c0d --- /dev/null +++ b/include/rapidjson/encodings.h @@ -0,0 +1,716 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ENCODINGS_H_ +#define RAPIDJSON_ENCODINGS_H_ + +#include "rapidjson.h" + +#if defined(_MSC_VER) && !defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4244) // conversion from 'type1' to 'type2', possible loss of data +RAPIDJSON_DIAG_OFF(4702) // unreachable code +#elif defined(__GNUC__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +RAPIDJSON_DIAG_OFF(overflow) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// Encoding + +/*! \class rapidjson::Encoding + \brief Concept for encoding of Unicode characters. + +\code +concept Encoding { + typename Ch; //! Type of character. A "character" is actually a code unit in unicode's definition. + + enum { supportUnicode = 1 }; // or 0 if not supporting unicode + + //! \brief Encode a Unicode codepoint to an output stream. + //! \param os Output stream. + //! \param codepoint An unicode codepoint, ranging from 0x0 to 0x10FFFF inclusively. + template + static void Encode(OutputStream& os, unsigned codepoint); + + //! \brief Decode a Unicode codepoint from an input stream. + //! \param is Input stream. + //! \param codepoint Output of the unicode codepoint. + //! \return true if a valid codepoint can be decoded from the stream. + template + static bool Decode(InputStream& is, unsigned* codepoint); + + //! \brief Validate one Unicode codepoint from an encoded stream. + //! \param is Input stream to obtain codepoint. + //! \param os Output for copying one codepoint. + //! \return true if it is valid. + //! \note This function just validating and copying the codepoint without actually decode it. + template + static bool Validate(InputStream& is, OutputStream& os); + + // The following functions are deal with byte streams. + + //! Take a character from input byte stream, skip BOM if exist. + template + static CharType TakeBOM(InputByteStream& is); + + //! Take a character from input byte stream. + template + static Ch Take(InputByteStream& is); + + //! Put BOM to output byte stream. + template + static void PutBOM(OutputByteStream& os); + + //! Put a character to output byte stream. + template + static void Put(OutputByteStream& os, Ch c); +}; +\endcode +*/ + +/////////////////////////////////////////////////////////////////////////////// +// UTF8 + +//! UTF-8 encoding. +/*! http://en.wikipedia.org/wiki/UTF-8 + http://tools.ietf.org/html/rfc3629 + \tparam CharType Code unit for storing 8-bit UTF-8 data. Default is char. + \note implements Encoding concept +*/ +template +struct UTF8 { + typedef CharType Ch; + + enum { supportUnicode = 1 }; + + template + static void Encode(OutputStream& os, unsigned codepoint) { + if (codepoint <= 0x7F) + os.Put(static_cast(codepoint & 0xFF)); + else if (codepoint <= 0x7FF) { + os.Put(static_cast(0xC0 | ((codepoint >> 6) & 0xFF))); + os.Put(static_cast(0x80 | ((codepoint & 0x3F)))); + } + else if (codepoint <= 0xFFFF) { + os.Put(static_cast(0xE0 | ((codepoint >> 12) & 0xFF))); + os.Put(static_cast(0x80 | ((codepoint >> 6) & 0x3F))); + os.Put(static_cast(0x80 | (codepoint & 0x3F))); + } + else { + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + os.Put(static_cast(0xF0 | ((codepoint >> 18) & 0xFF))); + os.Put(static_cast(0x80 | ((codepoint >> 12) & 0x3F))); + os.Put(static_cast(0x80 | ((codepoint >> 6) & 0x3F))); + os.Put(static_cast(0x80 | (codepoint & 0x3F))); + } + } + + template + static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { + if (codepoint <= 0x7F) + PutUnsafe(os, static_cast(codepoint & 0xFF)); + else if (codepoint <= 0x7FF) { + PutUnsafe(os, static_cast(0xC0 | ((codepoint >> 6) & 0xFF))); + PutUnsafe(os, static_cast(0x80 | ((codepoint & 0x3F)))); + } + else if (codepoint <= 0xFFFF) { + PutUnsafe(os, static_cast(0xE0 | ((codepoint >> 12) & 0xFF))); + PutUnsafe(os, static_cast(0x80 | ((codepoint >> 6) & 0x3F))); + PutUnsafe(os, static_cast(0x80 | (codepoint & 0x3F))); + } + else { + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + PutUnsafe(os, static_cast(0xF0 | ((codepoint >> 18) & 0xFF))); + PutUnsafe(os, static_cast(0x80 | ((codepoint >> 12) & 0x3F))); + PutUnsafe(os, static_cast(0x80 | ((codepoint >> 6) & 0x3F))); + PutUnsafe(os, static_cast(0x80 | (codepoint & 0x3F))); + } + } + + template + static bool Decode(InputStream& is, unsigned* codepoint) { +#define RAPIDJSON_COPY() c = is.Take(); *codepoint = (*codepoint << 6) | (static_cast(c) & 0x3Fu) +#define RAPIDJSON_TRANS(mask) result &= ((GetRange(static_cast(c)) & mask) != 0) +#define RAPIDJSON_TAIL() RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x70) + typename InputStream::Ch c = is.Take(); + if (!(c & 0x80)) { + *codepoint = static_cast(c); + return true; + } + + unsigned char type = GetRange(static_cast(c)); + if (type >= 32) { + *codepoint = 0; + } else { + *codepoint = (0xFFu >> type) & static_cast(c); + } + bool result = true; + switch (type) { + case 2: RAPIDJSON_TAIL(); return result; + case 3: RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 4: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x50); RAPIDJSON_TAIL(); return result; + case 5: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x10); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 6: RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 10: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x20); RAPIDJSON_TAIL(); return result; + case 11: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x60); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + default: return false; + } +#undef RAPIDJSON_COPY +#undef RAPIDJSON_TRANS +#undef RAPIDJSON_TAIL + } + + template + static bool Validate(InputStream& is, OutputStream& os) { +#define RAPIDJSON_COPY() if (c != '\0') os.Put(c = is.Take()) +#define RAPIDJSON_TRANS(mask) result &= ((GetRange(static_cast(c)) & mask) != 0) +#define RAPIDJSON_TAIL() RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x70) + Ch c = static_cast(-1); + RAPIDJSON_COPY(); + if (!(c & 0x80)) + return true; + + bool result = true; + switch (GetRange(static_cast(c))) { + case 2: RAPIDJSON_TAIL(); return result; + case 3: RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 4: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x50); RAPIDJSON_TAIL(); return result; + case 5: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x10); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 6: RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + case 10: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x20); RAPIDJSON_TAIL(); return result; + case 11: RAPIDJSON_COPY(); RAPIDJSON_TRANS(0x60); RAPIDJSON_TAIL(); RAPIDJSON_TAIL(); return result; + default: return false; + } +#undef RAPIDJSON_COPY +#undef RAPIDJSON_TRANS +#undef RAPIDJSON_TAIL + } + + static unsigned char GetRange(unsigned char c) { + // Referring to DFA of http://bjoern.hoehrmann.de/utf-8/decoder/dfa/ + // With new mapping 1 -> 0x10, 7 -> 0x20, 9 -> 0x40, such that AND operation can test multiple types. + static const unsigned char type[] = { + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, + 0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10,0x10, + 0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40,0x40, + 0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20, + 0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20,0x20, + 8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2, 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, + 10,3,3,3,3,3,3,3,3,3,3,3,3,4,3,3, 11,6,6,6,5,8,8,8,8,8,8,8,8,8,8,8, + }; + return type[c]; + } + + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + typename InputByteStream::Ch c = Take(is); + if (static_cast(c) != 0xEFu) return c; + c = is.Take(); + if (static_cast(c) != 0xBBu) return c; + c = is.Take(); + if (static_cast(c) != 0xBFu) return c; + c = is.Take(); + return c; + } + + template + static Ch Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + return static_cast(is.Take()); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(0xEFu)); + os.Put(static_cast(0xBBu)); + os.Put(static_cast(0xBFu)); + } + + template + static void Put(OutputByteStream& os, Ch c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(c)); + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// UTF16 + +//! UTF-16 encoding. +/*! http://en.wikipedia.org/wiki/UTF-16 + http://tools.ietf.org/html/rfc2781 + \tparam CharType Type for storing 16-bit UTF-16 data. Default is wchar_t. C++11 may use char16_t instead. + \note implements Encoding concept + + \note For in-memory access, no need to concern endianness. The code units and code points are represented by CPU's endianness. + For streaming, use UTF16LE and UTF16BE, which handle endianness. +*/ +template +struct UTF16 { + typedef CharType Ch; + RAPIDJSON_STATIC_ASSERT(sizeof(Ch) >= 2); + + enum { supportUnicode = 1 }; + + template + static void Encode(OutputStream& os, unsigned codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 2); + if (codepoint <= 0xFFFF) { + RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair + os.Put(static_cast(codepoint)); + } + else { + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + unsigned v = codepoint - 0x10000; + os.Put(static_cast((v >> 10) | 0xD800)); + os.Put(static_cast((v & 0x3FF) | 0xDC00)); + } + } + + + template + static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 2); + if (codepoint <= 0xFFFF) { + RAPIDJSON_ASSERT(codepoint < 0xD800 || codepoint > 0xDFFF); // Code point itself cannot be surrogate pair + PutUnsafe(os, static_cast(codepoint)); + } + else { + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + unsigned v = codepoint - 0x10000; + PutUnsafe(os, static_cast((v >> 10) | 0xD800)); + PutUnsafe(os, static_cast((v & 0x3FF) | 0xDC00)); + } + } + + template + static bool Decode(InputStream& is, unsigned* codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputStream::Ch) >= 2); + typename InputStream::Ch c = is.Take(); + if (c < 0xD800 || c > 0xDFFF) { + *codepoint = static_cast(c); + return true; + } + else if (c <= 0xDBFF) { + *codepoint = (static_cast(c) & 0x3FF) << 10; + c = is.Take(); + *codepoint |= (static_cast(c) & 0x3FF); + *codepoint += 0x10000; + return c >= 0xDC00 && c <= 0xDFFF; + } + return false; + } + + template + static bool Validate(InputStream& is, OutputStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputStream::Ch) >= 2); + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 2); + typename InputStream::Ch c; + os.Put(static_cast(c = is.Take())); + if (c < 0xD800 || c > 0xDFFF) + return true; + else if (c <= 0xDBFF) { + os.Put(c = is.Take()); + return c >= 0xDC00 && c <= 0xDFFF; + } + return false; + } +}; + +//! UTF-16 little endian encoding. +template +struct UTF16LE : UTF16 { + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + CharType c = Take(is); + return static_cast(c) == 0xFEFFu ? Take(is) : c; + } + + template + static CharType Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + unsigned c = static_cast(is.Take()); + c |= static_cast(static_cast(is.Take())) << 8; + return static_cast(c); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(0xFFu)); + os.Put(static_cast(0xFEu)); + } + + template + static void Put(OutputByteStream& os, CharType c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(static_cast(c) & 0xFFu)); + os.Put(static_cast((static_cast(c) >> 8) & 0xFFu)); + } +}; + +//! UTF-16 big endian encoding. +template +struct UTF16BE : UTF16 { + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + CharType c = Take(is); + return static_cast(c) == 0xFEFFu ? Take(is) : c; + } + + template + static CharType Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + unsigned c = static_cast(static_cast(is.Take())) << 8; + c |= static_cast(static_cast(is.Take())); + return static_cast(c); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(0xFEu)); + os.Put(static_cast(0xFFu)); + } + + template + static void Put(OutputByteStream& os, CharType c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast((static_cast(c) >> 8) & 0xFFu)); + os.Put(static_cast(static_cast(c) & 0xFFu)); + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// UTF32 + +//! UTF-32 encoding. +/*! http://en.wikipedia.org/wiki/UTF-32 + \tparam CharType Type for storing 32-bit UTF-32 data. Default is unsigned. C++11 may use char32_t instead. + \note implements Encoding concept + + \note For in-memory access, no need to concern endianness. The code units and code points are represented by CPU's endianness. + For streaming, use UTF32LE and UTF32BE, which handle endianness. +*/ +template +struct UTF32 { + typedef CharType Ch; + RAPIDJSON_STATIC_ASSERT(sizeof(Ch) >= 4); + + enum { supportUnicode = 1 }; + + template + static void Encode(OutputStream& os, unsigned codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 4); + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + os.Put(codepoint); + } + + template + static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputStream::Ch) >= 4); + RAPIDJSON_ASSERT(codepoint <= 0x10FFFF); + PutUnsafe(os, codepoint); + } + + template + static bool Decode(InputStream& is, unsigned* codepoint) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputStream::Ch) >= 4); + Ch c = is.Take(); + *codepoint = c; + return c <= 0x10FFFF; + } + + template + static bool Validate(InputStream& is, OutputStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputStream::Ch) >= 4); + Ch c; + os.Put(c = is.Take()); + return c <= 0x10FFFF; + } +}; + +//! UTF-32 little endian enocoding. +template +struct UTF32LE : UTF32 { + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + CharType c = Take(is); + return static_cast(c) == 0x0000FEFFu ? Take(is) : c; + } + + template + static CharType Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + unsigned c = static_cast(is.Take()); + c |= static_cast(static_cast(is.Take())) << 8; + c |= static_cast(static_cast(is.Take())) << 16; + c |= static_cast(static_cast(is.Take())) << 24; + return static_cast(c); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(0xFFu)); + os.Put(static_cast(0xFEu)); + os.Put(static_cast(0x00u)); + os.Put(static_cast(0x00u)); + } + + template + static void Put(OutputByteStream& os, CharType c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(c & 0xFFu)); + os.Put(static_cast((c >> 8) & 0xFFu)); + os.Put(static_cast((c >> 16) & 0xFFu)); + os.Put(static_cast((c >> 24) & 0xFFu)); + } +}; + +//! UTF-32 big endian encoding. +template +struct UTF32BE : UTF32 { + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + CharType c = Take(is); + return static_cast(c) == 0x0000FEFFu ? Take(is) : c; + } + + template + static CharType Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + unsigned c = static_cast(static_cast(is.Take())) << 24; + c |= static_cast(static_cast(is.Take())) << 16; + c |= static_cast(static_cast(is.Take())) << 8; + c |= static_cast(static_cast(is.Take())); + return static_cast(c); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(0x00u)); + os.Put(static_cast(0x00u)); + os.Put(static_cast(0xFEu)); + os.Put(static_cast(0xFFu)); + } + + template + static void Put(OutputByteStream& os, CharType c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast((c >> 24) & 0xFFu)); + os.Put(static_cast((c >> 16) & 0xFFu)); + os.Put(static_cast((c >> 8) & 0xFFu)); + os.Put(static_cast(c & 0xFFu)); + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// ASCII + +//! ASCII encoding. +/*! http://en.wikipedia.org/wiki/ASCII + \tparam CharType Code unit for storing 7-bit ASCII data. Default is char. + \note implements Encoding concept +*/ +template +struct ASCII { + typedef CharType Ch; + + enum { supportUnicode = 0 }; + + template + static void Encode(OutputStream& os, unsigned codepoint) { + RAPIDJSON_ASSERT(codepoint <= 0x7F); + os.Put(static_cast(codepoint & 0xFF)); + } + + template + static void EncodeUnsafe(OutputStream& os, unsigned codepoint) { + RAPIDJSON_ASSERT(codepoint <= 0x7F); + PutUnsafe(os, static_cast(codepoint & 0xFF)); + } + + template + static bool Decode(InputStream& is, unsigned* codepoint) { + uint8_t c = static_cast(is.Take()); + *codepoint = c; + return c <= 0X7F; + } + + template + static bool Validate(InputStream& is, OutputStream& os) { + uint8_t c = static_cast(is.Take()); + os.Put(static_cast(c)); + return c <= 0x7F; + } + + template + static CharType TakeBOM(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + uint8_t c = static_cast(Take(is)); + return static_cast(c); + } + + template + static Ch Take(InputByteStream& is) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename InputByteStream::Ch) == 1); + return static_cast(is.Take()); + } + + template + static void PutBOM(OutputByteStream& os) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + (void)os; + } + + template + static void Put(OutputByteStream& os, Ch c) { + RAPIDJSON_STATIC_ASSERT(sizeof(typename OutputByteStream::Ch) == 1); + os.Put(static_cast(c)); + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// AutoUTF + +//! Runtime-specified UTF encoding type of a stream. +enum UTFType { + kUTF8 = 0, //!< UTF-8. + kUTF16LE = 1, //!< UTF-16 little endian. + kUTF16BE = 2, //!< UTF-16 big endian. + kUTF32LE = 3, //!< UTF-32 little endian. + kUTF32BE = 4 //!< UTF-32 big endian. +}; + +//! Dynamically select encoding according to stream's runtime-specified UTF encoding type. +/*! \note This class can be used with AutoUTFInputtStream and AutoUTFOutputStream, which provides GetType(). +*/ +template +struct AutoUTF { + typedef CharType Ch; + + enum { supportUnicode = 1 }; + +#define RAPIDJSON_ENCODINGS_FUNC(x) UTF8::x, UTF16LE::x, UTF16BE::x, UTF32LE::x, UTF32BE::x + + template + static RAPIDJSON_FORCEINLINE void Encode(OutputStream& os, unsigned codepoint) { + typedef void (*EncodeFunc)(OutputStream&, unsigned); + static const EncodeFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Encode) }; + (*f[os.GetType()])(os, codepoint); + } + + template + static RAPIDJSON_FORCEINLINE void EncodeUnsafe(OutputStream& os, unsigned codepoint) { + typedef void (*EncodeFunc)(OutputStream&, unsigned); + static const EncodeFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(EncodeUnsafe) }; + (*f[os.GetType()])(os, codepoint); + } + + template + static RAPIDJSON_FORCEINLINE bool Decode(InputStream& is, unsigned* codepoint) { + typedef bool (*DecodeFunc)(InputStream&, unsigned*); + static const DecodeFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Decode) }; + return (*f[is.GetType()])(is, codepoint); + } + + template + static RAPIDJSON_FORCEINLINE bool Validate(InputStream& is, OutputStream& os) { + typedef bool (*ValidateFunc)(InputStream&, OutputStream&); + static const ValidateFunc f[] = { RAPIDJSON_ENCODINGS_FUNC(Validate) }; + return (*f[is.GetType()])(is, os); + } + +#undef RAPIDJSON_ENCODINGS_FUNC +}; + +/////////////////////////////////////////////////////////////////////////////// +// Transcoder + +//! Encoding conversion. +template +struct Transcoder { + //! Take one Unicode codepoint from source encoding, convert it to target encoding and put it to the output stream. + template + static RAPIDJSON_FORCEINLINE bool Transcode(InputStream& is, OutputStream& os) { + unsigned codepoint; + if (!SourceEncoding::Decode(is, &codepoint)) + return false; + TargetEncoding::Encode(os, codepoint); + return true; + } + + template + static RAPIDJSON_FORCEINLINE bool TranscodeUnsafe(InputStream& is, OutputStream& os) { + unsigned codepoint; + if (!SourceEncoding::Decode(is, &codepoint)) + return false; + TargetEncoding::EncodeUnsafe(os, codepoint); + return true; + } + + //! Validate one Unicode codepoint from an encoded stream. + template + static RAPIDJSON_FORCEINLINE bool Validate(InputStream& is, OutputStream& os) { + return Transcode(is, os); // Since source/target encoding is different, must transcode. + } +}; + +// Forward declaration. +template +inline void PutUnsafe(Stream& stream, typename Stream::Ch c); + +//! Specialization of Transcoder with same source and target encoding. +template +struct Transcoder { + template + static RAPIDJSON_FORCEINLINE bool Transcode(InputStream& is, OutputStream& os) { + os.Put(is.Take()); // Just copy one code unit. This semantic is different from primary template class. + return true; + } + + template + static RAPIDJSON_FORCEINLINE bool TranscodeUnsafe(InputStream& is, OutputStream& os) { + PutUnsafe(os, is.Take()); // Just copy one code unit. This semantic is different from primary template class. + return true; + } + + template + static RAPIDJSON_FORCEINLINE bool Validate(InputStream& is, OutputStream& os) { + return Encoding::Validate(is, os); // source/target encoding are the same + } +}; + +RAPIDJSON_NAMESPACE_END + +#if defined(__GNUC__) || (defined(_MSC_VER) && !defined(__clang__)) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_ENCODINGS_H_ diff --git a/include/rapidjson/error/en.h b/include/rapidjson/error/en.h new file mode 100644 index 0000000..c87b04e --- /dev/null +++ b/include/rapidjson/error/en.h @@ -0,0 +1,176 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ERROR_EN_H_ +#define RAPIDJSON_ERROR_EN_H_ + +#include "error.h" + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(switch-enum) +RAPIDJSON_DIAG_OFF(covered-switch-default) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Maps error code of parsing into error message. +/*! + \ingroup RAPIDJSON_ERRORS + \param parseErrorCode Error code obtained in parsing. + \return the error message. + \note User can make a copy of this function for localization. + Using switch-case is safer for future modification of error codes. +*/ +inline const RAPIDJSON_ERROR_CHARTYPE* GetParseError_En(ParseErrorCode parseErrorCode) { + switch (parseErrorCode) { + case kParseErrorNone: return RAPIDJSON_ERROR_STRING("No error."); + + case kParseErrorDocumentEmpty: return RAPIDJSON_ERROR_STRING("The document is empty."); + case kParseErrorDocumentRootNotSingular: return RAPIDJSON_ERROR_STRING("The document root must not be followed by other values."); + + case kParseErrorValueInvalid: return RAPIDJSON_ERROR_STRING("Invalid value."); + + case kParseErrorObjectMissName: return RAPIDJSON_ERROR_STRING("Missing a name for object member."); + case kParseErrorObjectMissColon: return RAPIDJSON_ERROR_STRING("Missing a colon after a name of object member."); + case kParseErrorObjectMissCommaOrCurlyBracket: return RAPIDJSON_ERROR_STRING("Missing a comma or '}' after an object member."); + + case kParseErrorArrayMissCommaOrSquareBracket: return RAPIDJSON_ERROR_STRING("Missing a comma or ']' after an array element."); + + case kParseErrorStringUnicodeEscapeInvalidHex: return RAPIDJSON_ERROR_STRING("Incorrect hex digit after \\u escape in string."); + case kParseErrorStringUnicodeSurrogateInvalid: return RAPIDJSON_ERROR_STRING("The surrogate pair in string is invalid."); + case kParseErrorStringEscapeInvalid: return RAPIDJSON_ERROR_STRING("Invalid escape character in string."); + case kParseErrorStringMissQuotationMark: return RAPIDJSON_ERROR_STRING("Missing a closing quotation mark in string."); + case kParseErrorStringInvalidEncoding: return RAPIDJSON_ERROR_STRING("Invalid encoding in string."); + + case kParseErrorNumberTooBig: return RAPIDJSON_ERROR_STRING("Number too big to be stored in double."); + case kParseErrorNumberMissFraction: return RAPIDJSON_ERROR_STRING("Miss fraction part in number."); + case kParseErrorNumberMissExponent: return RAPIDJSON_ERROR_STRING("Miss exponent in number."); + + case kParseErrorTermination: return RAPIDJSON_ERROR_STRING("Terminate parsing due to Handler error."); + case kParseErrorUnspecificSyntaxError: return RAPIDJSON_ERROR_STRING("Unspecific syntax error."); + + default: return RAPIDJSON_ERROR_STRING("Unknown error."); + } +} + +//! Maps error code of validation into error message. +/*! + \ingroup RAPIDJSON_ERRORS + \param validateErrorCode Error code obtained from validator. + \return the error message. + \note User can make a copy of this function for localization. + Using switch-case is safer for future modification of error codes. +*/ +inline const RAPIDJSON_ERROR_CHARTYPE* GetValidateError_En(ValidateErrorCode validateErrorCode) { + switch (validateErrorCode) { + case kValidateErrors: return RAPIDJSON_ERROR_STRING("One or more validation errors have occurred"); + case kValidateErrorNone: return RAPIDJSON_ERROR_STRING("No error."); + + case kValidateErrorMultipleOf: return RAPIDJSON_ERROR_STRING("Number '%actual' is not a multiple of the 'multipleOf' value '%expected'."); + case kValidateErrorMaximum: return RAPIDJSON_ERROR_STRING("Number '%actual' is greater than the 'maximum' value '%expected'."); + case kValidateErrorExclusiveMaximum: return RAPIDJSON_ERROR_STRING("Number '%actual' is greater than or equal to the 'exclusiveMaximum' value '%expected'."); + case kValidateErrorMinimum: return RAPIDJSON_ERROR_STRING("Number '%actual' is less than the 'minimum' value '%expected'."); + case kValidateErrorExclusiveMinimum: return RAPIDJSON_ERROR_STRING("Number '%actual' is less than or equal to the 'exclusiveMinimum' value '%expected'."); + + case kValidateErrorMaxLength: return RAPIDJSON_ERROR_STRING("String '%actual' is longer than the 'maxLength' value '%expected'."); + case kValidateErrorMinLength: return RAPIDJSON_ERROR_STRING("String '%actual' is shorter than the 'minLength' value '%expected'."); + case kValidateErrorPattern: return RAPIDJSON_ERROR_STRING("String '%actual' does not match the 'pattern' regular expression."); + + case kValidateErrorMaxItems: return RAPIDJSON_ERROR_STRING("Array of length '%actual' is longer than the 'maxItems' value '%expected'."); + case kValidateErrorMinItems: return RAPIDJSON_ERROR_STRING("Array of length '%actual' is shorter than the 'minItems' value '%expected'."); + case kValidateErrorUniqueItems: return RAPIDJSON_ERROR_STRING("Array has duplicate items at indices '%duplicates' but 'uniqueItems' is true."); + case kValidateErrorAdditionalItems: return RAPIDJSON_ERROR_STRING("Array has an additional item at index '%disallowed' that is not allowed by the schema."); + + case kValidateErrorMaxProperties: return RAPIDJSON_ERROR_STRING("Object has '%actual' members which is more than 'maxProperties' value '%expected'."); + case kValidateErrorMinProperties: return RAPIDJSON_ERROR_STRING("Object has '%actual' members which is less than 'minProperties' value '%expected'."); + case kValidateErrorRequired: return RAPIDJSON_ERROR_STRING("Object is missing the following members required by the schema: '%missing'."); + case kValidateErrorAdditionalProperties: return RAPIDJSON_ERROR_STRING("Object has an additional member '%disallowed' that is not allowed by the schema."); + case kValidateErrorPatternProperties: return RAPIDJSON_ERROR_STRING("Object has 'patternProperties' that are not allowed by the schema."); + case kValidateErrorDependencies: return RAPIDJSON_ERROR_STRING("Object has missing property or schema dependencies, refer to following errors."); + + case kValidateErrorEnum: return RAPIDJSON_ERROR_STRING("Property has a value that is not one of its allowed enumerated values."); + case kValidateErrorType: return RAPIDJSON_ERROR_STRING("Property has a type '%actual' that is not in the following list: '%expected'."); + + case kValidateErrorOneOf: return RAPIDJSON_ERROR_STRING("Property did not match any of the sub-schemas specified by 'oneOf', refer to following errors."); + case kValidateErrorOneOfMatch: return RAPIDJSON_ERROR_STRING("Property matched more than one of the sub-schemas specified by 'oneOf', indices '%matches'."); + case kValidateErrorAllOf: return RAPIDJSON_ERROR_STRING("Property did not match all of the sub-schemas specified by 'allOf', refer to following errors."); + case kValidateErrorAnyOf: return RAPIDJSON_ERROR_STRING("Property did not match any of the sub-schemas specified by 'anyOf', refer to following errors."); + case kValidateErrorNot: return RAPIDJSON_ERROR_STRING("Property matched the sub-schema specified by 'not'."); + + case kValidateErrorReadOnly: return RAPIDJSON_ERROR_STRING("Property is read-only but has been provided when validation is for writing."); + case kValidateErrorWriteOnly: return RAPIDJSON_ERROR_STRING("Property is write-only but has been provided when validation is for reading."); + + default: return RAPIDJSON_ERROR_STRING("Unknown error."); + } +} + +//! Maps error code of schema document compilation into error message. +/*! + \ingroup RAPIDJSON_ERRORS + \param schemaErrorCode Error code obtained from compiling the schema document. + \return the error message. + \note User can make a copy of this function for localization. + Using switch-case is safer for future modification of error codes. +*/ + inline const RAPIDJSON_ERROR_CHARTYPE* GetSchemaError_En(SchemaErrorCode schemaErrorCode) { + switch (schemaErrorCode) { + case kSchemaErrorNone: return RAPIDJSON_ERROR_STRING("No error."); + + case kSchemaErrorStartUnknown: return RAPIDJSON_ERROR_STRING("Pointer '%value' to start of schema does not resolve to a location in the document."); + case kSchemaErrorRefPlainName: return RAPIDJSON_ERROR_STRING("$ref fragment '%value' must be a JSON pointer."); + case kSchemaErrorRefInvalid: return RAPIDJSON_ERROR_STRING("$ref must not be an empty string."); + case kSchemaErrorRefPointerInvalid: return RAPIDJSON_ERROR_STRING("$ref fragment '%value' is not a valid JSON pointer at offset '%offset'."); + case kSchemaErrorRefUnknown: return RAPIDJSON_ERROR_STRING("$ref '%value' does not resolve to a location in the target document."); + case kSchemaErrorRefCyclical: return RAPIDJSON_ERROR_STRING("$ref '%value' is cyclical."); + case kSchemaErrorRefNoRemoteProvider: return RAPIDJSON_ERROR_STRING("$ref is remote but there is no remote provider."); + case kSchemaErrorRefNoRemoteSchema: return RAPIDJSON_ERROR_STRING("$ref '%value' is remote but the remote provider did not return a schema."); + case kSchemaErrorRegexInvalid: return RAPIDJSON_ERROR_STRING("Invalid regular expression '%value' in 'pattern' or 'patternProperties'."); + case kSchemaErrorSpecUnknown: return RAPIDJSON_ERROR_STRING("JSON schema draft or OpenAPI version is not recognized."); + case kSchemaErrorSpecUnsupported: return RAPIDJSON_ERROR_STRING("JSON schema draft or OpenAPI version is not supported."); + case kSchemaErrorSpecIllegal: return RAPIDJSON_ERROR_STRING("Both JSON schema draft and OpenAPI version found in document."); + case kSchemaErrorReadOnlyAndWriteOnly: return RAPIDJSON_ERROR_STRING("Property must not be both 'readOnly' and 'writeOnly'."); + + default: return RAPIDJSON_ERROR_STRING("Unknown error."); + } + } + +//! Maps error code of pointer parse into error message. +/*! + \ingroup RAPIDJSON_ERRORS + \param pointerParseErrorCode Error code obtained from pointer parse. + \return the error message. + \note User can make a copy of this function for localization. + Using switch-case is safer for future modification of error codes. +*/ +inline const RAPIDJSON_ERROR_CHARTYPE* GetPointerParseError_En(PointerParseErrorCode pointerParseErrorCode) { + switch (pointerParseErrorCode) { + case kPointerParseErrorNone: return RAPIDJSON_ERROR_STRING("No error."); + + case kPointerParseErrorTokenMustBeginWithSolidus: return RAPIDJSON_ERROR_STRING("A token must begin with a '/'."); + case kPointerParseErrorInvalidEscape: return RAPIDJSON_ERROR_STRING("Invalid escape."); + case kPointerParseErrorInvalidPercentEncoding: return RAPIDJSON_ERROR_STRING("Invalid percent encoding in URI fragment."); + case kPointerParseErrorCharacterMustPercentEncode: return RAPIDJSON_ERROR_STRING("A character must be percent encoded in a URI fragment."); + + default: return RAPIDJSON_ERROR_STRING("Unknown error."); + } +} + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_ERROR_EN_H_ diff --git a/include/rapidjson/error/error.h b/include/rapidjson/error/error.h new file mode 100644 index 0000000..cae345d --- /dev/null +++ b/include/rapidjson/error/error.h @@ -0,0 +1,285 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ERROR_ERROR_H_ +#define RAPIDJSON_ERROR_ERROR_H_ + +#include "../rapidjson.h" + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +#endif + +/*! \file error.h */ + +/*! \defgroup RAPIDJSON_ERRORS RapidJSON error handling */ + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_ERROR_CHARTYPE + +//! Character type of error messages. +/*! \ingroup RAPIDJSON_ERRORS + The default character type is \c char. + On Windows, user can define this macro as \c TCHAR for supporting both + unicode/non-unicode settings. +*/ +#ifndef RAPIDJSON_ERROR_CHARTYPE +#define RAPIDJSON_ERROR_CHARTYPE char +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_ERROR_STRING + +//! Macro for converting string literal to \ref RAPIDJSON_ERROR_CHARTYPE[]. +/*! \ingroup RAPIDJSON_ERRORS + By default this conversion macro does nothing. + On Windows, user can define this macro as \c _T(x) for supporting both + unicode/non-unicode settings. +*/ +#ifndef RAPIDJSON_ERROR_STRING +#define RAPIDJSON_ERROR_STRING(x) x +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// ParseErrorCode + +//! Error code of parsing. +/*! \ingroup RAPIDJSON_ERRORS + \see GenericReader::Parse, GenericReader::GetParseErrorCode +*/ +enum ParseErrorCode { + kParseErrorNone = 0, //!< No error. + + kParseErrorDocumentEmpty, //!< The document is empty. + kParseErrorDocumentRootNotSingular, //!< The document root must not follow by other values. + + kParseErrorValueInvalid, //!< Invalid value. + + kParseErrorObjectMissName, //!< Missing a name for object member. + kParseErrorObjectMissColon, //!< Missing a colon after a name of object member. + kParseErrorObjectMissCommaOrCurlyBracket, //!< Missing a comma or '}' after an object member. + + kParseErrorArrayMissCommaOrSquareBracket, //!< Missing a comma or ']' after an array element. + + kParseErrorStringUnicodeEscapeInvalidHex, //!< Incorrect hex digit after \\u escape in string. + kParseErrorStringUnicodeSurrogateInvalid, //!< The surrogate pair in string is invalid. + kParseErrorStringEscapeInvalid, //!< Invalid escape character in string. + kParseErrorStringMissQuotationMark, //!< Missing a closing quotation mark in string. + kParseErrorStringInvalidEncoding, //!< Invalid encoding in string. + + kParseErrorNumberTooBig, //!< Number too big to be stored in double. + kParseErrorNumberMissFraction, //!< Miss fraction part in number. + kParseErrorNumberMissExponent, //!< Miss exponent in number. + + kParseErrorTermination, //!< Parsing was terminated. + kParseErrorUnspecificSyntaxError //!< Unspecific syntax error. +}; + +//! Result of parsing (wraps ParseErrorCode) +/*! + \ingroup RAPIDJSON_ERRORS + \code + Document doc; + ParseResult ok = doc.Parse("[42]"); + if (!ok) { + fprintf(stderr, "JSON parse error: %s (%u)", + GetParseError_En(ok.Code()), ok.Offset()); + exit(EXIT_FAILURE); + } + \endcode + \see GenericReader::Parse, GenericDocument::Parse +*/ +struct ParseResult { + //!! Unspecified boolean type + typedef bool (ParseResult::*BooleanType)() const; +public: + //! Default constructor, no error. + ParseResult() : code_(kParseErrorNone), offset_(0) {} + //! Constructor to set an error. + ParseResult(ParseErrorCode code, size_t offset) : code_(code), offset_(offset) {} + + //! Get the error code. + ParseErrorCode Code() const { return code_; } + //! Get the error offset, if \ref IsError(), 0 otherwise. + size_t Offset() const { return offset_; } + + //! Explicit conversion to \c bool, returns \c true, iff !\ref IsError(). + operator BooleanType() const { return !IsError() ? &ParseResult::IsError : NULL; } + //! Whether the result is an error. + bool IsError() const { return code_ != kParseErrorNone; } + + bool operator==(const ParseResult& that) const { return code_ == that.code_; } + bool operator==(ParseErrorCode code) const { return code_ == code; } + friend bool operator==(ParseErrorCode code, const ParseResult & err) { return code == err.code_; } + + bool operator!=(const ParseResult& that) const { return !(*this == that); } + bool operator!=(ParseErrorCode code) const { return !(*this == code); } + friend bool operator!=(ParseErrorCode code, const ParseResult & err) { return err != code; } + + //! Reset error code. + void Clear() { Set(kParseErrorNone); } + //! Update error code and offset. + void Set(ParseErrorCode code, size_t offset = 0) { code_ = code; offset_ = offset; } + +private: + ParseErrorCode code_; + size_t offset_; +}; + +//! Function pointer type of GetParseError(). +/*! \ingroup RAPIDJSON_ERRORS + + This is the prototype for \c GetParseError_X(), where \c X is a locale. + User can dynamically change locale in runtime, e.g.: +\code + GetParseErrorFunc GetParseError = GetParseError_En; // or whatever + const RAPIDJSON_ERROR_CHARTYPE* s = GetParseError(document.GetParseErrorCode()); +\endcode +*/ +typedef const RAPIDJSON_ERROR_CHARTYPE* (*GetParseErrorFunc)(ParseErrorCode); + +/////////////////////////////////////////////////////////////////////////////// +// ValidateErrorCode + +//! Error codes when validating. +/*! \ingroup RAPIDJSON_ERRORS + \see GenericSchemaValidator +*/ +enum ValidateErrorCode { + kValidateErrors = -1, //!< Top level error code when kValidateContinueOnErrorsFlag set. + kValidateErrorNone = 0, //!< No error. + + kValidateErrorMultipleOf, //!< Number is not a multiple of the 'multipleOf' value. + kValidateErrorMaximum, //!< Number is greater than the 'maximum' value. + kValidateErrorExclusiveMaximum, //!< Number is greater than or equal to the 'maximum' value. + kValidateErrorMinimum, //!< Number is less than the 'minimum' value. + kValidateErrorExclusiveMinimum, //!< Number is less than or equal to the 'minimum' value. + + kValidateErrorMaxLength, //!< String is longer than the 'maxLength' value. + kValidateErrorMinLength, //!< String is longer than the 'maxLength' value. + kValidateErrorPattern, //!< String does not match the 'pattern' regular expression. + + kValidateErrorMaxItems, //!< Array is longer than the 'maxItems' value. + kValidateErrorMinItems, //!< Array is shorter than the 'minItems' value. + kValidateErrorUniqueItems, //!< Array has duplicate items but 'uniqueItems' is true. + kValidateErrorAdditionalItems, //!< Array has additional items that are not allowed by the schema. + + kValidateErrorMaxProperties, //!< Object has more members than 'maxProperties' value. + kValidateErrorMinProperties, //!< Object has less members than 'minProperties' value. + kValidateErrorRequired, //!< Object is missing one or more members required by the schema. + kValidateErrorAdditionalProperties, //!< Object has additional members that are not allowed by the schema. + kValidateErrorPatternProperties, //!< See other errors. + kValidateErrorDependencies, //!< Object has missing property or schema dependencies. + + kValidateErrorEnum, //!< Property has a value that is not one of its allowed enumerated values. + kValidateErrorType, //!< Property has a type that is not allowed by the schema. + + kValidateErrorOneOf, //!< Property did not match any of the sub-schemas specified by 'oneOf'. + kValidateErrorOneOfMatch, //!< Property matched more than one of the sub-schemas specified by 'oneOf'. + kValidateErrorAllOf, //!< Property did not match all of the sub-schemas specified by 'allOf'. + kValidateErrorAnyOf, //!< Property did not match any of the sub-schemas specified by 'anyOf'. + kValidateErrorNot, //!< Property matched the sub-schema specified by 'not'. + + kValidateErrorReadOnly, //!< Property is read-only but has been provided when validation is for writing + kValidateErrorWriteOnly //!< Property is write-only but has been provided when validation is for reading +}; + +//! Function pointer type of GetValidateError(). +/*! \ingroup RAPIDJSON_ERRORS + + This is the prototype for \c GetValidateError_X(), where \c X is a locale. + User can dynamically change locale in runtime, e.g.: +\code + GetValidateErrorFunc GetValidateError = GetValidateError_En; // or whatever + const RAPIDJSON_ERROR_CHARTYPE* s = GetValidateError(validator.GetInvalidSchemaCode()); +\endcode +*/ +typedef const RAPIDJSON_ERROR_CHARTYPE* (*GetValidateErrorFunc)(ValidateErrorCode); + +/////////////////////////////////////////////////////////////////////////////// +// SchemaErrorCode + +//! Error codes when validating. +/*! \ingroup RAPIDJSON_ERRORS + \see GenericSchemaValidator +*/ +enum SchemaErrorCode { + kSchemaErrorNone = 0, //!< No error. + + kSchemaErrorStartUnknown, //!< Pointer to start of schema does not resolve to a location in the document + kSchemaErrorRefPlainName, //!< $ref fragment must be a JSON pointer + kSchemaErrorRefInvalid, //!< $ref must not be an empty string + kSchemaErrorRefPointerInvalid, //!< $ref fragment is not a valid JSON pointer at offset + kSchemaErrorRefUnknown, //!< $ref does not resolve to a location in the target document + kSchemaErrorRefCyclical, //!< $ref is cyclical + kSchemaErrorRefNoRemoteProvider, //!< $ref is remote but there is no remote provider + kSchemaErrorRefNoRemoteSchema, //!< $ref is remote but the remote provider did not return a schema + kSchemaErrorRegexInvalid, //!< Invalid regular expression in 'pattern' or 'patternProperties' + kSchemaErrorSpecUnknown, //!< JSON schema draft or OpenAPI version is not recognized + kSchemaErrorSpecUnsupported, //!< JSON schema draft or OpenAPI version is not supported + kSchemaErrorSpecIllegal, //!< Both JSON schema draft and OpenAPI version found in document + kSchemaErrorReadOnlyAndWriteOnly //!< Property must not be both 'readOnly' and 'writeOnly' +}; + +//! Function pointer type of GetSchemaError(). +/*! \ingroup RAPIDJSON_ERRORS + + This is the prototype for \c GetSchemaError_X(), where \c X is a locale. + User can dynamically change locale in runtime, e.g.: +\code + GetSchemaErrorFunc GetSchemaError = GetSchemaError_En; // or whatever + const RAPIDJSON_ERROR_CHARTYPE* s = GetSchemaError(validator.GetInvalidSchemaCode()); +\endcode +*/ +typedef const RAPIDJSON_ERROR_CHARTYPE* (*GetSchemaErrorFunc)(SchemaErrorCode); + +/////////////////////////////////////////////////////////////////////////////// +// PointerParseErrorCode + +//! Error code of JSON pointer parsing. +/*! \ingroup RAPIDJSON_ERRORS + \see GenericPointer::GenericPointer, GenericPointer::GetParseErrorCode +*/ +enum PointerParseErrorCode { + kPointerParseErrorNone = 0, //!< The parse is successful + + kPointerParseErrorTokenMustBeginWithSolidus, //!< A token must begin with a '/' + kPointerParseErrorInvalidEscape, //!< Invalid escape + kPointerParseErrorInvalidPercentEncoding, //!< Invalid percent encoding in URI fragment + kPointerParseErrorCharacterMustPercentEncode //!< A character must percent encoded in URI fragment +}; + +//! Function pointer type of GetPointerParseError(). +/*! \ingroup RAPIDJSON_ERRORS + + This is the prototype for \c GetPointerParseError_X(), where \c X is a locale. + User can dynamically change locale in runtime, e.g.: +\code + GetPointerParseErrorFunc GetPointerParseError = GetPointerParseError_En; // or whatever + const RAPIDJSON_ERROR_CHARTYPE* s = GetPointerParseError(pointer.GetParseErrorCode()); +\endcode +*/ +typedef const RAPIDJSON_ERROR_CHARTYPE* (*GetPointerParseErrorFunc)(PointerParseErrorCode); + + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_ERROR_ERROR_H_ diff --git a/include/rapidjson/filereadstream.h b/include/rapidjson/filereadstream.h new file mode 100644 index 0000000..f8bb43c --- /dev/null +++ b/include/rapidjson/filereadstream.h @@ -0,0 +1,99 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_FILEREADSTREAM_H_ +#define RAPIDJSON_FILEREADSTREAM_H_ + +#include "stream.h" +#include + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +RAPIDJSON_DIAG_OFF(unreachable-code) +RAPIDJSON_DIAG_OFF(missing-noreturn) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! File byte stream for input using fread(). +/*! + \note implements Stream concept +*/ +class FileReadStream { +public: + typedef char Ch; //!< Character type (byte). + + //! Constructor. + /*! + \param fp File pointer opened for read. + \param buffer user-supplied buffer. + \param bufferSize size of buffer in bytes. Must >=4 bytes. + */ + FileReadStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferSize_(bufferSize), bufferLast_(0), current_(buffer_), readCount_(0), count_(0), eof_(false) { + RAPIDJSON_ASSERT(fp_ != 0); + RAPIDJSON_ASSERT(bufferSize >= 4); + Read(); + } + + Ch Peek() const { return *current_; } + Ch Take() { Ch c = *current_; Read(); return c; } + size_t Tell() const { return count_ + static_cast(current_ - buffer_); } + + // Not implemented + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + + // For encoding detection only. + const Ch* Peek4() const { + return (current_ + 4 - !eof_ <= bufferLast_) ? current_ : 0; + } + +private: + void Read() { + if (current_ < bufferLast_) + ++current_; + else if (!eof_) { + count_ += readCount_; + readCount_ = std::fread(buffer_, 1, bufferSize_, fp_); + bufferLast_ = buffer_ + readCount_ - 1; + current_ = buffer_; + + if (readCount_ < bufferSize_) { + buffer_[readCount_] = '\0'; + ++bufferLast_; + eof_ = true; + } + } + } + + std::FILE* fp_; + Ch *buffer_; + size_t bufferSize_; + Ch *bufferLast_; + Ch *current_; + size_t readCount_; + size_t count_; //!< Number of characters read + bool eof_; +}; + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_FILESTREAM_H_ diff --git a/include/rapidjson/filewritestream.h b/include/rapidjson/filewritestream.h new file mode 100644 index 0000000..5d89588 --- /dev/null +++ b/include/rapidjson/filewritestream.h @@ -0,0 +1,104 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_FILEWRITESTREAM_H_ +#define RAPIDJSON_FILEWRITESTREAM_H_ + +#include "stream.h" +#include + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(unreachable-code) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Wrapper of C file stream for output using fwrite(). +/*! + \note implements Stream concept +*/ +class FileWriteStream { +public: + typedef char Ch; //!< Character type. Only support char. + + FileWriteStream(std::FILE* fp, char* buffer, size_t bufferSize) : fp_(fp), buffer_(buffer), bufferEnd_(buffer + bufferSize), current_(buffer_) { + RAPIDJSON_ASSERT(fp_ != 0); + } + + void Put(char c) { + if (current_ >= bufferEnd_) + Flush(); + + *current_++ = c; + } + + void PutN(char c, size_t n) { + size_t avail = static_cast(bufferEnd_ - current_); + while (n > avail) { + std::memset(current_, c, avail); + current_ += avail; + Flush(); + n -= avail; + avail = static_cast(bufferEnd_ - current_); + } + + if (n > 0) { + std::memset(current_, c, n); + current_ += n; + } + } + + void Flush() { + if (current_ != buffer_) { + size_t result = std::fwrite(buffer_, 1, static_cast(current_ - buffer_), fp_); + if (result < static_cast(current_ - buffer_)) { + // failure deliberately ignored at this time + // added to avoid warn_unused_result build errors + } + current_ = buffer_; + } + } + + // Not implemented + char Peek() const { RAPIDJSON_ASSERT(false); return 0; } + char Take() { RAPIDJSON_ASSERT(false); return 0; } + size_t Tell() const { RAPIDJSON_ASSERT(false); return 0; } + char* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(char*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + // Prohibit copy constructor & assignment operator. + FileWriteStream(const FileWriteStream&); + FileWriteStream& operator=(const FileWriteStream&); + + std::FILE* fp_; + char *buffer_; + char *bufferEnd_; + char *current_; +}; + +//! Implement specialized version of PutN() with memset() for better performance. +template<> +inline void PutN(FileWriteStream& stream, char c, size_t n) { + stream.PutN(c, n); +} + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_FILESTREAM_H_ diff --git a/include/rapidjson/fwd.h b/include/rapidjson/fwd.h new file mode 100644 index 0000000..d62f77f --- /dev/null +++ b/include/rapidjson/fwd.h @@ -0,0 +1,151 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_FWD_H_ +#define RAPIDJSON_FWD_H_ + +#include "rapidjson.h" + +RAPIDJSON_NAMESPACE_BEGIN + +// encodings.h + +template struct UTF8; +template struct UTF16; +template struct UTF16BE; +template struct UTF16LE; +template struct UTF32; +template struct UTF32BE; +template struct UTF32LE; +template struct ASCII; +template struct AutoUTF; + +template +struct Transcoder; + +// allocators.h + +class CrtAllocator; + +template +class MemoryPoolAllocator; + +// stream.h + +template +struct GenericStringStream; + +typedef GenericStringStream > StringStream; + +template +struct GenericInsituStringStream; + +typedef GenericInsituStringStream > InsituStringStream; + +// stringbuffer.h + +template +class GenericStringBuffer; + +typedef GenericStringBuffer, CrtAllocator> StringBuffer; + +// filereadstream.h + +class FileReadStream; + +// filewritestream.h + +class FileWriteStream; + +// memorybuffer.h + +template +struct GenericMemoryBuffer; + +typedef GenericMemoryBuffer MemoryBuffer; + +// memorystream.h + +struct MemoryStream; + +// reader.h + +template +struct BaseReaderHandler; + +template +class GenericReader; + +typedef GenericReader, UTF8, CrtAllocator> Reader; + +// writer.h + +template +class Writer; + +// prettywriter.h + +template +class PrettyWriter; + +// document.h + +template +class GenericMember; + +template +class GenericMemberIterator; + +template +struct GenericStringRef; + +template +class GenericValue; + +typedef GenericValue, MemoryPoolAllocator > Value; + +template +class GenericDocument; + +typedef GenericDocument, MemoryPoolAllocator, CrtAllocator> Document; + +// pointer.h + +template +class GenericPointer; + +typedef GenericPointer Pointer; + +// schema.h + +template +class IGenericRemoteSchemaDocumentProvider; + +template +class GenericSchemaDocument; + +typedef GenericSchemaDocument SchemaDocument; +typedef IGenericRemoteSchemaDocumentProvider IRemoteSchemaDocumentProvider; + +template < + typename SchemaDocumentType, + typename OutputHandler, + typename StateAllocator> +class GenericSchemaValidator; + +typedef GenericSchemaValidator, void>, CrtAllocator> SchemaValidator; + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_RAPIDJSONFWD_H_ diff --git a/include/rapidjson/internal/biginteger.h b/include/rapidjson/internal/biginteger.h new file mode 100644 index 0000000..4930043 --- /dev/null +++ b/include/rapidjson/internal/biginteger.h @@ -0,0 +1,297 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_BIGINTEGER_H_ +#define RAPIDJSON_BIGINTEGER_H_ + +#include "../rapidjson.h" + +#if defined(_MSC_VER) && !defined(__INTEL_COMPILER) && defined(_M_AMD64) +#include // for _umul128 +#if !defined(_ARM64EC_) +#pragma intrinsic(_umul128) +#else +#pragma comment(lib,"softintrin") +#endif +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +class BigInteger { +public: + typedef uint64_t Type; + + BigInteger(const BigInteger& rhs) : count_(rhs.count_) { + std::memcpy(digits_, rhs.digits_, count_ * sizeof(Type)); + } + + explicit BigInteger(uint64_t u) : count_(1) { + digits_[0] = u; + } + + template + BigInteger(const Ch* decimals, size_t length) : count_(1) { + RAPIDJSON_ASSERT(length > 0); + digits_[0] = 0; + size_t i = 0; + const size_t kMaxDigitPerIteration = 19; // 2^64 = 18446744073709551616 > 10^19 + while (length >= kMaxDigitPerIteration) { + AppendDecimal64(decimals + i, decimals + i + kMaxDigitPerIteration); + length -= kMaxDigitPerIteration; + i += kMaxDigitPerIteration; + } + + if (length > 0) + AppendDecimal64(decimals + i, decimals + i + length); + } + + BigInteger& operator=(const BigInteger &rhs) + { + if (this != &rhs) { + count_ = rhs.count_; + std::memcpy(digits_, rhs.digits_, count_ * sizeof(Type)); + } + return *this; + } + + BigInteger& operator=(uint64_t u) { + digits_[0] = u; + count_ = 1; + return *this; + } + + BigInteger& operator+=(uint64_t u) { + Type backup = digits_[0]; + digits_[0] += u; + for (size_t i = 0; i < count_ - 1; i++) { + if (digits_[i] >= backup) + return *this; // no carry + backup = digits_[i + 1]; + digits_[i + 1] += 1; + } + + // Last carry + if (digits_[count_ - 1] < backup) + PushBack(1); + + return *this; + } + + BigInteger& operator*=(uint64_t u) { + if (u == 0) return *this = 0; + if (u == 1) return *this; + if (*this == 1) return *this = u; + + uint64_t k = 0; + for (size_t i = 0; i < count_; i++) { + uint64_t hi; + digits_[i] = MulAdd64(digits_[i], u, k, &hi); + k = hi; + } + + if (k > 0) + PushBack(k); + + return *this; + } + + BigInteger& operator*=(uint32_t u) { + if (u == 0) return *this = 0; + if (u == 1) return *this; + if (*this == 1) return *this = u; + + uint64_t k = 0; + for (size_t i = 0; i < count_; i++) { + const uint64_t c = digits_[i] >> 32; + const uint64_t d = digits_[i] & 0xFFFFFFFF; + const uint64_t uc = u * c; + const uint64_t ud = u * d; + const uint64_t p0 = ud + k; + const uint64_t p1 = uc + (p0 >> 32); + digits_[i] = (p0 & 0xFFFFFFFF) | (p1 << 32); + k = p1 >> 32; + } + + if (k > 0) + PushBack(k); + + return *this; + } + + BigInteger& operator<<=(size_t shift) { + if (IsZero() || shift == 0) return *this; + + size_t offset = shift / kTypeBit; + size_t interShift = shift % kTypeBit; + RAPIDJSON_ASSERT(count_ + offset <= kCapacity); + + if (interShift == 0) { + std::memmove(digits_ + offset, digits_, count_ * sizeof(Type)); + count_ += offset; + } + else { + digits_[count_] = 0; + for (size_t i = count_; i > 0; i--) + digits_[i + offset] = (digits_[i] << interShift) | (digits_[i - 1] >> (kTypeBit - interShift)); + digits_[offset] = digits_[0] << interShift; + count_ += offset; + if (digits_[count_]) + count_++; + } + + std::memset(digits_, 0, offset * sizeof(Type)); + + return *this; + } + + bool operator==(const BigInteger& rhs) const { + return count_ == rhs.count_ && std::memcmp(digits_, rhs.digits_, count_ * sizeof(Type)) == 0; + } + + bool operator==(const Type rhs) const { + return count_ == 1 && digits_[0] == rhs; + } + + BigInteger& MultiplyPow5(unsigned exp) { + static const uint32_t kPow5[12] = { + 5, + 5 * 5, + 5 * 5 * 5, + 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5, + 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 * 5 + }; + if (exp == 0) return *this; + for (; exp >= 27; exp -= 27) *this *= RAPIDJSON_UINT64_C2(0X6765C793, 0XFA10079D); // 5^27 + for (; exp >= 13; exp -= 13) *this *= static_cast(1220703125u); // 5^13 + if (exp > 0) *this *= kPow5[exp - 1]; + return *this; + } + + // Compute absolute difference of this and rhs. + // Assume this != rhs + bool Difference(const BigInteger& rhs, BigInteger* out) const { + int cmp = Compare(rhs); + RAPIDJSON_ASSERT(cmp != 0); + const BigInteger *a, *b; // Makes a > b + bool ret; + if (cmp < 0) { a = &rhs; b = this; ret = true; } + else { a = this; b = &rhs; ret = false; } + + Type borrow = 0; + for (size_t i = 0; i < a->count_; i++) { + Type d = a->digits_[i] - borrow; + if (i < b->count_) + d -= b->digits_[i]; + borrow = (d > a->digits_[i]) ? 1 : 0; + out->digits_[i] = d; + if (d != 0) + out->count_ = i + 1; + } + + return ret; + } + + int Compare(const BigInteger& rhs) const { + if (count_ != rhs.count_) + return count_ < rhs.count_ ? -1 : 1; + + for (size_t i = count_; i-- > 0;) + if (digits_[i] != rhs.digits_[i]) + return digits_[i] < rhs.digits_[i] ? -1 : 1; + + return 0; + } + + size_t GetCount() const { return count_; } + Type GetDigit(size_t index) const { RAPIDJSON_ASSERT(index < count_); return digits_[index]; } + bool IsZero() const { return count_ == 1 && digits_[0] == 0; } + +private: + template + void AppendDecimal64(const Ch* begin, const Ch* end) { + uint64_t u = ParseUint64(begin, end); + if (IsZero()) + *this = u; + else { + unsigned exp = static_cast(end - begin); + (MultiplyPow5(exp) <<= exp) += u; // *this = *this * 10^exp + u + } + } + + void PushBack(Type digit) { + RAPIDJSON_ASSERT(count_ < kCapacity); + digits_[count_++] = digit; + } + + template + static uint64_t ParseUint64(const Ch* begin, const Ch* end) { + uint64_t r = 0; + for (const Ch* p = begin; p != end; ++p) { + RAPIDJSON_ASSERT(*p >= Ch('0') && *p <= Ch('9')); + r = r * 10u + static_cast(*p - Ch('0')); + } + return r; + } + + // Assume a * b + k < 2^128 + static uint64_t MulAdd64(uint64_t a, uint64_t b, uint64_t k, uint64_t* outHigh) { +#if defined(_MSC_VER) && defined(_M_AMD64) + uint64_t low = _umul128(a, b, outHigh) + k; + if (low < k) + (*outHigh)++; + return low; +#elif defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 6)) && defined(__x86_64__) + __extension__ typedef unsigned __int128 uint128; + uint128 p = static_cast(a) * static_cast(b); + p += k; + *outHigh = static_cast(p >> 64); + return static_cast(p); +#else + const uint64_t a0 = a & 0xFFFFFFFF, a1 = a >> 32, b0 = b & 0xFFFFFFFF, b1 = b >> 32; + uint64_t x0 = a0 * b0, x1 = a0 * b1, x2 = a1 * b0, x3 = a1 * b1; + x1 += (x0 >> 32); // can't give carry + x1 += x2; + if (x1 < x2) + x3 += (static_cast(1) << 32); + uint64_t lo = (x1 << 32) + (x0 & 0xFFFFFFFF); + uint64_t hi = x3 + (x1 >> 32); + + lo += k; + if (lo < k) + hi++; + *outHigh = hi; + return lo; +#endif + } + + static const size_t kBitCount = 3328; // 64bit * 54 > 10^1000 + static const size_t kCapacity = kBitCount / sizeof(Type); + static const size_t kTypeBit = sizeof(Type) * 8; + + Type digits_[kCapacity]; + size_t count_; +}; + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_BIGINTEGER_H_ diff --git a/include/rapidjson/internal/clzll.h b/include/rapidjson/internal/clzll.h new file mode 100644 index 0000000..8fc5118 --- /dev/null +++ b/include/rapidjson/internal/clzll.h @@ -0,0 +1,71 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_CLZLL_H_ +#define RAPIDJSON_CLZLL_H_ + +#include "../rapidjson.h" + +#if defined(_MSC_VER) && !defined(UNDER_CE) +#include +#if defined(_WIN64) +#pragma intrinsic(_BitScanReverse64) +#else +#pragma intrinsic(_BitScanReverse) +#endif +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +inline uint32_t clzll(uint64_t x) { + // Passing 0 to __builtin_clzll is UB in GCC and results in an + // infinite loop in the software implementation. + RAPIDJSON_ASSERT(x != 0); + +#if defined(_MSC_VER) && !defined(UNDER_CE) + unsigned long r = 0; +#if defined(_WIN64) + _BitScanReverse64(&r, x); +#else + // Scan the high 32 bits. + if (_BitScanReverse(&r, static_cast(x >> 32))) + return 63 - (r + 32); + + // Scan the low 32 bits. + _BitScanReverse(&r, static_cast(x & 0xFFFFFFFF)); +#endif // _WIN64 + + return 63 - r; +#elif (defined(__GNUC__) && __GNUC__ >= 4) || RAPIDJSON_HAS_BUILTIN(__builtin_clzll) + // __builtin_clzll wrapper + return static_cast(__builtin_clzll(x)); +#else + // naive version + uint32_t r = 0; + while (!(x & (static_cast(1) << 63))) { + x <<= 1; + ++r; + } + + return r; +#endif // _MSC_VER +} + +#define RAPIDJSON_CLZLL RAPIDJSON_NAMESPACE::internal::clzll + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_CLZLL_H_ diff --git a/include/rapidjson/internal/diyfp.h b/include/rapidjson/internal/diyfp.h new file mode 100644 index 0000000..1f60fb6 --- /dev/null +++ b/include/rapidjson/internal/diyfp.h @@ -0,0 +1,261 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +// This is a C++ header-only implementation of Grisu2 algorithm from the publication: +// Loitsch, Florian. "Printing floating-point numbers quickly and accurately with +// integers." ACM Sigplan Notices 45.6 (2010): 233-243. + +#ifndef RAPIDJSON_DIYFP_H_ +#define RAPIDJSON_DIYFP_H_ + +#include "../rapidjson.h" +#include "clzll.h" +#include + +#if defined(_MSC_VER) && defined(_M_AMD64) && !defined(__INTEL_COMPILER) +#include +#if !defined(_ARM64EC_) +#pragma intrinsic(_umul128) +#else +#pragma comment(lib,"softintrin") +#endif +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +#endif + +struct DiyFp { + DiyFp() : f(), e() {} + + DiyFp(uint64_t fp, int exp) : f(fp), e(exp) {} + + explicit DiyFp(double d) { + union { + double d; + uint64_t u64; + } u = { d }; + + int biased_e = static_cast((u.u64 & kDpExponentMask) >> kDpSignificandSize); + uint64_t significand = (u.u64 & kDpSignificandMask); + if (biased_e != 0) { + f = significand + kDpHiddenBit; + e = biased_e - kDpExponentBias; + } + else { + f = significand; + e = kDpMinExponent + 1; + } + } + + DiyFp operator-(const DiyFp& rhs) const { + return DiyFp(f - rhs.f, e); + } + + DiyFp operator*(const DiyFp& rhs) const { +#if defined(_MSC_VER) && defined(_M_AMD64) + uint64_t h; + uint64_t l = _umul128(f, rhs.f, &h); + if (l & (uint64_t(1) << 63)) // rounding + h++; + return DiyFp(h, e + rhs.e + 64); +#elif defined(__GNUC__) && (__GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 6)) && defined(__x86_64__) + __extension__ typedef unsigned __int128 uint128; + uint128 p = static_cast(f) * static_cast(rhs.f); + uint64_t h = static_cast(p >> 64); + uint64_t l = static_cast(p); + if (l & (uint64_t(1) << 63)) // rounding + h++; + return DiyFp(h, e + rhs.e + 64); +#else + const uint64_t M32 = 0xFFFFFFFF; + const uint64_t a = f >> 32; + const uint64_t b = f & M32; + const uint64_t c = rhs.f >> 32; + const uint64_t d = rhs.f & M32; + const uint64_t ac = a * c; + const uint64_t bc = b * c; + const uint64_t ad = a * d; + const uint64_t bd = b * d; + uint64_t tmp = (bd >> 32) + (ad & M32) + (bc & M32); + tmp += 1U << 31; /// mult_round + return DiyFp(ac + (ad >> 32) + (bc >> 32) + (tmp >> 32), e + rhs.e + 64); +#endif + } + + DiyFp Normalize() const { + int s = static_cast(clzll(f)); + return DiyFp(f << s, e - s); + } + + DiyFp NormalizeBoundary() const { + DiyFp res = *this; + while (!(res.f & (kDpHiddenBit << 1))) { + res.f <<= 1; + res.e--; + } + res.f <<= (kDiySignificandSize - kDpSignificandSize - 2); + res.e = res.e - (kDiySignificandSize - kDpSignificandSize - 2); + return res; + } + + void NormalizedBoundaries(DiyFp* minus, DiyFp* plus) const { + DiyFp pl = DiyFp((f << 1) + 1, e - 1).NormalizeBoundary(); + DiyFp mi = (f == kDpHiddenBit) ? DiyFp((f << 2) - 1, e - 2) : DiyFp((f << 1) - 1, e - 1); + mi.f <<= mi.e - pl.e; + mi.e = pl.e; + *plus = pl; + *minus = mi; + } + + double ToDouble() const { + union { + double d; + uint64_t u64; + }u; + RAPIDJSON_ASSERT(f <= kDpHiddenBit + kDpSignificandMask); + if (e < kDpDenormalExponent) { + // Underflow. + return 0.0; + } + if (e >= kDpMaxExponent) { + // Overflow. + return std::numeric_limits::infinity(); + } + const uint64_t be = (e == kDpDenormalExponent && (f & kDpHiddenBit) == 0) ? 0 : + static_cast(e + kDpExponentBias); + u.u64 = (f & kDpSignificandMask) | (be << kDpSignificandSize); + return u.d; + } + + static const int kDiySignificandSize = 64; + static const int kDpSignificandSize = 52; + static const int kDpExponentBias = 0x3FF + kDpSignificandSize; + static const int kDpMaxExponent = 0x7FF - kDpExponentBias; + static const int kDpMinExponent = -kDpExponentBias; + static const int kDpDenormalExponent = -kDpExponentBias + 1; + static const uint64_t kDpExponentMask = RAPIDJSON_UINT64_C2(0x7FF00000, 0x00000000); + static const uint64_t kDpSignificandMask = RAPIDJSON_UINT64_C2(0x000FFFFF, 0xFFFFFFFF); + static const uint64_t kDpHiddenBit = RAPIDJSON_UINT64_C2(0x00100000, 0x00000000); + + uint64_t f; + int e; +}; + +inline DiyFp GetCachedPowerByIndex(size_t index) { + // 10^-348, 10^-340, ..., 10^340 + static const uint64_t kCachedPowers_F[] = { + RAPIDJSON_UINT64_C2(0xfa8fd5a0, 0x081c0288), RAPIDJSON_UINT64_C2(0xbaaee17f, 0xa23ebf76), + RAPIDJSON_UINT64_C2(0x8b16fb20, 0x3055ac76), RAPIDJSON_UINT64_C2(0xcf42894a, 0x5dce35ea), + RAPIDJSON_UINT64_C2(0x9a6bb0aa, 0x55653b2d), RAPIDJSON_UINT64_C2(0xe61acf03, 0x3d1a45df), + RAPIDJSON_UINT64_C2(0xab70fe17, 0xc79ac6ca), RAPIDJSON_UINT64_C2(0xff77b1fc, 0xbebcdc4f), + RAPIDJSON_UINT64_C2(0xbe5691ef, 0x416bd60c), RAPIDJSON_UINT64_C2(0x8dd01fad, 0x907ffc3c), + RAPIDJSON_UINT64_C2(0xd3515c28, 0x31559a83), RAPIDJSON_UINT64_C2(0x9d71ac8f, 0xada6c9b5), + RAPIDJSON_UINT64_C2(0xea9c2277, 0x23ee8bcb), RAPIDJSON_UINT64_C2(0xaecc4991, 0x4078536d), + RAPIDJSON_UINT64_C2(0x823c1279, 0x5db6ce57), RAPIDJSON_UINT64_C2(0xc2109436, 0x4dfb5637), + RAPIDJSON_UINT64_C2(0x9096ea6f, 0x3848984f), RAPIDJSON_UINT64_C2(0xd77485cb, 0x25823ac7), + RAPIDJSON_UINT64_C2(0xa086cfcd, 0x97bf97f4), RAPIDJSON_UINT64_C2(0xef340a98, 0x172aace5), + RAPIDJSON_UINT64_C2(0xb23867fb, 0x2a35b28e), RAPIDJSON_UINT64_C2(0x84c8d4df, 0xd2c63f3b), + RAPIDJSON_UINT64_C2(0xc5dd4427, 0x1ad3cdba), RAPIDJSON_UINT64_C2(0x936b9fce, 0xbb25c996), + RAPIDJSON_UINT64_C2(0xdbac6c24, 0x7d62a584), RAPIDJSON_UINT64_C2(0xa3ab6658, 0x0d5fdaf6), + RAPIDJSON_UINT64_C2(0xf3e2f893, 0xdec3f126), RAPIDJSON_UINT64_C2(0xb5b5ada8, 0xaaff80b8), + RAPIDJSON_UINT64_C2(0x87625f05, 0x6c7c4a8b), RAPIDJSON_UINT64_C2(0xc9bcff60, 0x34c13053), + RAPIDJSON_UINT64_C2(0x964e858c, 0x91ba2655), RAPIDJSON_UINT64_C2(0xdff97724, 0x70297ebd), + RAPIDJSON_UINT64_C2(0xa6dfbd9f, 0xb8e5b88f), RAPIDJSON_UINT64_C2(0xf8a95fcf, 0x88747d94), + RAPIDJSON_UINT64_C2(0xb9447093, 0x8fa89bcf), RAPIDJSON_UINT64_C2(0x8a08f0f8, 0xbf0f156b), + RAPIDJSON_UINT64_C2(0xcdb02555, 0x653131b6), RAPIDJSON_UINT64_C2(0x993fe2c6, 0xd07b7fac), + RAPIDJSON_UINT64_C2(0xe45c10c4, 0x2a2b3b06), RAPIDJSON_UINT64_C2(0xaa242499, 0x697392d3), + RAPIDJSON_UINT64_C2(0xfd87b5f2, 0x8300ca0e), RAPIDJSON_UINT64_C2(0xbce50864, 0x92111aeb), + RAPIDJSON_UINT64_C2(0x8cbccc09, 0x6f5088cc), RAPIDJSON_UINT64_C2(0xd1b71758, 0xe219652c), + RAPIDJSON_UINT64_C2(0x9c400000, 0x00000000), RAPIDJSON_UINT64_C2(0xe8d4a510, 0x00000000), + RAPIDJSON_UINT64_C2(0xad78ebc5, 0xac620000), RAPIDJSON_UINT64_C2(0x813f3978, 0xf8940984), + RAPIDJSON_UINT64_C2(0xc097ce7b, 0xc90715b3), RAPIDJSON_UINT64_C2(0x8f7e32ce, 0x7bea5c70), + RAPIDJSON_UINT64_C2(0xd5d238a4, 0xabe98068), RAPIDJSON_UINT64_C2(0x9f4f2726, 0x179a2245), + RAPIDJSON_UINT64_C2(0xed63a231, 0xd4c4fb27), RAPIDJSON_UINT64_C2(0xb0de6538, 0x8cc8ada8), + RAPIDJSON_UINT64_C2(0x83c7088e, 0x1aab65db), RAPIDJSON_UINT64_C2(0xc45d1df9, 0x42711d9a), + RAPIDJSON_UINT64_C2(0x924d692c, 0xa61be758), RAPIDJSON_UINT64_C2(0xda01ee64, 0x1a708dea), + RAPIDJSON_UINT64_C2(0xa26da399, 0x9aef774a), RAPIDJSON_UINT64_C2(0xf209787b, 0xb47d6b85), + RAPIDJSON_UINT64_C2(0xb454e4a1, 0x79dd1877), RAPIDJSON_UINT64_C2(0x865b8692, 0x5b9bc5c2), + RAPIDJSON_UINT64_C2(0xc83553c5, 0xc8965d3d), RAPIDJSON_UINT64_C2(0x952ab45c, 0xfa97a0b3), + RAPIDJSON_UINT64_C2(0xde469fbd, 0x99a05fe3), RAPIDJSON_UINT64_C2(0xa59bc234, 0xdb398c25), + RAPIDJSON_UINT64_C2(0xf6c69a72, 0xa3989f5c), RAPIDJSON_UINT64_C2(0xb7dcbf53, 0x54e9bece), + RAPIDJSON_UINT64_C2(0x88fcf317, 0xf22241e2), RAPIDJSON_UINT64_C2(0xcc20ce9b, 0xd35c78a5), + RAPIDJSON_UINT64_C2(0x98165af3, 0x7b2153df), RAPIDJSON_UINT64_C2(0xe2a0b5dc, 0x971f303a), + RAPIDJSON_UINT64_C2(0xa8d9d153, 0x5ce3b396), RAPIDJSON_UINT64_C2(0xfb9b7cd9, 0xa4a7443c), + RAPIDJSON_UINT64_C2(0xbb764c4c, 0xa7a44410), RAPIDJSON_UINT64_C2(0x8bab8eef, 0xb6409c1a), + RAPIDJSON_UINT64_C2(0xd01fef10, 0xa657842c), RAPIDJSON_UINT64_C2(0x9b10a4e5, 0xe9913129), + RAPIDJSON_UINT64_C2(0xe7109bfb, 0xa19c0c9d), RAPIDJSON_UINT64_C2(0xac2820d9, 0x623bf429), + RAPIDJSON_UINT64_C2(0x80444b5e, 0x7aa7cf85), RAPIDJSON_UINT64_C2(0xbf21e440, 0x03acdd2d), + RAPIDJSON_UINT64_C2(0x8e679c2f, 0x5e44ff8f), RAPIDJSON_UINT64_C2(0xd433179d, 0x9c8cb841), + RAPIDJSON_UINT64_C2(0x9e19db92, 0xb4e31ba9), RAPIDJSON_UINT64_C2(0xeb96bf6e, 0xbadf77d9), + RAPIDJSON_UINT64_C2(0xaf87023b, 0x9bf0ee6b) + }; + static const int16_t kCachedPowers_E[] = { + -1220, -1193, -1166, -1140, -1113, -1087, -1060, -1034, -1007, -980, + -954, -927, -901, -874, -847, -821, -794, -768, -741, -715, + -688, -661, -635, -608, -582, -555, -529, -502, -475, -449, + -422, -396, -369, -343, -316, -289, -263, -236, -210, -183, + -157, -130, -103, -77, -50, -24, 3, 30, 56, 83, + 109, 136, 162, 189, 216, 242, 269, 295, 322, 348, + 375, 402, 428, 455, 481, 508, 534, 561, 588, 614, + 641, 667, 694, 720, 747, 774, 800, 827, 853, 880, + 907, 933, 960, 986, 1013, 1039, 1066 + }; + RAPIDJSON_ASSERT(index < 87); + return DiyFp(kCachedPowers_F[index], kCachedPowers_E[index]); +} + +inline DiyFp GetCachedPower(int e, int* K) { + + //int k = static_cast(ceil((-61 - e) * 0.30102999566398114)) + 374; + double dk = (-61 - e) * 0.30102999566398114 + 347; // dk must be positive, so can do ceiling in positive + int k = static_cast(dk); + if (dk - k > 0.0) + k++; + + unsigned index = static_cast((k >> 3) + 1); + *K = -(-348 + static_cast(index << 3)); // decimal exponent no need lookup table + + return GetCachedPowerByIndex(index); +} + +inline DiyFp GetCachedPower10(int exp, int *outExp) { + RAPIDJSON_ASSERT(exp >= -348); + unsigned index = static_cast(exp + 348) / 8u; + *outExp = -348 + static_cast(index) * 8; + return GetCachedPowerByIndex(index); +} + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +RAPIDJSON_DIAG_OFF(padded) +#endif + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_DIYFP_H_ diff --git a/include/rapidjson/internal/dtoa.h b/include/rapidjson/internal/dtoa.h new file mode 100644 index 0000000..cd45672 --- /dev/null +++ b/include/rapidjson/internal/dtoa.h @@ -0,0 +1,249 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +// This is a C++ header-only implementation of Grisu2 algorithm from the publication: +// Loitsch, Florian. "Printing floating-point numbers quickly and accurately with +// integers." ACM Sigplan Notices 45.6 (2010): 233-243. + +#ifndef RAPIDJSON_DTOA_ +#define RAPIDJSON_DTOA_ + +#include "itoa.h" // GetDigitsLut() +#include "diyfp.h" +#include "ieee754.h" + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +RAPIDJSON_DIAG_OFF(array-bounds) // some gcc versions generate wrong warnings https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59124 +#endif + +inline void GrisuRound(char* buffer, int len, uint64_t delta, uint64_t rest, uint64_t ten_kappa, uint64_t wp_w) { + while (rest < wp_w && delta - rest >= ten_kappa && + (rest + ten_kappa < wp_w || /// closer + wp_w - rest > rest + ten_kappa - wp_w)) { + buffer[len - 1]--; + rest += ten_kappa; + } +} + +inline int CountDecimalDigit32(uint32_t n) { + // Simple pure C++ implementation was faster than __builtin_clz version in this situation. + if (n < 10) return 1; + if (n < 100) return 2; + if (n < 1000) return 3; + if (n < 10000) return 4; + if (n < 100000) return 5; + if (n < 1000000) return 6; + if (n < 10000000) return 7; + if (n < 100000000) return 8; + // Will not reach 10 digits in DigitGen() + //if (n < 1000000000) return 9; + //return 10; + return 9; +} + +inline void DigitGen(const DiyFp& W, const DiyFp& Mp, uint64_t delta, char* buffer, int* len, int* K) { + static const uint64_t kPow10[] = { 1ULL, 10ULL, 100ULL, 1000ULL, 10000ULL, 100000ULL, 1000000ULL, 10000000ULL, 100000000ULL, + 1000000000ULL, 10000000000ULL, 100000000000ULL, 1000000000000ULL, + 10000000000000ULL, 100000000000000ULL, 1000000000000000ULL, + 10000000000000000ULL, 100000000000000000ULL, 1000000000000000000ULL, + 10000000000000000000ULL }; + const DiyFp one(uint64_t(1) << -Mp.e, Mp.e); + const DiyFp wp_w = Mp - W; + uint32_t p1 = static_cast(Mp.f >> -one.e); + uint64_t p2 = Mp.f & (one.f - 1); + int kappa = CountDecimalDigit32(p1); // kappa in [0, 9] + *len = 0; + + while (kappa > 0) { + uint32_t d = 0; + switch (kappa) { + case 9: d = p1 / 100000000; p1 %= 100000000; break; + case 8: d = p1 / 10000000; p1 %= 10000000; break; + case 7: d = p1 / 1000000; p1 %= 1000000; break; + case 6: d = p1 / 100000; p1 %= 100000; break; + case 5: d = p1 / 10000; p1 %= 10000; break; + case 4: d = p1 / 1000; p1 %= 1000; break; + case 3: d = p1 / 100; p1 %= 100; break; + case 2: d = p1 / 10; p1 %= 10; break; + case 1: d = p1; p1 = 0; break; + default:; + } + if (d || *len) + buffer[(*len)++] = static_cast('0' + static_cast(d)); + kappa--; + uint64_t tmp = (static_cast(p1) << -one.e) + p2; + if (tmp <= delta) { + *K += kappa; + GrisuRound(buffer, *len, delta, tmp, kPow10[kappa] << -one.e, wp_w.f); + return; + } + } + + // kappa = 0 + for (;;) { + p2 *= 10; + delta *= 10; + char d = static_cast(p2 >> -one.e); + if (d || *len) + buffer[(*len)++] = static_cast('0' + d); + p2 &= one.f - 1; + kappa--; + if (p2 < delta) { + *K += kappa; + int index = -kappa; + GrisuRound(buffer, *len, delta, p2, one.f, wp_w.f * (index < 20 ? kPow10[index] : 0)); + return; + } + } +} + +inline void Grisu2(double value, char* buffer, int* length, int* K) { + const DiyFp v(value); + DiyFp w_m, w_p; + v.NormalizedBoundaries(&w_m, &w_p); + + const DiyFp c_mk = GetCachedPower(w_p.e, K); + const DiyFp W = v.Normalize() * c_mk; + DiyFp Wp = w_p * c_mk; + DiyFp Wm = w_m * c_mk; + Wm.f++; + Wp.f--; + DigitGen(W, Wp, Wp.f - Wm.f, buffer, length, K); +} + +inline char* WriteExponent(int K, char* buffer) { + if (K < 0) { + *buffer++ = '-'; + K = -K; + } + + if (K >= 100) { + *buffer++ = static_cast('0' + static_cast(K / 100)); + K %= 100; + const char* d = GetDigitsLut() + K * 2; + *buffer++ = d[0]; + *buffer++ = d[1]; + } + else if (K >= 10) { + const char* d = GetDigitsLut() + K * 2; + *buffer++ = d[0]; + *buffer++ = d[1]; + } + else + *buffer++ = static_cast('0' + static_cast(K)); + + return buffer; +} + +inline char* Prettify(char* buffer, int length, int k, int maxDecimalPlaces) { + const int kk = length + k; // 10^(kk-1) <= v < 10^kk + + if (0 <= k && kk <= 21) { + // 1234e7 -> 12340000000 + for (int i = length; i < kk; i++) + buffer[i] = '0'; + buffer[kk] = '.'; + buffer[kk + 1] = '0'; + return &buffer[kk + 2]; + } + else if (0 < kk && kk <= 21) { + // 1234e-2 -> 12.34 + std::memmove(&buffer[kk + 1], &buffer[kk], static_cast(length - kk)); + buffer[kk] = '.'; + if (0 > k + maxDecimalPlaces) { + // When maxDecimalPlaces = 2, 1.2345 -> 1.23, 1.102 -> 1.1 + // Remove extra trailing zeros (at least one) after truncation. + for (int i = kk + maxDecimalPlaces; i > kk + 1; i--) + if (buffer[i] != '0') + return &buffer[i + 1]; + return &buffer[kk + 2]; // Reserve one zero + } + else + return &buffer[length + 1]; + } + else if (-6 < kk && kk <= 0) { + // 1234e-6 -> 0.001234 + const int offset = 2 - kk; + std::memmove(&buffer[offset], &buffer[0], static_cast(length)); + buffer[0] = '0'; + buffer[1] = '.'; + for (int i = 2; i < offset; i++) + buffer[i] = '0'; + if (length - kk > maxDecimalPlaces) { + // When maxDecimalPlaces = 2, 0.123 -> 0.12, 0.102 -> 0.1 + // Remove extra trailing zeros (at least one) after truncation. + for (int i = maxDecimalPlaces + 1; i > 2; i--) + if (buffer[i] != '0') + return &buffer[i + 1]; + return &buffer[3]; // Reserve one zero + } + else + return &buffer[length + offset]; + } + else if (kk < -maxDecimalPlaces) { + // Truncate to zero + buffer[0] = '0'; + buffer[1] = '.'; + buffer[2] = '0'; + return &buffer[3]; + } + else if (length == 1) { + // 1e30 + buffer[1] = 'e'; + return WriteExponent(kk - 1, &buffer[2]); + } + else { + // 1234e30 -> 1.234e33 + std::memmove(&buffer[2], &buffer[1], static_cast(length - 1)); + buffer[1] = '.'; + buffer[length + 1] = 'e'; + return WriteExponent(kk - 1, &buffer[0 + length + 2]); + } +} + +inline char* dtoa(double value, char* buffer, int maxDecimalPlaces = 324) { + RAPIDJSON_ASSERT(maxDecimalPlaces >= 1); + Double d(value); + if (d.IsZero()) { + if (d.Sign()) + *buffer++ = '-'; // -0.0, Issue #289 + buffer[0] = '0'; + buffer[1] = '.'; + buffer[2] = '0'; + return &buffer[3]; + } + else { + if (value < 0) { + *buffer++ = '-'; + value = -value; + } + int length, K; + Grisu2(value, buffer, &length, &K); + return Prettify(buffer, length, K, maxDecimalPlaces); + } +} + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_DTOA_ diff --git a/include/rapidjson/internal/ieee754.h b/include/rapidjson/internal/ieee754.h new file mode 100644 index 0000000..68c9e96 --- /dev/null +++ b/include/rapidjson/internal/ieee754.h @@ -0,0 +1,78 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_IEEE754_ +#define RAPIDJSON_IEEE754_ + +#include "../rapidjson.h" + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +class Double { +public: + Double() {} + Double(double d) : d_(d) {} + Double(uint64_t u) : u_(u) {} + + double Value() const { return d_; } + uint64_t Uint64Value() const { return u_; } + + double NextPositiveDouble() const { + RAPIDJSON_ASSERT(!Sign()); + return Double(u_ + 1).Value(); + } + + bool Sign() const { return (u_ & kSignMask) != 0; } + uint64_t Significand() const { return u_ & kSignificandMask; } + int Exponent() const { return static_cast(((u_ & kExponentMask) >> kSignificandSize) - kExponentBias); } + + bool IsNan() const { return (u_ & kExponentMask) == kExponentMask && Significand() != 0; } + bool IsInf() const { return (u_ & kExponentMask) == kExponentMask && Significand() == 0; } + bool IsNanOrInf() const { return (u_ & kExponentMask) == kExponentMask; } + bool IsNormal() const { return (u_ & kExponentMask) != 0 || Significand() == 0; } + bool IsZero() const { return (u_ & (kExponentMask | kSignificandMask)) == 0; } + + uint64_t IntegerSignificand() const { return IsNormal() ? Significand() | kHiddenBit : Significand(); } + int IntegerExponent() const { return (IsNormal() ? Exponent() : kDenormalExponent) - kSignificandSize; } + uint64_t ToBias() const { return (u_ & kSignMask) ? ~u_ + 1 : u_ | kSignMask; } + + static int EffectiveSignificandSize(int order) { + if (order >= -1021) + return 53; + else if (order <= -1074) + return 0; + else + return order + 1074; + } + +private: + static const int kSignificandSize = 52; + static const int kExponentBias = 0x3FF; + static const int kDenormalExponent = 1 - kExponentBias; + static const uint64_t kSignMask = RAPIDJSON_UINT64_C2(0x80000000, 0x00000000); + static const uint64_t kExponentMask = RAPIDJSON_UINT64_C2(0x7FF00000, 0x00000000); + static const uint64_t kSignificandMask = RAPIDJSON_UINT64_C2(0x000FFFFF, 0xFFFFFFFF); + static const uint64_t kHiddenBit = RAPIDJSON_UINT64_C2(0x00100000, 0x00000000); + + union { + double d_; + uint64_t u_; + }; +}; + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_IEEE754_ diff --git a/include/rapidjson/internal/itoa.h b/include/rapidjson/internal/itoa.h new file mode 100644 index 0000000..9fe8c93 --- /dev/null +++ b/include/rapidjson/internal/itoa.h @@ -0,0 +1,308 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ITOA_ +#define RAPIDJSON_ITOA_ + +#include "../rapidjson.h" + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +inline const char* GetDigitsLut() { + static const char cDigitsLut[200] = { + '0','0','0','1','0','2','0','3','0','4','0','5','0','6','0','7','0','8','0','9', + '1','0','1','1','1','2','1','3','1','4','1','5','1','6','1','7','1','8','1','9', + '2','0','2','1','2','2','2','3','2','4','2','5','2','6','2','7','2','8','2','9', + '3','0','3','1','3','2','3','3','3','4','3','5','3','6','3','7','3','8','3','9', + '4','0','4','1','4','2','4','3','4','4','4','5','4','6','4','7','4','8','4','9', + '5','0','5','1','5','2','5','3','5','4','5','5','5','6','5','7','5','8','5','9', + '6','0','6','1','6','2','6','3','6','4','6','5','6','6','6','7','6','8','6','9', + '7','0','7','1','7','2','7','3','7','4','7','5','7','6','7','7','7','8','7','9', + '8','0','8','1','8','2','8','3','8','4','8','5','8','6','8','7','8','8','8','9', + '9','0','9','1','9','2','9','3','9','4','9','5','9','6','9','7','9','8','9','9' + }; + return cDigitsLut; +} + +inline char* u32toa(uint32_t value, char* buffer) { + RAPIDJSON_ASSERT(buffer != 0); + + const char* cDigitsLut = GetDigitsLut(); + + if (value < 10000) { + const uint32_t d1 = (value / 100) << 1; + const uint32_t d2 = (value % 100) << 1; + + if (value >= 1000) + *buffer++ = cDigitsLut[d1]; + if (value >= 100) + *buffer++ = cDigitsLut[d1 + 1]; + if (value >= 10) + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + } + else if (value < 100000000) { + // value = bbbbcccc + const uint32_t b = value / 10000; + const uint32_t c = value % 10000; + + const uint32_t d1 = (b / 100) << 1; + const uint32_t d2 = (b % 100) << 1; + + const uint32_t d3 = (c / 100) << 1; + const uint32_t d4 = (c % 100) << 1; + + if (value >= 10000000) + *buffer++ = cDigitsLut[d1]; + if (value >= 1000000) + *buffer++ = cDigitsLut[d1 + 1]; + if (value >= 100000) + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + + *buffer++ = cDigitsLut[d3]; + *buffer++ = cDigitsLut[d3 + 1]; + *buffer++ = cDigitsLut[d4]; + *buffer++ = cDigitsLut[d4 + 1]; + } + else { + // value = aabbbbcccc in decimal + + const uint32_t a = value / 100000000; // 1 to 42 + value %= 100000000; + + if (a >= 10) { + const unsigned i = a << 1; + *buffer++ = cDigitsLut[i]; + *buffer++ = cDigitsLut[i + 1]; + } + else + *buffer++ = static_cast('0' + static_cast(a)); + + const uint32_t b = value / 10000; // 0 to 9999 + const uint32_t c = value % 10000; // 0 to 9999 + + const uint32_t d1 = (b / 100) << 1; + const uint32_t d2 = (b % 100) << 1; + + const uint32_t d3 = (c / 100) << 1; + const uint32_t d4 = (c % 100) << 1; + + *buffer++ = cDigitsLut[d1]; + *buffer++ = cDigitsLut[d1 + 1]; + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + *buffer++ = cDigitsLut[d3]; + *buffer++ = cDigitsLut[d3 + 1]; + *buffer++ = cDigitsLut[d4]; + *buffer++ = cDigitsLut[d4 + 1]; + } + return buffer; +} + +inline char* i32toa(int32_t value, char* buffer) { + RAPIDJSON_ASSERT(buffer != 0); + uint32_t u = static_cast(value); + if (value < 0) { + *buffer++ = '-'; + u = ~u + 1; + } + + return u32toa(u, buffer); +} + +inline char* u64toa(uint64_t value, char* buffer) { + RAPIDJSON_ASSERT(buffer != 0); + const char* cDigitsLut = GetDigitsLut(); + const uint64_t kTen8 = 100000000; + const uint64_t kTen9 = kTen8 * 10; + const uint64_t kTen10 = kTen8 * 100; + const uint64_t kTen11 = kTen8 * 1000; + const uint64_t kTen12 = kTen8 * 10000; + const uint64_t kTen13 = kTen8 * 100000; + const uint64_t kTen14 = kTen8 * 1000000; + const uint64_t kTen15 = kTen8 * 10000000; + const uint64_t kTen16 = kTen8 * kTen8; + + if (value < kTen8) { + uint32_t v = static_cast(value); + if (v < 10000) { + const uint32_t d1 = (v / 100) << 1; + const uint32_t d2 = (v % 100) << 1; + + if (v >= 1000) + *buffer++ = cDigitsLut[d1]; + if (v >= 100) + *buffer++ = cDigitsLut[d1 + 1]; + if (v >= 10) + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + } + else { + // value = bbbbcccc + const uint32_t b = v / 10000; + const uint32_t c = v % 10000; + + const uint32_t d1 = (b / 100) << 1; + const uint32_t d2 = (b % 100) << 1; + + const uint32_t d3 = (c / 100) << 1; + const uint32_t d4 = (c % 100) << 1; + + if (value >= 10000000) + *buffer++ = cDigitsLut[d1]; + if (value >= 1000000) + *buffer++ = cDigitsLut[d1 + 1]; + if (value >= 100000) + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + + *buffer++ = cDigitsLut[d3]; + *buffer++ = cDigitsLut[d3 + 1]; + *buffer++ = cDigitsLut[d4]; + *buffer++ = cDigitsLut[d4 + 1]; + } + } + else if (value < kTen16) { + const uint32_t v0 = static_cast(value / kTen8); + const uint32_t v1 = static_cast(value % kTen8); + + const uint32_t b0 = v0 / 10000; + const uint32_t c0 = v0 % 10000; + + const uint32_t d1 = (b0 / 100) << 1; + const uint32_t d2 = (b0 % 100) << 1; + + const uint32_t d3 = (c0 / 100) << 1; + const uint32_t d4 = (c0 % 100) << 1; + + const uint32_t b1 = v1 / 10000; + const uint32_t c1 = v1 % 10000; + + const uint32_t d5 = (b1 / 100) << 1; + const uint32_t d6 = (b1 % 100) << 1; + + const uint32_t d7 = (c1 / 100) << 1; + const uint32_t d8 = (c1 % 100) << 1; + + if (value >= kTen15) + *buffer++ = cDigitsLut[d1]; + if (value >= kTen14) + *buffer++ = cDigitsLut[d1 + 1]; + if (value >= kTen13) + *buffer++ = cDigitsLut[d2]; + if (value >= kTen12) + *buffer++ = cDigitsLut[d2 + 1]; + if (value >= kTen11) + *buffer++ = cDigitsLut[d3]; + if (value >= kTen10) + *buffer++ = cDigitsLut[d3 + 1]; + if (value >= kTen9) + *buffer++ = cDigitsLut[d4]; + + *buffer++ = cDigitsLut[d4 + 1]; + *buffer++ = cDigitsLut[d5]; + *buffer++ = cDigitsLut[d5 + 1]; + *buffer++ = cDigitsLut[d6]; + *buffer++ = cDigitsLut[d6 + 1]; + *buffer++ = cDigitsLut[d7]; + *buffer++ = cDigitsLut[d7 + 1]; + *buffer++ = cDigitsLut[d8]; + *buffer++ = cDigitsLut[d8 + 1]; + } + else { + const uint32_t a = static_cast(value / kTen16); // 1 to 1844 + value %= kTen16; + + if (a < 10) + *buffer++ = static_cast('0' + static_cast(a)); + else if (a < 100) { + const uint32_t i = a << 1; + *buffer++ = cDigitsLut[i]; + *buffer++ = cDigitsLut[i + 1]; + } + else if (a < 1000) { + *buffer++ = static_cast('0' + static_cast(a / 100)); + + const uint32_t i = (a % 100) << 1; + *buffer++ = cDigitsLut[i]; + *buffer++ = cDigitsLut[i + 1]; + } + else { + const uint32_t i = (a / 100) << 1; + const uint32_t j = (a % 100) << 1; + *buffer++ = cDigitsLut[i]; + *buffer++ = cDigitsLut[i + 1]; + *buffer++ = cDigitsLut[j]; + *buffer++ = cDigitsLut[j + 1]; + } + + const uint32_t v0 = static_cast(value / kTen8); + const uint32_t v1 = static_cast(value % kTen8); + + const uint32_t b0 = v0 / 10000; + const uint32_t c0 = v0 % 10000; + + const uint32_t d1 = (b0 / 100) << 1; + const uint32_t d2 = (b0 % 100) << 1; + + const uint32_t d3 = (c0 / 100) << 1; + const uint32_t d4 = (c0 % 100) << 1; + + const uint32_t b1 = v1 / 10000; + const uint32_t c1 = v1 % 10000; + + const uint32_t d5 = (b1 / 100) << 1; + const uint32_t d6 = (b1 % 100) << 1; + + const uint32_t d7 = (c1 / 100) << 1; + const uint32_t d8 = (c1 % 100) << 1; + + *buffer++ = cDigitsLut[d1]; + *buffer++ = cDigitsLut[d1 + 1]; + *buffer++ = cDigitsLut[d2]; + *buffer++ = cDigitsLut[d2 + 1]; + *buffer++ = cDigitsLut[d3]; + *buffer++ = cDigitsLut[d3 + 1]; + *buffer++ = cDigitsLut[d4]; + *buffer++ = cDigitsLut[d4 + 1]; + *buffer++ = cDigitsLut[d5]; + *buffer++ = cDigitsLut[d5 + 1]; + *buffer++ = cDigitsLut[d6]; + *buffer++ = cDigitsLut[d6 + 1]; + *buffer++ = cDigitsLut[d7]; + *buffer++ = cDigitsLut[d7 + 1]; + *buffer++ = cDigitsLut[d8]; + *buffer++ = cDigitsLut[d8 + 1]; + } + + return buffer; +} + +inline char* i64toa(int64_t value, char* buffer) { + RAPIDJSON_ASSERT(buffer != 0); + uint64_t u = static_cast(value); + if (value < 0) { + *buffer++ = '-'; + u = ~u + 1; + } + + return u64toa(u, buffer); +} + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_ITOA_ diff --git a/include/rapidjson/internal/meta.h b/include/rapidjson/internal/meta.h new file mode 100644 index 0000000..27092dc --- /dev/null +++ b/include/rapidjson/internal/meta.h @@ -0,0 +1,186 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_INTERNAL_META_H_ +#define RAPIDJSON_INTERNAL_META_H_ + +#include "../rapidjson.h" + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#if defined(_MSC_VER) && !defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(6334) +#endif + +#if RAPIDJSON_HAS_CXX11_TYPETRAITS +#include +#endif + +//@cond RAPIDJSON_INTERNAL +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +// Helper to wrap/convert arbitrary types to void, useful for arbitrary type matching +template struct Void { typedef void Type; }; + +/////////////////////////////////////////////////////////////////////////////// +// BoolType, TrueType, FalseType +// +template struct BoolType { + static const bool Value = Cond; + typedef BoolType Type; +}; +typedef BoolType TrueType; +typedef BoolType FalseType; + + +/////////////////////////////////////////////////////////////////////////////// +// SelectIf, BoolExpr, NotExpr, AndExpr, OrExpr +// + +template struct SelectIfImpl { template struct Apply { typedef T1 Type; }; }; +template <> struct SelectIfImpl { template struct Apply { typedef T2 Type; }; }; +template struct SelectIfCond : SelectIfImpl::template Apply {}; +template struct SelectIf : SelectIfCond {}; + +template struct AndExprCond : FalseType {}; +template <> struct AndExprCond : TrueType {}; +template struct OrExprCond : TrueType {}; +template <> struct OrExprCond : FalseType {}; + +template struct BoolExpr : SelectIf::Type {}; +template struct NotExpr : SelectIf::Type {}; +template struct AndExpr : AndExprCond::Type {}; +template struct OrExpr : OrExprCond::Type {}; + + +/////////////////////////////////////////////////////////////////////////////// +// AddConst, MaybeAddConst, RemoveConst +template struct AddConst { typedef const T Type; }; +template struct MaybeAddConst : SelectIfCond {}; +template struct RemoveConst { typedef T Type; }; +template struct RemoveConst { typedef T Type; }; + + +/////////////////////////////////////////////////////////////////////////////// +// IsSame, IsConst, IsMoreConst, IsPointer +// +template struct IsSame : FalseType {}; +template struct IsSame : TrueType {}; + +template struct IsConst : FalseType {}; +template struct IsConst : TrueType {}; + +template +struct IsMoreConst + : AndExpr::Type, typename RemoveConst::Type>, + BoolType::Value >= IsConst::Value> >::Type {}; + +template struct IsPointer : FalseType {}; +template struct IsPointer : TrueType {}; + +/////////////////////////////////////////////////////////////////////////////// +// IsBaseOf +// +#if RAPIDJSON_HAS_CXX11_TYPETRAITS + +template struct IsBaseOf + : BoolType< ::std::is_base_of::value> {}; + +#else // simplified version adopted from Boost + +template struct IsBaseOfImpl { + RAPIDJSON_STATIC_ASSERT(sizeof(B) != 0); + RAPIDJSON_STATIC_ASSERT(sizeof(D) != 0); + + typedef char (&Yes)[1]; + typedef char (&No) [2]; + + template + static Yes Check(const D*, T); + static No Check(const B*, int); + + struct Host { + operator const B*() const; + operator const D*(); + }; + + enum { Value = (sizeof(Check(Host(), 0)) == sizeof(Yes)) }; +}; + +template struct IsBaseOf + : OrExpr, BoolExpr > >::Type {}; + +#endif // RAPIDJSON_HAS_CXX11_TYPETRAITS + + +////////////////////////////////////////////////////////////////////////// +// EnableIf / DisableIf +// +template struct EnableIfCond { typedef T Type; }; +template struct EnableIfCond { /* empty */ }; + +template struct DisableIfCond { typedef T Type; }; +template struct DisableIfCond { /* empty */ }; + +template +struct EnableIf : EnableIfCond {}; + +template +struct DisableIf : DisableIfCond {}; + +// SFINAE helpers +struct SfinaeTag {}; +template struct RemoveSfinaeTag; +template struct RemoveSfinaeTag { typedef T Type; }; + +#define RAPIDJSON_REMOVEFPTR_(type) \ + typename ::RAPIDJSON_NAMESPACE::internal::RemoveSfinaeTag \ + < ::RAPIDJSON_NAMESPACE::internal::SfinaeTag&(*) type>::Type + +#define RAPIDJSON_ENABLEIF(cond) \ + typename ::RAPIDJSON_NAMESPACE::internal::EnableIf \ + ::Type * = NULL + +#define RAPIDJSON_DISABLEIF(cond) \ + typename ::RAPIDJSON_NAMESPACE::internal::DisableIf \ + ::Type * = NULL + +#define RAPIDJSON_ENABLEIF_RETURN(cond,returntype) \ + typename ::RAPIDJSON_NAMESPACE::internal::EnableIf \ + ::Type + +#define RAPIDJSON_DISABLEIF_RETURN(cond,returntype) \ + typename ::RAPIDJSON_NAMESPACE::internal::DisableIf \ + ::Type + +} // namespace internal +RAPIDJSON_NAMESPACE_END +//@endcond + +#if defined(_MSC_VER) && !defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_INTERNAL_META_H_ diff --git a/include/rapidjson/internal/pow10.h b/include/rapidjson/internal/pow10.h new file mode 100644 index 0000000..eae1a43 --- /dev/null +++ b/include/rapidjson/internal/pow10.h @@ -0,0 +1,55 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_POW10_ +#define RAPIDJSON_POW10_ + +#include "../rapidjson.h" + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +//! Computes integer powers of 10 in double (10.0^n). +/*! This function uses lookup table for fast and accurate results. + \param n non-negative exponent. Must <= 308. + \return 10.0^n +*/ +inline double Pow10(int n) { + static const double e[] = { // 1e-0...1e308: 309 * 8 bytes = 2472 bytes + 1e+0, + 1e+1, 1e+2, 1e+3, 1e+4, 1e+5, 1e+6, 1e+7, 1e+8, 1e+9, 1e+10, 1e+11, 1e+12, 1e+13, 1e+14, 1e+15, 1e+16, 1e+17, 1e+18, 1e+19, 1e+20, + 1e+21, 1e+22, 1e+23, 1e+24, 1e+25, 1e+26, 1e+27, 1e+28, 1e+29, 1e+30, 1e+31, 1e+32, 1e+33, 1e+34, 1e+35, 1e+36, 1e+37, 1e+38, 1e+39, 1e+40, + 1e+41, 1e+42, 1e+43, 1e+44, 1e+45, 1e+46, 1e+47, 1e+48, 1e+49, 1e+50, 1e+51, 1e+52, 1e+53, 1e+54, 1e+55, 1e+56, 1e+57, 1e+58, 1e+59, 1e+60, + 1e+61, 1e+62, 1e+63, 1e+64, 1e+65, 1e+66, 1e+67, 1e+68, 1e+69, 1e+70, 1e+71, 1e+72, 1e+73, 1e+74, 1e+75, 1e+76, 1e+77, 1e+78, 1e+79, 1e+80, + 1e+81, 1e+82, 1e+83, 1e+84, 1e+85, 1e+86, 1e+87, 1e+88, 1e+89, 1e+90, 1e+91, 1e+92, 1e+93, 1e+94, 1e+95, 1e+96, 1e+97, 1e+98, 1e+99, 1e+100, + 1e+101,1e+102,1e+103,1e+104,1e+105,1e+106,1e+107,1e+108,1e+109,1e+110,1e+111,1e+112,1e+113,1e+114,1e+115,1e+116,1e+117,1e+118,1e+119,1e+120, + 1e+121,1e+122,1e+123,1e+124,1e+125,1e+126,1e+127,1e+128,1e+129,1e+130,1e+131,1e+132,1e+133,1e+134,1e+135,1e+136,1e+137,1e+138,1e+139,1e+140, + 1e+141,1e+142,1e+143,1e+144,1e+145,1e+146,1e+147,1e+148,1e+149,1e+150,1e+151,1e+152,1e+153,1e+154,1e+155,1e+156,1e+157,1e+158,1e+159,1e+160, + 1e+161,1e+162,1e+163,1e+164,1e+165,1e+166,1e+167,1e+168,1e+169,1e+170,1e+171,1e+172,1e+173,1e+174,1e+175,1e+176,1e+177,1e+178,1e+179,1e+180, + 1e+181,1e+182,1e+183,1e+184,1e+185,1e+186,1e+187,1e+188,1e+189,1e+190,1e+191,1e+192,1e+193,1e+194,1e+195,1e+196,1e+197,1e+198,1e+199,1e+200, + 1e+201,1e+202,1e+203,1e+204,1e+205,1e+206,1e+207,1e+208,1e+209,1e+210,1e+211,1e+212,1e+213,1e+214,1e+215,1e+216,1e+217,1e+218,1e+219,1e+220, + 1e+221,1e+222,1e+223,1e+224,1e+225,1e+226,1e+227,1e+228,1e+229,1e+230,1e+231,1e+232,1e+233,1e+234,1e+235,1e+236,1e+237,1e+238,1e+239,1e+240, + 1e+241,1e+242,1e+243,1e+244,1e+245,1e+246,1e+247,1e+248,1e+249,1e+250,1e+251,1e+252,1e+253,1e+254,1e+255,1e+256,1e+257,1e+258,1e+259,1e+260, + 1e+261,1e+262,1e+263,1e+264,1e+265,1e+266,1e+267,1e+268,1e+269,1e+270,1e+271,1e+272,1e+273,1e+274,1e+275,1e+276,1e+277,1e+278,1e+279,1e+280, + 1e+281,1e+282,1e+283,1e+284,1e+285,1e+286,1e+287,1e+288,1e+289,1e+290,1e+291,1e+292,1e+293,1e+294,1e+295,1e+296,1e+297,1e+298,1e+299,1e+300, + 1e+301,1e+302,1e+303,1e+304,1e+305,1e+306,1e+307,1e+308 + }; + RAPIDJSON_ASSERT(n >= 0 && n <= 308); + return e[n]; +} + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_POW10_ diff --git a/include/rapidjson/internal/regex.h b/include/rapidjson/internal/regex.h new file mode 100644 index 0000000..7740dcd --- /dev/null +++ b/include/rapidjson/internal/regex.h @@ -0,0 +1,739 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_INTERNAL_REGEX_H_ +#define RAPIDJSON_INTERNAL_REGEX_H_ + +#include "../allocators.h" +#include "../stream.h" +#include "stack.h" + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +RAPIDJSON_DIAG_OFF(switch-enum) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#ifndef RAPIDJSON_REGEX_VERBOSE +#define RAPIDJSON_REGEX_VERBOSE 0 +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +/////////////////////////////////////////////////////////////////////////////// +// DecodedStream + +template +class DecodedStream { +public: + DecodedStream(SourceStream& ss) : ss_(ss), codepoint_() { Decode(); } + unsigned Peek() { return codepoint_; } + unsigned Take() { + unsigned c = codepoint_; + if (c) // No further decoding when '\0' + Decode(); + return c; + } + +private: + void Decode() { + if (!Encoding::Decode(ss_, &codepoint_)) + codepoint_ = 0; + } + + SourceStream& ss_; + unsigned codepoint_; +}; + +/////////////////////////////////////////////////////////////////////////////// +// GenericRegex + +static const SizeType kRegexInvalidState = ~SizeType(0); //!< Represents an invalid index in GenericRegex::State::out, out1 +static const SizeType kRegexInvalidRange = ~SizeType(0); + +template +class GenericRegexSearch; + +//! Regular expression engine with subset of ECMAscript grammar. +/*! + Supported regular expression syntax: + - \c ab Concatenation + - \c a|b Alternation + - \c a? Zero or one + - \c a* Zero or more + - \c a+ One or more + - \c a{3} Exactly 3 times + - \c a{3,} At least 3 times + - \c a{3,5} 3 to 5 times + - \c (ab) Grouping + - \c ^a At the beginning + - \c a$ At the end + - \c . Any character + - \c [abc] Character classes + - \c [a-c] Character class range + - \c [a-z0-9_] Character class combination + - \c [^abc] Negated character classes + - \c [^a-c] Negated character class range + - \c [\b] Backspace (U+0008) + - \c \\| \\\\ ... Escape characters + - \c \\f Form feed (U+000C) + - \c \\n Line feed (U+000A) + - \c \\r Carriage return (U+000D) + - \c \\t Tab (U+0009) + - \c \\v Vertical tab (U+000B) + + \note This is a Thompson NFA engine, implemented with reference to + Cox, Russ. "Regular Expression Matching Can Be Simple And Fast (but is slow in Java, Perl, PHP, Python, Ruby,...).", + https://swtch.com/~rsc/regexp/regexp1.html +*/ +template +class GenericRegex { +public: + typedef Encoding EncodingType; + typedef typename Encoding::Ch Ch; + template friend class GenericRegexSearch; + + GenericRegex(const Ch* source, Allocator* allocator = 0) : + ownAllocator_(allocator ? 0 : RAPIDJSON_NEW(Allocator)()), allocator_(allocator ? allocator : ownAllocator_), + states_(allocator_, 256), ranges_(allocator_, 256), root_(kRegexInvalidState), stateCount_(), rangeCount_(), + anchorBegin_(), anchorEnd_() + { + GenericStringStream ss(source); + DecodedStream, Encoding> ds(ss); + Parse(ds); + } + + ~GenericRegex() + { + RAPIDJSON_DELETE(ownAllocator_); + } + + bool IsValid() const { + return root_ != kRegexInvalidState; + } + +private: + enum Operator { + kZeroOrOne, + kZeroOrMore, + kOneOrMore, + kConcatenation, + kAlternation, + kLeftParenthesis + }; + + static const unsigned kAnyCharacterClass = 0xFFFFFFFF; //!< For '.' + static const unsigned kRangeCharacterClass = 0xFFFFFFFE; + static const unsigned kRangeNegationFlag = 0x80000000; + + struct Range { + unsigned start; // + unsigned end; + SizeType next; + }; + + struct State { + SizeType out; //!< Equals to kInvalid for matching state + SizeType out1; //!< Equals to non-kInvalid for split + SizeType rangeStart; + unsigned codepoint; + }; + + struct Frag { + Frag(SizeType s, SizeType o, SizeType m) : start(s), out(o), minIndex(m) {} + SizeType start; + SizeType out; //!< link-list of all output states + SizeType minIndex; + }; + + State& GetState(SizeType index) { + RAPIDJSON_ASSERT(index < stateCount_); + return states_.template Bottom()[index]; + } + + const State& GetState(SizeType index) const { + RAPIDJSON_ASSERT(index < stateCount_); + return states_.template Bottom()[index]; + } + + Range& GetRange(SizeType index) { + RAPIDJSON_ASSERT(index < rangeCount_); + return ranges_.template Bottom()[index]; + } + + const Range& GetRange(SizeType index) const { + RAPIDJSON_ASSERT(index < rangeCount_); + return ranges_.template Bottom()[index]; + } + + template + void Parse(DecodedStream& ds) { + Stack operandStack(allocator_, 256); // Frag + Stack operatorStack(allocator_, 256); // Operator + Stack atomCountStack(allocator_, 256); // unsigned (Atom per parenthesis) + + *atomCountStack.template Push() = 0; + + unsigned codepoint; + while (ds.Peek() != 0) { + switch (codepoint = ds.Take()) { + case '^': + anchorBegin_ = true; + break; + + case '$': + anchorEnd_ = true; + break; + + case '|': + while (!operatorStack.Empty() && *operatorStack.template Top() < kAlternation) + if (!Eval(operandStack, *operatorStack.template Pop(1))) + return; + *operatorStack.template Push() = kAlternation; + *atomCountStack.template Top() = 0; + break; + + case '(': + *operatorStack.template Push() = kLeftParenthesis; + *atomCountStack.template Push() = 0; + break; + + case ')': + while (!operatorStack.Empty() && *operatorStack.template Top() != kLeftParenthesis) + if (!Eval(operandStack, *operatorStack.template Pop(1))) + return; + if (operatorStack.Empty()) + return; + operatorStack.template Pop(1); + atomCountStack.template Pop(1); + ImplicitConcatenation(atomCountStack, operatorStack); + break; + + case '?': + if (!Eval(operandStack, kZeroOrOne)) + return; + break; + + case '*': + if (!Eval(operandStack, kZeroOrMore)) + return; + break; + + case '+': + if (!Eval(operandStack, kOneOrMore)) + return; + break; + + case '{': + { + unsigned n, m; + if (!ParseUnsigned(ds, &n)) + return; + + if (ds.Peek() == ',') { + ds.Take(); + if (ds.Peek() == '}') + m = kInfinityQuantifier; + else if (!ParseUnsigned(ds, &m) || m < n) + return; + } + else + m = n; + + if (!EvalQuantifier(operandStack, n, m) || ds.Peek() != '}') + return; + ds.Take(); + } + break; + + case '.': + PushOperand(operandStack, kAnyCharacterClass); + ImplicitConcatenation(atomCountStack, operatorStack); + break; + + case '[': + { + SizeType range; + if (!ParseRange(ds, &range)) + return; + SizeType s = NewState(kRegexInvalidState, kRegexInvalidState, kRangeCharacterClass); + GetState(s).rangeStart = range; + *operandStack.template Push() = Frag(s, s, s); + } + ImplicitConcatenation(atomCountStack, operatorStack); + break; + + case '\\': // Escape character + if (!CharacterEscape(ds, &codepoint)) + return; // Unsupported escape character + // fall through to default + RAPIDJSON_DELIBERATE_FALLTHROUGH; + + default: // Pattern character + PushOperand(operandStack, codepoint); + ImplicitConcatenation(atomCountStack, operatorStack); + } + } + + while (!operatorStack.Empty()) + if (!Eval(operandStack, *operatorStack.template Pop(1))) + return; + + // Link the operand to matching state. + if (operandStack.GetSize() == sizeof(Frag)) { + Frag* e = operandStack.template Pop(1); + Patch(e->out, NewState(kRegexInvalidState, kRegexInvalidState, 0)); + root_ = e->start; + +#if RAPIDJSON_REGEX_VERBOSE + printf("root: %d\n", root_); + for (SizeType i = 0; i < stateCount_ ; i++) { + State& s = GetState(i); + printf("[%2d] out: %2d out1: %2d c: '%c'\n", i, s.out, s.out1, (char)s.codepoint); + } + printf("\n"); +#endif + } + } + + SizeType NewState(SizeType out, SizeType out1, unsigned codepoint) { + State* s = states_.template Push(); + s->out = out; + s->out1 = out1; + s->codepoint = codepoint; + s->rangeStart = kRegexInvalidRange; + return stateCount_++; + } + + void PushOperand(Stack& operandStack, unsigned codepoint) { + SizeType s = NewState(kRegexInvalidState, kRegexInvalidState, codepoint); + *operandStack.template Push() = Frag(s, s, s); + } + + void ImplicitConcatenation(Stack& atomCountStack, Stack& operatorStack) { + if (*atomCountStack.template Top()) + *operatorStack.template Push() = kConcatenation; + (*atomCountStack.template Top())++; + } + + SizeType Append(SizeType l1, SizeType l2) { + SizeType old = l1; + while (GetState(l1).out != kRegexInvalidState) + l1 = GetState(l1).out; + GetState(l1).out = l2; + return old; + } + + void Patch(SizeType l, SizeType s) { + for (SizeType next; l != kRegexInvalidState; l = next) { + next = GetState(l).out; + GetState(l).out = s; + } + } + + bool Eval(Stack& operandStack, Operator op) { + switch (op) { + case kConcatenation: + RAPIDJSON_ASSERT(operandStack.GetSize() >= sizeof(Frag) * 2); + { + Frag e2 = *operandStack.template Pop(1); + Frag e1 = *operandStack.template Pop(1); + Patch(e1.out, e2.start); + *operandStack.template Push() = Frag(e1.start, e2.out, Min(e1.minIndex, e2.minIndex)); + } + return true; + + case kAlternation: + if (operandStack.GetSize() >= sizeof(Frag) * 2) { + Frag e2 = *operandStack.template Pop(1); + Frag e1 = *operandStack.template Pop(1); + SizeType s = NewState(e1.start, e2.start, 0); + *operandStack.template Push() = Frag(s, Append(e1.out, e2.out), Min(e1.minIndex, e2.minIndex)); + return true; + } + return false; + + case kZeroOrOne: + if (operandStack.GetSize() >= sizeof(Frag)) { + Frag e = *operandStack.template Pop(1); + SizeType s = NewState(kRegexInvalidState, e.start, 0); + *operandStack.template Push() = Frag(s, Append(e.out, s), e.minIndex); + return true; + } + return false; + + case kZeroOrMore: + if (operandStack.GetSize() >= sizeof(Frag)) { + Frag e = *operandStack.template Pop(1); + SizeType s = NewState(kRegexInvalidState, e.start, 0); + Patch(e.out, s); + *operandStack.template Push() = Frag(s, s, e.minIndex); + return true; + } + return false; + + case kOneOrMore: + if (operandStack.GetSize() >= sizeof(Frag)) { + Frag e = *operandStack.template Pop(1); + SizeType s = NewState(kRegexInvalidState, e.start, 0); + Patch(e.out, s); + *operandStack.template Push() = Frag(e.start, s, e.minIndex); + return true; + } + return false; + + default: + // syntax error (e.g. unclosed kLeftParenthesis) + return false; + } + } + + bool EvalQuantifier(Stack& operandStack, unsigned n, unsigned m) { + RAPIDJSON_ASSERT(n <= m); + RAPIDJSON_ASSERT(operandStack.GetSize() >= sizeof(Frag)); + + if (n == 0) { + if (m == 0) // a{0} not support + return false; + else if (m == kInfinityQuantifier) + Eval(operandStack, kZeroOrMore); // a{0,} -> a* + else { + Eval(operandStack, kZeroOrOne); // a{0,5} -> a? + for (unsigned i = 0; i < m - 1; i++) + CloneTopOperand(operandStack); // a{0,5} -> a? a? a? a? a? + for (unsigned i = 0; i < m - 1; i++) + Eval(operandStack, kConcatenation); // a{0,5} -> a?a?a?a?a? + } + return true; + } + + for (unsigned i = 0; i < n - 1; i++) // a{3} -> a a a + CloneTopOperand(operandStack); + + if (m == kInfinityQuantifier) + Eval(operandStack, kOneOrMore); // a{3,} -> a a a+ + else if (m > n) { + CloneTopOperand(operandStack); // a{3,5} -> a a a a + Eval(operandStack, kZeroOrOne); // a{3,5} -> a a a a? + for (unsigned i = n; i < m - 1; i++) + CloneTopOperand(operandStack); // a{3,5} -> a a a a? a? + for (unsigned i = n; i < m; i++) + Eval(operandStack, kConcatenation); // a{3,5} -> a a aa?a? + } + + for (unsigned i = 0; i < n - 1; i++) + Eval(operandStack, kConcatenation); // a{3} -> aaa, a{3,} -> aaa+, a{3.5} -> aaaa?a? + + return true; + } + + static SizeType Min(SizeType a, SizeType b) { return a < b ? a : b; } + + void CloneTopOperand(Stack& operandStack) { + const Frag src = *operandStack.template Top(); // Copy constructor to prevent invalidation + SizeType count = stateCount_ - src.minIndex; // Assumes top operand contains states in [src->minIndex, stateCount_) + State* s = states_.template Push(count); + memcpy(s, &GetState(src.minIndex), count * sizeof(State)); + for (SizeType j = 0; j < count; j++) { + if (s[j].out != kRegexInvalidState) + s[j].out += count; + if (s[j].out1 != kRegexInvalidState) + s[j].out1 += count; + } + *operandStack.template Push() = Frag(src.start + count, src.out + count, src.minIndex + count); + stateCount_ += count; + } + + template + bool ParseUnsigned(DecodedStream& ds, unsigned* u) { + unsigned r = 0; + if (ds.Peek() < '0' || ds.Peek() > '9') + return false; + while (ds.Peek() >= '0' && ds.Peek() <= '9') { + if (r >= 429496729 && ds.Peek() > '5') // 2^32 - 1 = 4294967295 + return false; // overflow + r = r * 10 + (ds.Take() - '0'); + } + *u = r; + return true; + } + + template + bool ParseRange(DecodedStream& ds, SizeType* range) { + bool isBegin = true; + bool negate = false; + int step = 0; + SizeType start = kRegexInvalidRange; + SizeType current = kRegexInvalidRange; + unsigned codepoint; + while ((codepoint = ds.Take()) != 0) { + if (isBegin) { + isBegin = false; + if (codepoint == '^') { + negate = true; + continue; + } + } + + switch (codepoint) { + case ']': + if (start == kRegexInvalidRange) + return false; // Error: nothing inside [] + if (step == 2) { // Add trailing '-' + SizeType r = NewRange('-'); + RAPIDJSON_ASSERT(current != kRegexInvalidRange); + GetRange(current).next = r; + } + if (negate) + GetRange(start).start |= kRangeNegationFlag; + *range = start; + return true; + + case '\\': + if (ds.Peek() == 'b') { + ds.Take(); + codepoint = 0x0008; // Escape backspace character + } + else if (!CharacterEscape(ds, &codepoint)) + return false; + // fall through to default + RAPIDJSON_DELIBERATE_FALLTHROUGH; + + default: + switch (step) { + case 1: + if (codepoint == '-') { + step++; + break; + } + // fall through to step 0 for other characters + RAPIDJSON_DELIBERATE_FALLTHROUGH; + + case 0: + { + SizeType r = NewRange(codepoint); + if (current != kRegexInvalidRange) + GetRange(current).next = r; + if (start == kRegexInvalidRange) + start = r; + current = r; + } + step = 1; + break; + + default: + RAPIDJSON_ASSERT(step == 2); + GetRange(current).end = codepoint; + step = 0; + } + } + } + return false; + } + + SizeType NewRange(unsigned codepoint) { + Range* r = ranges_.template Push(); + r->start = r->end = codepoint; + r->next = kRegexInvalidRange; + return rangeCount_++; + } + + template + bool CharacterEscape(DecodedStream& ds, unsigned* escapedCodepoint) { + unsigned codepoint; + switch (codepoint = ds.Take()) { + case '^': + case '$': + case '|': + case '(': + case ')': + case '?': + case '*': + case '+': + case '.': + case '[': + case ']': + case '{': + case '}': + case '\\': + *escapedCodepoint = codepoint; return true; + case 'f': *escapedCodepoint = 0x000C; return true; + case 'n': *escapedCodepoint = 0x000A; return true; + case 'r': *escapedCodepoint = 0x000D; return true; + case 't': *escapedCodepoint = 0x0009; return true; + case 'v': *escapedCodepoint = 0x000B; return true; + default: + return false; // Unsupported escape character + } + } + + Allocator* ownAllocator_; + Allocator* allocator_; + Stack states_; + Stack ranges_; + SizeType root_; + SizeType stateCount_; + SizeType rangeCount_; + + static const unsigned kInfinityQuantifier = ~0u; + + // For SearchWithAnchoring() + bool anchorBegin_; + bool anchorEnd_; +}; + +template +class GenericRegexSearch { +public: + typedef typename RegexType::EncodingType Encoding; + typedef typename Encoding::Ch Ch; + + GenericRegexSearch(const RegexType& regex, Allocator* allocator = 0) : + regex_(regex), allocator_(allocator), ownAllocator_(0), + state0_(allocator, 0), state1_(allocator, 0), stateSet_() + { + RAPIDJSON_ASSERT(regex_.IsValid()); + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + stateSet_ = static_cast(allocator_->Malloc(GetStateSetSize())); + state0_.template Reserve(regex_.stateCount_); + state1_.template Reserve(regex_.stateCount_); + } + + ~GenericRegexSearch() { + Allocator::Free(stateSet_); + RAPIDJSON_DELETE(ownAllocator_); + } + + template + bool Match(InputStream& is) { + return SearchWithAnchoring(is, true, true); + } + + bool Match(const Ch* s) { + GenericStringStream is(s); + return Match(is); + } + + template + bool Search(InputStream& is) { + return SearchWithAnchoring(is, regex_.anchorBegin_, regex_.anchorEnd_); + } + + bool Search(const Ch* s) { + GenericStringStream is(s); + return Search(is); + } + +private: + typedef typename RegexType::State State; + typedef typename RegexType::Range Range; + + template + bool SearchWithAnchoring(InputStream& is, bool anchorBegin, bool anchorEnd) { + DecodedStream ds(is); + + state0_.Clear(); + Stack *current = &state0_, *next = &state1_; + const size_t stateSetSize = GetStateSetSize(); + std::memset(stateSet_, 0, stateSetSize); + + bool matched = AddState(*current, regex_.root_); + unsigned codepoint; + while (!current->Empty() && (codepoint = ds.Take()) != 0) { + std::memset(stateSet_, 0, stateSetSize); + next->Clear(); + matched = false; + for (const SizeType* s = current->template Bottom(); s != current->template End(); ++s) { + const State& sr = regex_.GetState(*s); + if (sr.codepoint == codepoint || + sr.codepoint == RegexType::kAnyCharacterClass || + (sr.codepoint == RegexType::kRangeCharacterClass && MatchRange(sr.rangeStart, codepoint))) + { + matched = AddState(*next, sr.out) || matched; + if (!anchorEnd && matched) + return true; + } + if (!anchorBegin) + AddState(*next, regex_.root_); + } + internal::Swap(current, next); + } + + return matched; + } + + size_t GetStateSetSize() const { + return (regex_.stateCount_ + 31) / 32 * 4; + } + + // Return whether the added states is a match state + bool AddState(Stack& l, SizeType index) { + RAPIDJSON_ASSERT(index != kRegexInvalidState); + + const State& s = regex_.GetState(index); + if (s.out1 != kRegexInvalidState) { // Split + bool matched = AddState(l, s.out); + return AddState(l, s.out1) || matched; + } + else if (!(stateSet_[index >> 5] & (1u << (index & 31)))) { + stateSet_[index >> 5] |= (1u << (index & 31)); + *l.template PushUnsafe() = index; + } + return s.out == kRegexInvalidState; // by using PushUnsafe() above, we can ensure s is not validated due to reallocation. + } + + bool MatchRange(SizeType rangeIndex, unsigned codepoint) const { + bool yes = (regex_.GetRange(rangeIndex).start & RegexType::kRangeNegationFlag) == 0; + while (rangeIndex != kRegexInvalidRange) { + const Range& r = regex_.GetRange(rangeIndex); + if (codepoint >= (r.start & ~RegexType::kRangeNegationFlag) && codepoint <= r.end) + return yes; + rangeIndex = r.next; + } + return !yes; + } + + const RegexType& regex_; + Allocator* allocator_; + Allocator* ownAllocator_; + Stack state0_; + Stack state1_; + uint32_t* stateSet_; +}; + +typedef GenericRegex > Regex; +typedef GenericRegexSearch RegexSearch; + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#if defined(__clang__) || defined(_MSC_VER) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_INTERNAL_REGEX_H_ diff --git a/include/rapidjson/internal/stack.h b/include/rapidjson/internal/stack.h new file mode 100644 index 0000000..73abd70 --- /dev/null +++ b/include/rapidjson/internal/stack.h @@ -0,0 +1,232 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_INTERNAL_STACK_H_ +#define RAPIDJSON_INTERNAL_STACK_H_ + +#include "../allocators.h" +#include "swap.h" +#include + +#if defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(c++98-compat) +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +/////////////////////////////////////////////////////////////////////////////// +// Stack + +//! A type-unsafe stack for storing different types of data. +/*! \tparam Allocator Allocator for allocating stack memory. +*/ +template +class Stack { +public: + // Optimization note: Do not allocate memory for stack_ in constructor. + // Do it lazily when first Push() -> Expand() -> Resize(). + Stack(Allocator* allocator, size_t stackCapacity) : allocator_(allocator), ownAllocator_(0), stack_(0), stackTop_(0), stackEnd_(0), initialCapacity_(stackCapacity) { + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + Stack(Stack&& rhs) + : allocator_(rhs.allocator_), + ownAllocator_(rhs.ownAllocator_), + stack_(rhs.stack_), + stackTop_(rhs.stackTop_), + stackEnd_(rhs.stackEnd_), + initialCapacity_(rhs.initialCapacity_) + { + rhs.allocator_ = 0; + rhs.ownAllocator_ = 0; + rhs.stack_ = 0; + rhs.stackTop_ = 0; + rhs.stackEnd_ = 0; + rhs.initialCapacity_ = 0; + } +#endif + + ~Stack() { + Destroy(); + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + Stack& operator=(Stack&& rhs) { + if (&rhs != this) + { + Destroy(); + + allocator_ = rhs.allocator_; + ownAllocator_ = rhs.ownAllocator_; + stack_ = rhs.stack_; + stackTop_ = rhs.stackTop_; + stackEnd_ = rhs.stackEnd_; + initialCapacity_ = rhs.initialCapacity_; + + rhs.allocator_ = 0; + rhs.ownAllocator_ = 0; + rhs.stack_ = 0; + rhs.stackTop_ = 0; + rhs.stackEnd_ = 0; + rhs.initialCapacity_ = 0; + } + return *this; + } +#endif + + void Swap(Stack& rhs) RAPIDJSON_NOEXCEPT { + internal::Swap(allocator_, rhs.allocator_); + internal::Swap(ownAllocator_, rhs.ownAllocator_); + internal::Swap(stack_, rhs.stack_); + internal::Swap(stackTop_, rhs.stackTop_); + internal::Swap(stackEnd_, rhs.stackEnd_); + internal::Swap(initialCapacity_, rhs.initialCapacity_); + } + + void Clear() { stackTop_ = stack_; } + + void ShrinkToFit() { + if (Empty()) { + // If the stack is empty, completely deallocate the memory. + Allocator::Free(stack_); // NOLINT (+clang-analyzer-unix.Malloc) + stack_ = 0; + stackTop_ = 0; + stackEnd_ = 0; + } + else + Resize(GetSize()); + } + + // Optimization note: try to minimize the size of this function for force inline. + // Expansion is run very infrequently, so it is moved to another (probably non-inline) function. + template + RAPIDJSON_FORCEINLINE void Reserve(size_t count = 1) { + // Expand the stack if needed + if (RAPIDJSON_UNLIKELY(static_cast(sizeof(T) * count) > (stackEnd_ - stackTop_))) + Expand(count); + } + + template + RAPIDJSON_FORCEINLINE T* Push(size_t count = 1) { + Reserve(count); + return PushUnsafe(count); + } + + template + RAPIDJSON_FORCEINLINE T* PushUnsafe(size_t count = 1) { + RAPIDJSON_ASSERT(stackTop_); + RAPIDJSON_ASSERT(static_cast(sizeof(T) * count) <= (stackEnd_ - stackTop_)); + T* ret = reinterpret_cast(stackTop_); + stackTop_ += sizeof(T) * count; + return ret; + } + + template + T* Pop(size_t count) { + RAPIDJSON_ASSERT(GetSize() >= count * sizeof(T)); + stackTop_ -= count * sizeof(T); + return reinterpret_cast(stackTop_); + } + + template + T* Top() { + RAPIDJSON_ASSERT(GetSize() >= sizeof(T)); + return reinterpret_cast(stackTop_ - sizeof(T)); + } + + template + const T* Top() const { + RAPIDJSON_ASSERT(GetSize() >= sizeof(T)); + return reinterpret_cast(stackTop_ - sizeof(T)); + } + + template + T* End() { return reinterpret_cast(stackTop_); } + + template + const T* End() const { return reinterpret_cast(stackTop_); } + + template + T* Bottom() { return reinterpret_cast(stack_); } + + template + const T* Bottom() const { return reinterpret_cast(stack_); } + + bool HasAllocator() const { + return allocator_ != 0; + } + + Allocator& GetAllocator() { + RAPIDJSON_ASSERT(allocator_); + return *allocator_; + } + + bool Empty() const { return stackTop_ == stack_; } + size_t GetSize() const { return static_cast(stackTop_ - stack_); } + size_t GetCapacity() const { return static_cast(stackEnd_ - stack_); } + +private: + template + void Expand(size_t count) { + // Only expand the capacity if the current stack exists. Otherwise just create a stack with initial capacity. + size_t newCapacity; + if (stack_ == 0) { + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + newCapacity = initialCapacity_; + } else { + newCapacity = GetCapacity(); + newCapacity += (newCapacity + 1) / 2; + } + size_t newSize = GetSize() + sizeof(T) * count; + if (newCapacity < newSize) + newCapacity = newSize; + + Resize(newCapacity); + } + + void Resize(size_t newCapacity) { + const size_t size = GetSize(); // Backup the current size + stack_ = static_cast(allocator_->Realloc(stack_, GetCapacity(), newCapacity)); + stackTop_ = stack_ + size; + stackEnd_ = stack_ + newCapacity; + } + + void Destroy() { + Allocator::Free(stack_); + RAPIDJSON_DELETE(ownAllocator_); // Only delete if it is owned by the stack + } + + // Prohibit copy constructor & assignment operator. + Stack(const Stack&); + Stack& operator=(const Stack&); + + Allocator* allocator_; + Allocator* ownAllocator_; + char *stack_; + char *stackTop_; + char *stackEnd_; + size_t initialCapacity_; +}; + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_STACK_H_ diff --git a/include/rapidjson/internal/strfunc.h b/include/rapidjson/internal/strfunc.h new file mode 100644 index 0000000..b698a8f --- /dev/null +++ b/include/rapidjson/internal/strfunc.h @@ -0,0 +1,83 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_INTERNAL_STRFUNC_H_ +#define RAPIDJSON_INTERNAL_STRFUNC_H_ + +#include "../stream.h" +#include + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +//! Custom strlen() which works on different character types. +/*! \tparam Ch Character type (e.g. char, wchar_t, short) + \param s Null-terminated input string. + \return Number of characters in the string. + \note This has the same semantics as strlen(), the return value is not number of Unicode codepoints. +*/ +template +inline SizeType StrLen(const Ch* s) { + RAPIDJSON_ASSERT(s != 0); + const Ch* p = s; + while (*p) ++p; + return SizeType(p - s); +} + +template <> +inline SizeType StrLen(const char* s) { + return SizeType(std::strlen(s)); +} + +template <> +inline SizeType StrLen(const wchar_t* s) { + return SizeType(std::wcslen(s)); +} + +//! Custom strcmpn() which works on different character types. +/*! \tparam Ch Character type (e.g. char, wchar_t, short) + \param s1 Null-terminated input string. + \param s2 Null-terminated input string. + \return 0 if equal +*/ +template +inline int StrCmp(const Ch* s1, const Ch* s2) { + RAPIDJSON_ASSERT(s1 != 0); + RAPIDJSON_ASSERT(s2 != 0); + while(*s1 && (*s1 == *s2)) { s1++; s2++; } + return static_cast(*s1) < static_cast(*s2) ? -1 : static_cast(*s1) > static_cast(*s2); +} + +//! Returns number of code points in a encoded string. +template +bool CountStringCodePoint(const typename Encoding::Ch* s, SizeType length, SizeType* outCount) { + RAPIDJSON_ASSERT(s != 0); + RAPIDJSON_ASSERT(outCount != 0); + GenericStringStream is(s); + const typename Encoding::Ch* end = s + length; + SizeType count = 0; + while (is.src_ < end) { + unsigned codepoint; + if (!Encoding::Decode(is, &codepoint)) + return false; + count++; + } + *outCount = count; + return true; +} + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_INTERNAL_STRFUNC_H_ diff --git a/include/rapidjson/internal/strtod.h b/include/rapidjson/internal/strtod.h new file mode 100644 index 0000000..57c8418 --- /dev/null +++ b/include/rapidjson/internal/strtod.h @@ -0,0 +1,293 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_STRTOD_ +#define RAPIDJSON_STRTOD_ + +#include "ieee754.h" +#include "biginteger.h" +#include "diyfp.h" +#include "pow10.h" +#include +#include + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +inline double FastPath(double significand, int exp) { + if (exp < -308) + return 0.0; + else if (exp >= 0) + return significand * internal::Pow10(exp); + else + return significand / internal::Pow10(-exp); +} + +inline double StrtodNormalPrecision(double d, int p) { + if (p < -308) { + // Prevent expSum < -308, making Pow10(p) = 0 + d = FastPath(d, -308); + d = FastPath(d, p + 308); + } + else + d = FastPath(d, p); + return d; +} + +template +inline T Min3(T a, T b, T c) { + T m = a; + if (m > b) m = b; + if (m > c) m = c; + return m; +} + +inline int CheckWithinHalfULP(double b, const BigInteger& d, int dExp) { + const Double db(b); + const uint64_t bInt = db.IntegerSignificand(); + const int bExp = db.IntegerExponent(); + const int hExp = bExp - 1; + + int dS_Exp2 = 0, dS_Exp5 = 0, bS_Exp2 = 0, bS_Exp5 = 0, hS_Exp2 = 0, hS_Exp5 = 0; + + // Adjust for decimal exponent + if (dExp >= 0) { + dS_Exp2 += dExp; + dS_Exp5 += dExp; + } + else { + bS_Exp2 -= dExp; + bS_Exp5 -= dExp; + hS_Exp2 -= dExp; + hS_Exp5 -= dExp; + } + + // Adjust for binary exponent + if (bExp >= 0) + bS_Exp2 += bExp; + else { + dS_Exp2 -= bExp; + hS_Exp2 -= bExp; + } + + // Adjust for half ulp exponent + if (hExp >= 0) + hS_Exp2 += hExp; + else { + dS_Exp2 -= hExp; + bS_Exp2 -= hExp; + } + + // Remove common power of two factor from all three scaled values + int common_Exp2 = Min3(dS_Exp2, bS_Exp2, hS_Exp2); + dS_Exp2 -= common_Exp2; + bS_Exp2 -= common_Exp2; + hS_Exp2 -= common_Exp2; + + BigInteger dS = d; + dS.MultiplyPow5(static_cast(dS_Exp5)) <<= static_cast(dS_Exp2); + + BigInteger bS(bInt); + bS.MultiplyPow5(static_cast(bS_Exp5)) <<= static_cast(bS_Exp2); + + BigInteger hS(1); + hS.MultiplyPow5(static_cast(hS_Exp5)) <<= static_cast(hS_Exp2); + + BigInteger delta(0); + dS.Difference(bS, &delta); + + return delta.Compare(hS); +} + +inline bool StrtodFast(double d, int p, double* result) { + // Use fast path for string-to-double conversion if possible + // see http://www.exploringbinary.com/fast-path-decimal-to-floating-point-conversion/ + if (p > 22 && p < 22 + 16) { + // Fast Path Cases In Disguise + d *= internal::Pow10(p - 22); + p = 22; + } + + if (p >= -22 && p <= 22 && d <= 9007199254740991.0) { // 2^53 - 1 + *result = FastPath(d, p); + return true; + } + else + return false; +} + +// Compute an approximation and see if it is within 1/2 ULP +template +inline bool StrtodDiyFp(const Ch* decimals, int dLen, int dExp, double* result) { + uint64_t significand = 0; + int i = 0; // 2^64 - 1 = 18446744073709551615, 1844674407370955161 = 0x1999999999999999 + for (; i < dLen; i++) { + if (significand > RAPIDJSON_UINT64_C2(0x19999999, 0x99999999) || + (significand == RAPIDJSON_UINT64_C2(0x19999999, 0x99999999) && decimals[i] >= Ch('5'))) + break; + significand = significand * 10u + static_cast(decimals[i] - Ch('0')); + } + + if (i < dLen && decimals[i] >= Ch('5')) // Rounding + significand++; + + int remaining = dLen - i; + const int kUlpShift = 3; + const int kUlp = 1 << kUlpShift; + int64_t error = (remaining == 0) ? 0 : kUlp / 2; + + DiyFp v(significand, 0); + v = v.Normalize(); + error <<= -v.e; + + dExp += remaining; + + int actualExp; + DiyFp cachedPower = GetCachedPower10(dExp, &actualExp); + if (actualExp != dExp) { + static const DiyFp kPow10[] = { + DiyFp(RAPIDJSON_UINT64_C2(0xa0000000, 0x00000000), -60), // 10^1 + DiyFp(RAPIDJSON_UINT64_C2(0xc8000000, 0x00000000), -57), // 10^2 + DiyFp(RAPIDJSON_UINT64_C2(0xfa000000, 0x00000000), -54), // 10^3 + DiyFp(RAPIDJSON_UINT64_C2(0x9c400000, 0x00000000), -50), // 10^4 + DiyFp(RAPIDJSON_UINT64_C2(0xc3500000, 0x00000000), -47), // 10^5 + DiyFp(RAPIDJSON_UINT64_C2(0xf4240000, 0x00000000), -44), // 10^6 + DiyFp(RAPIDJSON_UINT64_C2(0x98968000, 0x00000000), -40) // 10^7 + }; + int adjustment = dExp - actualExp; + RAPIDJSON_ASSERT(adjustment >= 1 && adjustment < 8); + v = v * kPow10[adjustment - 1]; + if (dLen + adjustment > 19) // has more digits than decimal digits in 64-bit + error += kUlp / 2; + } + + v = v * cachedPower; + + error += kUlp + (error == 0 ? 0 : 1); + + const int oldExp = v.e; + v = v.Normalize(); + error <<= oldExp - v.e; + + const int effectiveSignificandSize = Double::EffectiveSignificandSize(64 + v.e); + int precisionSize = 64 - effectiveSignificandSize; + if (precisionSize + kUlpShift >= 64) { + int scaleExp = (precisionSize + kUlpShift) - 63; + v.f >>= scaleExp; + v.e += scaleExp; + error = (error >> scaleExp) + 1 + kUlp; + precisionSize -= scaleExp; + } + + DiyFp rounded(v.f >> precisionSize, v.e + precisionSize); + const uint64_t precisionBits = (v.f & ((uint64_t(1) << precisionSize) - 1)) * kUlp; + const uint64_t halfWay = (uint64_t(1) << (precisionSize - 1)) * kUlp; + if (precisionBits >= halfWay + static_cast(error)) { + rounded.f++; + if (rounded.f & (DiyFp::kDpHiddenBit << 1)) { // rounding overflows mantissa (issue #340) + rounded.f >>= 1; + rounded.e++; + } + } + + *result = rounded.ToDouble(); + + return halfWay - static_cast(error) >= precisionBits || precisionBits >= halfWay + static_cast(error); +} + +template +inline double StrtodBigInteger(double approx, const Ch* decimals, int dLen, int dExp) { + RAPIDJSON_ASSERT(dLen >= 0); + const BigInteger dInt(decimals, static_cast(dLen)); + Double a(approx); + int cmp = CheckWithinHalfULP(a.Value(), dInt, dExp); + if (cmp < 0) + return a.Value(); // within half ULP + else if (cmp == 0) { + // Round towards even + if (a.Significand() & 1) + return a.NextPositiveDouble(); + else + return a.Value(); + } + else // adjustment + return a.NextPositiveDouble(); +} + +template +inline double StrtodFullPrecision(double d, int p, const Ch* decimals, size_t length, size_t decimalPosition, int exp) { + RAPIDJSON_ASSERT(d >= 0.0); + RAPIDJSON_ASSERT(length >= 1); + + double result = 0.0; + if (StrtodFast(d, p, &result)) + return result; + + RAPIDJSON_ASSERT(length <= INT_MAX); + int dLen = static_cast(length); + + RAPIDJSON_ASSERT(length >= decimalPosition); + RAPIDJSON_ASSERT(length - decimalPosition <= INT_MAX); + int dExpAdjust = static_cast(length - decimalPosition); + + RAPIDJSON_ASSERT(exp >= INT_MIN + dExpAdjust); + int dExp = exp - dExpAdjust; + + // Make sure length+dExp does not overflow + RAPIDJSON_ASSERT(dExp <= INT_MAX - dLen); + + // Trim leading zeros + while (dLen > 0 && *decimals == '0') { + dLen--; + decimals++; + } + + // Trim trailing zeros + while (dLen > 0 && decimals[dLen - 1] == '0') { + dLen--; + dExp++; + } + + if (dLen == 0) { // Buffer only contains zeros. + return 0.0; + } + + // Trim right-most digits + const int kMaxDecimalDigit = 767 + 1; + if (dLen > kMaxDecimalDigit) { + dExp += dLen - kMaxDecimalDigit; + dLen = kMaxDecimalDigit; + } + + // If too small, underflow to zero. + // Any x <= 10^-324 is interpreted as zero. + if (dLen + dExp <= -324) + return 0.0; + + // If too large, overflow to infinity. + // Any x >= 10^309 is interpreted as +infinity. + if (dLen + dExp > 309) + return std::numeric_limits::infinity(); + + if (StrtodDiyFp(decimals, dLen, dExp, &result)) + return result; + + // Use approximation from StrtodDiyFp and make adjustment with BigInteger comparison + return StrtodBigInteger(result, decimals, dLen, dExp); +} + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_STRTOD_ diff --git a/include/rapidjson/internal/swap.h b/include/rapidjson/internal/swap.h new file mode 100644 index 0000000..2cf92f9 --- /dev/null +++ b/include/rapidjson/internal/swap.h @@ -0,0 +1,46 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_INTERNAL_SWAP_H_ +#define RAPIDJSON_INTERNAL_SWAP_H_ + +#include "../rapidjson.h" + +#if defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(c++98-compat) +#endif + +RAPIDJSON_NAMESPACE_BEGIN +namespace internal { + +//! Custom swap() to avoid dependency on C++ header +/*! \tparam T Type of the arguments to swap, should be instantiated with primitive C++ types only. + \note This has the same semantics as std::swap(). +*/ +template +inline void Swap(T& a, T& b) RAPIDJSON_NOEXCEPT { + T tmp = a; + a = b; + b = tmp; +} + +} // namespace internal +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_INTERNAL_SWAP_H_ diff --git a/include/rapidjson/istreamwrapper.h b/include/rapidjson/istreamwrapper.h new file mode 100644 index 0000000..01437ec --- /dev/null +++ b/include/rapidjson/istreamwrapper.h @@ -0,0 +1,128 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_ISTREAMWRAPPER_H_ +#define RAPIDJSON_ISTREAMWRAPPER_H_ + +#include "stream.h" +#include +#include + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4351) // new behavior: elements of array 'array' will be default initialized +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Wrapper of \c std::basic_istream into RapidJSON's Stream concept. +/*! + The classes can be wrapped including but not limited to: + + - \c std::istringstream + - \c std::stringstream + - \c std::wistringstream + - \c std::wstringstream + - \c std::ifstream + - \c std::fstream + - \c std::wifstream + - \c std::wfstream + + \tparam StreamType Class derived from \c std::basic_istream. +*/ + +template +class BasicIStreamWrapper { +public: + typedef typename StreamType::char_type Ch; + + //! Constructor. + /*! + \param stream stream opened for read. + */ + BasicIStreamWrapper(StreamType &stream) : stream_(stream), buffer_(peekBuffer_), bufferSize_(4), bufferLast_(0), current_(buffer_), readCount_(0), count_(0), eof_(false) { + Read(); + } + + //! Constructor. + /*! + \param stream stream opened for read. + \param buffer user-supplied buffer. + \param bufferSize size of buffer in bytes. Must >=4 bytes. + */ + BasicIStreamWrapper(StreamType &stream, char* buffer, size_t bufferSize) : stream_(stream), buffer_(buffer), bufferSize_(bufferSize), bufferLast_(0), current_(buffer_), readCount_(0), count_(0), eof_(false) { + RAPIDJSON_ASSERT(bufferSize >= 4); + Read(); + } + + Ch Peek() const { return *current_; } + Ch Take() { Ch c = *current_; Read(); return c; } + size_t Tell() const { return count_ + static_cast(current_ - buffer_); } + + // Not implemented + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + + // For encoding detection only. + const Ch* Peek4() const { + return (current_ + 4 - !eof_ <= bufferLast_) ? current_ : 0; + } + +private: + BasicIStreamWrapper(); + BasicIStreamWrapper(const BasicIStreamWrapper&); + BasicIStreamWrapper& operator=(const BasicIStreamWrapper&); + + void Read() { + if (current_ < bufferLast_) + ++current_; + else if (!eof_) { + count_ += readCount_; + readCount_ = bufferSize_; + bufferLast_ = buffer_ + readCount_ - 1; + current_ = buffer_; + + if (!stream_.read(buffer_, static_cast(bufferSize_))) { + readCount_ = static_cast(stream_.gcount()); + *(bufferLast_ = buffer_ + readCount_) = '\0'; + eof_ = true; + } + } + } + + StreamType &stream_; + Ch peekBuffer_[4], *buffer_; + size_t bufferSize_; + Ch *bufferLast_; + Ch *current_; + size_t readCount_; + size_t count_; //!< Number of characters read + bool eof_; +}; + +typedef BasicIStreamWrapper IStreamWrapper; +typedef BasicIStreamWrapper WIStreamWrapper; + +#if defined(__clang__) || defined(_MSC_VER) +RAPIDJSON_DIAG_POP +#endif + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_ISTREAMWRAPPER_H_ diff --git a/include/rapidjson/memorybuffer.h b/include/rapidjson/memorybuffer.h new file mode 100644 index 0000000..ffbc41e --- /dev/null +++ b/include/rapidjson/memorybuffer.h @@ -0,0 +1,70 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_MEMORYBUFFER_H_ +#define RAPIDJSON_MEMORYBUFFER_H_ + +#include "stream.h" +#include "internal/stack.h" + +RAPIDJSON_NAMESPACE_BEGIN + +//! Represents an in-memory output byte stream. +/*! + This class is mainly for being wrapped by EncodedOutputStream or AutoUTFOutputStream. + + It is similar to FileWriteBuffer but the destination is an in-memory buffer instead of a file. + + Differences between MemoryBuffer and StringBuffer: + 1. StringBuffer has Encoding but MemoryBuffer is only a byte buffer. + 2. StringBuffer::GetString() returns a null-terminated string. MemoryBuffer::GetBuffer() returns a buffer without terminator. + + \tparam Allocator type for allocating memory buffer. + \note implements Stream concept +*/ +template +struct GenericMemoryBuffer { + typedef char Ch; // byte + + GenericMemoryBuffer(Allocator* allocator = 0, size_t capacity = kDefaultCapacity) : stack_(allocator, capacity) {} + + void Put(Ch c) { *stack_.template Push() = c; } + void Flush() {} + + void Clear() { stack_.Clear(); } + void ShrinkToFit() { stack_.ShrinkToFit(); } + Ch* Push(size_t count) { return stack_.template Push(count); } + void Pop(size_t count) { stack_.template Pop(count); } + + const Ch* GetBuffer() const { + return stack_.template Bottom(); + } + + size_t GetSize() const { return stack_.GetSize(); } + + static const size_t kDefaultCapacity = 256; + mutable internal::Stack stack_; +}; + +typedef GenericMemoryBuffer<> MemoryBuffer; + +//! Implement specialized version of PutN() with memset() for better performance. +template<> +inline void PutN(MemoryBuffer& memoryBuffer, char c, size_t n) { + std::memset(memoryBuffer.stack_.Push(n), c, n * sizeof(c)); +} + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_MEMORYBUFFER_H_ diff --git a/include/rapidjson/memorystream.h b/include/rapidjson/memorystream.h new file mode 100644 index 0000000..77af6c9 --- /dev/null +++ b/include/rapidjson/memorystream.h @@ -0,0 +1,71 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_MEMORYSTREAM_H_ +#define RAPIDJSON_MEMORYSTREAM_H_ + +#include "stream.h" + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(unreachable-code) +RAPIDJSON_DIAG_OFF(missing-noreturn) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Represents an in-memory input byte stream. +/*! + This class is mainly for being wrapped by EncodedInputStream or AutoUTFInputStream. + + It is similar to FileReadBuffer but the source is an in-memory buffer instead of a file. + + Differences between MemoryStream and StringStream: + 1. StringStream has encoding but MemoryStream is a byte stream. + 2. MemoryStream needs size of the source buffer and the buffer don't need to be null terminated. StringStream assume null-terminated string as source. + 3. MemoryStream supports Peek4() for encoding detection. StringStream is specified with an encoding so it should not have Peek4(). + \note implements Stream concept +*/ +struct MemoryStream { + typedef char Ch; // byte + + MemoryStream(const Ch *src, size_t size) : src_(src), begin_(src), end_(src + size), size_(size) {} + + Ch Peek() const { return RAPIDJSON_UNLIKELY(src_ == end_) ? '\0' : *src_; } + Ch Take() { return RAPIDJSON_UNLIKELY(src_ == end_) ? '\0' : *src_++; } + size_t Tell() const { return static_cast(src_ - begin_); } + + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + + // For encoding detection only. + const Ch* Peek4() const { + return Tell() + 4 <= size_ ? src_ : 0; + } + + const Ch* src_; //!< Current read position. + const Ch* begin_; //!< Original head of the string. + const Ch* end_; //!< End of stream. + size_t size_; //!< Size of the stream. +}; + +RAPIDJSON_NAMESPACE_END + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_MEMORYBUFFER_H_ diff --git a/include/rapidjson/msinttypes/inttypes.h b/include/rapidjson/msinttypes/inttypes.h new file mode 100644 index 0000000..1811128 --- /dev/null +++ b/include/rapidjson/msinttypes/inttypes.h @@ -0,0 +1,316 @@ +// ISO C9x compliant inttypes.h for Microsoft Visual Studio +// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 +// +// Copyright (c) 2006-2013 Alexander Chemeris +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are met: +// +// 1. Redistributions of source code must retain the above copyright notice, +// this list of conditions and the following disclaimer. +// +// 2. Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// +// 3. Neither the name of the product nor the names of its contributors may +// be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED +// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO +// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; +// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR +// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF +// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +// +/////////////////////////////////////////////////////////////////////////////// + +// The above software in this distribution may have been modified by +// THL A29 Limited ("Tencent Modifications"). +// All Tencent Modifications are Copyright (C) 2015 THL A29 Limited. + +#ifndef _MSC_VER // [ +#error "Use this header only with Microsoft Visual C++ compilers!" +#endif // _MSC_VER ] + +#ifndef _MSC_INTTYPES_H_ // [ +#define _MSC_INTTYPES_H_ + +#if _MSC_VER > 1000 +#pragma once +#endif + +#include "stdint.h" + +// miloyip: VC supports inttypes.h since VC2013 +#if _MSC_VER >= 1800 +#include +#else + +// 7.8 Format conversion of integer types + +typedef struct { + intmax_t quot; + intmax_t rem; +} imaxdiv_t; + +// 7.8.1 Macros for format specifiers + +#if !defined(__cplusplus) || defined(__STDC_FORMAT_MACROS) // [ See footnote 185 at page 198 + +// The fprintf macros for signed integers are: +#define PRId8 "d" +#define PRIi8 "i" +#define PRIdLEAST8 "d" +#define PRIiLEAST8 "i" +#define PRIdFAST8 "d" +#define PRIiFAST8 "i" + +#define PRId16 "hd" +#define PRIi16 "hi" +#define PRIdLEAST16 "hd" +#define PRIiLEAST16 "hi" +#define PRIdFAST16 "hd" +#define PRIiFAST16 "hi" + +#define PRId32 "I32d" +#define PRIi32 "I32i" +#define PRIdLEAST32 "I32d" +#define PRIiLEAST32 "I32i" +#define PRIdFAST32 "I32d" +#define PRIiFAST32 "I32i" + +#define PRId64 "I64d" +#define PRIi64 "I64i" +#define PRIdLEAST64 "I64d" +#define PRIiLEAST64 "I64i" +#define PRIdFAST64 "I64d" +#define PRIiFAST64 "I64i" + +#define PRIdMAX "I64d" +#define PRIiMAX "I64i" + +#define PRIdPTR "Id" +#define PRIiPTR "Ii" + +// The fprintf macros for unsigned integers are: +#define PRIo8 "o" +#define PRIu8 "u" +#define PRIx8 "x" +#define PRIX8 "X" +#define PRIoLEAST8 "o" +#define PRIuLEAST8 "u" +#define PRIxLEAST8 "x" +#define PRIXLEAST8 "X" +#define PRIoFAST8 "o" +#define PRIuFAST8 "u" +#define PRIxFAST8 "x" +#define PRIXFAST8 "X" + +#define PRIo16 "ho" +#define PRIu16 "hu" +#define PRIx16 "hx" +#define PRIX16 "hX" +#define PRIoLEAST16 "ho" +#define PRIuLEAST16 "hu" +#define PRIxLEAST16 "hx" +#define PRIXLEAST16 "hX" +#define PRIoFAST16 "ho" +#define PRIuFAST16 "hu" +#define PRIxFAST16 "hx" +#define PRIXFAST16 "hX" + +#define PRIo32 "I32o" +#define PRIu32 "I32u" +#define PRIx32 "I32x" +#define PRIX32 "I32X" +#define PRIoLEAST32 "I32o" +#define PRIuLEAST32 "I32u" +#define PRIxLEAST32 "I32x" +#define PRIXLEAST32 "I32X" +#define PRIoFAST32 "I32o" +#define PRIuFAST32 "I32u" +#define PRIxFAST32 "I32x" +#define PRIXFAST32 "I32X" + +#define PRIo64 "I64o" +#define PRIu64 "I64u" +#define PRIx64 "I64x" +#define PRIX64 "I64X" +#define PRIoLEAST64 "I64o" +#define PRIuLEAST64 "I64u" +#define PRIxLEAST64 "I64x" +#define PRIXLEAST64 "I64X" +#define PRIoFAST64 "I64o" +#define PRIuFAST64 "I64u" +#define PRIxFAST64 "I64x" +#define PRIXFAST64 "I64X" + +#define PRIoMAX "I64o" +#define PRIuMAX "I64u" +#define PRIxMAX "I64x" +#define PRIXMAX "I64X" + +#define PRIoPTR "Io" +#define PRIuPTR "Iu" +#define PRIxPTR "Ix" +#define PRIXPTR "IX" + +// The fscanf macros for signed integers are: +#define SCNd8 "d" +#define SCNi8 "i" +#define SCNdLEAST8 "d" +#define SCNiLEAST8 "i" +#define SCNdFAST8 "d" +#define SCNiFAST8 "i" + +#define SCNd16 "hd" +#define SCNi16 "hi" +#define SCNdLEAST16 "hd" +#define SCNiLEAST16 "hi" +#define SCNdFAST16 "hd" +#define SCNiFAST16 "hi" + +#define SCNd32 "ld" +#define SCNi32 "li" +#define SCNdLEAST32 "ld" +#define SCNiLEAST32 "li" +#define SCNdFAST32 "ld" +#define SCNiFAST32 "li" + +#define SCNd64 "I64d" +#define SCNi64 "I64i" +#define SCNdLEAST64 "I64d" +#define SCNiLEAST64 "I64i" +#define SCNdFAST64 "I64d" +#define SCNiFAST64 "I64i" + +#define SCNdMAX "I64d" +#define SCNiMAX "I64i" + +#ifdef _WIN64 // [ +# define SCNdPTR "I64d" +# define SCNiPTR "I64i" +#else // _WIN64 ][ +# define SCNdPTR "ld" +# define SCNiPTR "li" +#endif // _WIN64 ] + +// The fscanf macros for unsigned integers are: +#define SCNo8 "o" +#define SCNu8 "u" +#define SCNx8 "x" +#define SCNX8 "X" +#define SCNoLEAST8 "o" +#define SCNuLEAST8 "u" +#define SCNxLEAST8 "x" +#define SCNXLEAST8 "X" +#define SCNoFAST8 "o" +#define SCNuFAST8 "u" +#define SCNxFAST8 "x" +#define SCNXFAST8 "X" + +#define SCNo16 "ho" +#define SCNu16 "hu" +#define SCNx16 "hx" +#define SCNX16 "hX" +#define SCNoLEAST16 "ho" +#define SCNuLEAST16 "hu" +#define SCNxLEAST16 "hx" +#define SCNXLEAST16 "hX" +#define SCNoFAST16 "ho" +#define SCNuFAST16 "hu" +#define SCNxFAST16 "hx" +#define SCNXFAST16 "hX" + +#define SCNo32 "lo" +#define SCNu32 "lu" +#define SCNx32 "lx" +#define SCNX32 "lX" +#define SCNoLEAST32 "lo" +#define SCNuLEAST32 "lu" +#define SCNxLEAST32 "lx" +#define SCNXLEAST32 "lX" +#define SCNoFAST32 "lo" +#define SCNuFAST32 "lu" +#define SCNxFAST32 "lx" +#define SCNXFAST32 "lX" + +#define SCNo64 "I64o" +#define SCNu64 "I64u" +#define SCNx64 "I64x" +#define SCNX64 "I64X" +#define SCNoLEAST64 "I64o" +#define SCNuLEAST64 "I64u" +#define SCNxLEAST64 "I64x" +#define SCNXLEAST64 "I64X" +#define SCNoFAST64 "I64o" +#define SCNuFAST64 "I64u" +#define SCNxFAST64 "I64x" +#define SCNXFAST64 "I64X" + +#define SCNoMAX "I64o" +#define SCNuMAX "I64u" +#define SCNxMAX "I64x" +#define SCNXMAX "I64X" + +#ifdef _WIN64 // [ +# define SCNoPTR "I64o" +# define SCNuPTR "I64u" +# define SCNxPTR "I64x" +# define SCNXPTR "I64X" +#else // _WIN64 ][ +# define SCNoPTR "lo" +# define SCNuPTR "lu" +# define SCNxPTR "lx" +# define SCNXPTR "lX" +#endif // _WIN64 ] + +#endif // __STDC_FORMAT_MACROS ] + +// 7.8.2 Functions for greatest-width integer types + +// 7.8.2.1 The imaxabs function +#define imaxabs _abs64 + +// 7.8.2.2 The imaxdiv function + +// This is modified version of div() function from Microsoft's div.c found +// in %MSVC.NET%\crt\src\div.c +#ifdef STATIC_IMAXDIV // [ +static +#else // STATIC_IMAXDIV ][ +_inline +#endif // STATIC_IMAXDIV ] +imaxdiv_t __cdecl imaxdiv(intmax_t numer, intmax_t denom) +{ + imaxdiv_t result; + + result.quot = numer / denom; + result.rem = numer % denom; + + if (numer < 0 && result.rem > 0) { + // did division wrong; must fix up + ++result.quot; + result.rem -= denom; + } + + return result; +} + +// 7.8.2.3 The strtoimax and strtoumax functions +#define strtoimax _strtoi64 +#define strtoumax _strtoui64 + +// 7.8.2.4 The wcstoimax and wcstoumax functions +#define wcstoimax _wcstoi64 +#define wcstoumax _wcstoui64 + +#endif // _MSC_VER >= 1800 + +#endif // _MSC_INTTYPES_H_ ] diff --git a/include/rapidjson/msinttypes/stdint.h b/include/rapidjson/msinttypes/stdint.h new file mode 100644 index 0000000..3d4477b --- /dev/null +++ b/include/rapidjson/msinttypes/stdint.h @@ -0,0 +1,300 @@ +// ISO C9x compliant stdint.h for Microsoft Visual Studio +// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124 +// +// Copyright (c) 2006-2013 Alexander Chemeris +// +// Redistribution and use in source and binary forms, with or without +// modification, are permitted provided that the following conditions are met: +// +// 1. Redistributions of source code must retain the above copyright notice, +// this list of conditions and the following disclaimer. +// +// 2. Redistributions in binary form must reproduce the above copyright +// notice, this list of conditions and the following disclaimer in the +// documentation and/or other materials provided with the distribution. +// +// 3. Neither the name of the product nor the names of its contributors may +// be used to endorse or promote products derived from this software +// without specific prior written permission. +// +// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED +// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO +// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; +// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, +// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR +// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF +// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +// +/////////////////////////////////////////////////////////////////////////////// + +// The above software in this distribution may have been modified by +// THL A29 Limited ("Tencent Modifications"). +// All Tencent Modifications are Copyright (C) 2015 THL A29 Limited. + +#ifndef _MSC_VER // [ +#error "Use this header only with Microsoft Visual C++ compilers!" +#endif // _MSC_VER ] + +#ifndef _MSC_STDINT_H_ // [ +#define _MSC_STDINT_H_ + +#if _MSC_VER > 1000 +#pragma once +#endif + +// miloyip: Originally Visual Studio 2010 uses its own stdint.h. However it generates warning with INT64_C(), so change to use this file for vs2010. +#if _MSC_VER >= 1600 // [ +#include + +#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) // [ See footnote 224 at page 260 + +#undef INT8_C +#undef INT16_C +#undef INT32_C +#undef INT64_C +#undef UINT8_C +#undef UINT16_C +#undef UINT32_C +#undef UINT64_C + +// 7.18.4.1 Macros for minimum-width integer constants + +#define INT8_C(val) val##i8 +#define INT16_C(val) val##i16 +#define INT32_C(val) val##i32 +#define INT64_C(val) val##i64 + +#define UINT8_C(val) val##ui8 +#define UINT16_C(val) val##ui16 +#define UINT32_C(val) val##ui32 +#define UINT64_C(val) val##ui64 + +// 7.18.4.2 Macros for greatest-width integer constants +// These #ifndef's are needed to prevent collisions with . +// Check out Issue 9 for the details. +#ifndef INTMAX_C // [ +# define INTMAX_C INT64_C +#endif // INTMAX_C ] +#ifndef UINTMAX_C // [ +# define UINTMAX_C UINT64_C +#endif // UINTMAX_C ] + +#endif // __STDC_CONSTANT_MACROS ] + +#else // ] _MSC_VER >= 1700 [ + +#include + +// For Visual Studio 6 in C++ mode and for many Visual Studio versions when +// compiling for ARM we have to wrap include with 'extern "C++" {}' +// or compiler would give many errors like this: +// error C2733: second C linkage of overloaded function 'wmemchr' not allowed +#if defined(__cplusplus) && !defined(_M_ARM) +extern "C" { +#endif +# include +#if defined(__cplusplus) && !defined(_M_ARM) +} +#endif + +// Define _W64 macros to mark types changing their size, like intptr_t. +#ifndef _W64 +# if !defined(__midl) && (defined(_X86_) || defined(_M_IX86)) && _MSC_VER >= 1300 +# define _W64 __w64 +# else +# define _W64 +# endif +#endif + + +// 7.18.1 Integer types + +// 7.18.1.1 Exact-width integer types + +// Visual Studio 6 and Embedded Visual C++ 4 doesn't +// realize that, e.g. char has the same size as __int8 +// so we give up on __intX for them. +#if (_MSC_VER < 1300) + typedef signed char int8_t; + typedef signed short int16_t; + typedef signed int int32_t; + typedef unsigned char uint8_t; + typedef unsigned short uint16_t; + typedef unsigned int uint32_t; +#else + typedef signed __int8 int8_t; + typedef signed __int16 int16_t; + typedef signed __int32 int32_t; + typedef unsigned __int8 uint8_t; + typedef unsigned __int16 uint16_t; + typedef unsigned __int32 uint32_t; +#endif +typedef signed __int64 int64_t; +typedef unsigned __int64 uint64_t; + + +// 7.18.1.2 Minimum-width integer types +typedef int8_t int_least8_t; +typedef int16_t int_least16_t; +typedef int32_t int_least32_t; +typedef int64_t int_least64_t; +typedef uint8_t uint_least8_t; +typedef uint16_t uint_least16_t; +typedef uint32_t uint_least32_t; +typedef uint64_t uint_least64_t; + +// 7.18.1.3 Fastest minimum-width integer types +typedef int8_t int_fast8_t; +typedef int16_t int_fast16_t; +typedef int32_t int_fast32_t; +typedef int64_t int_fast64_t; +typedef uint8_t uint_fast8_t; +typedef uint16_t uint_fast16_t; +typedef uint32_t uint_fast32_t; +typedef uint64_t uint_fast64_t; + +// 7.18.1.4 Integer types capable of holding object pointers +#ifdef _WIN64 // [ + typedef signed __int64 intptr_t; + typedef unsigned __int64 uintptr_t; +#else // _WIN64 ][ + typedef _W64 signed int intptr_t; + typedef _W64 unsigned int uintptr_t; +#endif // _WIN64 ] + +// 7.18.1.5 Greatest-width integer types +typedef int64_t intmax_t; +typedef uint64_t uintmax_t; + + +// 7.18.2 Limits of specified-width integer types + +#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) // [ See footnote 220 at page 257 and footnote 221 at page 259 + +// 7.18.2.1 Limits of exact-width integer types +#define INT8_MIN ((int8_t)_I8_MIN) +#define INT8_MAX _I8_MAX +#define INT16_MIN ((int16_t)_I16_MIN) +#define INT16_MAX _I16_MAX +#define INT32_MIN ((int32_t)_I32_MIN) +#define INT32_MAX _I32_MAX +#define INT64_MIN ((int64_t)_I64_MIN) +#define INT64_MAX _I64_MAX +#define UINT8_MAX _UI8_MAX +#define UINT16_MAX _UI16_MAX +#define UINT32_MAX _UI32_MAX +#define UINT64_MAX _UI64_MAX + +// 7.18.2.2 Limits of minimum-width integer types +#define INT_LEAST8_MIN INT8_MIN +#define INT_LEAST8_MAX INT8_MAX +#define INT_LEAST16_MIN INT16_MIN +#define INT_LEAST16_MAX INT16_MAX +#define INT_LEAST32_MIN INT32_MIN +#define INT_LEAST32_MAX INT32_MAX +#define INT_LEAST64_MIN INT64_MIN +#define INT_LEAST64_MAX INT64_MAX +#define UINT_LEAST8_MAX UINT8_MAX +#define UINT_LEAST16_MAX UINT16_MAX +#define UINT_LEAST32_MAX UINT32_MAX +#define UINT_LEAST64_MAX UINT64_MAX + +// 7.18.2.3 Limits of fastest minimum-width integer types +#define INT_FAST8_MIN INT8_MIN +#define INT_FAST8_MAX INT8_MAX +#define INT_FAST16_MIN INT16_MIN +#define INT_FAST16_MAX INT16_MAX +#define INT_FAST32_MIN INT32_MIN +#define INT_FAST32_MAX INT32_MAX +#define INT_FAST64_MIN INT64_MIN +#define INT_FAST64_MAX INT64_MAX +#define UINT_FAST8_MAX UINT8_MAX +#define UINT_FAST16_MAX UINT16_MAX +#define UINT_FAST32_MAX UINT32_MAX +#define UINT_FAST64_MAX UINT64_MAX + +// 7.18.2.4 Limits of integer types capable of holding object pointers +#ifdef _WIN64 // [ +# define INTPTR_MIN INT64_MIN +# define INTPTR_MAX INT64_MAX +# define UINTPTR_MAX UINT64_MAX +#else // _WIN64 ][ +# define INTPTR_MIN INT32_MIN +# define INTPTR_MAX INT32_MAX +# define UINTPTR_MAX UINT32_MAX +#endif // _WIN64 ] + +// 7.18.2.5 Limits of greatest-width integer types +#define INTMAX_MIN INT64_MIN +#define INTMAX_MAX INT64_MAX +#define UINTMAX_MAX UINT64_MAX + +// 7.18.3 Limits of other integer types + +#ifdef _WIN64 // [ +# define PTRDIFF_MIN _I64_MIN +# define PTRDIFF_MAX _I64_MAX +#else // _WIN64 ][ +# define PTRDIFF_MIN _I32_MIN +# define PTRDIFF_MAX _I32_MAX +#endif // _WIN64 ] + +#define SIG_ATOMIC_MIN INT_MIN +#define SIG_ATOMIC_MAX INT_MAX + +#ifndef SIZE_MAX // [ +# ifdef _WIN64 // [ +# define SIZE_MAX _UI64_MAX +# else // _WIN64 ][ +# define SIZE_MAX _UI32_MAX +# endif // _WIN64 ] +#endif // SIZE_MAX ] + +// WCHAR_MIN and WCHAR_MAX are also defined in +#ifndef WCHAR_MIN // [ +# define WCHAR_MIN 0 +#endif // WCHAR_MIN ] +#ifndef WCHAR_MAX // [ +# define WCHAR_MAX _UI16_MAX +#endif // WCHAR_MAX ] + +#define WINT_MIN 0 +#define WINT_MAX _UI16_MAX + +#endif // __STDC_LIMIT_MACROS ] + + +// 7.18.4 Limits of other integer types + +#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) // [ See footnote 224 at page 260 + +// 7.18.4.1 Macros for minimum-width integer constants + +#define INT8_C(val) val##i8 +#define INT16_C(val) val##i16 +#define INT32_C(val) val##i32 +#define INT64_C(val) val##i64 + +#define UINT8_C(val) val##ui8 +#define UINT16_C(val) val##ui16 +#define UINT32_C(val) val##ui32 +#define UINT64_C(val) val##ui64 + +// 7.18.4.2 Macros for greatest-width integer constants +// These #ifndef's are needed to prevent collisions with . +// Check out Issue 9 for the details. +#ifndef INTMAX_C // [ +# define INTMAX_C INT64_C +#endif // INTMAX_C ] +#ifndef UINTMAX_C // [ +# define UINTMAX_C UINT64_C +#endif // UINTMAX_C ] + +#endif // __STDC_CONSTANT_MACROS ] + +#endif // _MSC_VER >= 1600 ] + +#endif // _MSC_STDINT_H_ ] diff --git a/include/rapidjson/ostreamwrapper.h b/include/rapidjson/ostreamwrapper.h new file mode 100644 index 0000000..11ed4d3 --- /dev/null +++ b/include/rapidjson/ostreamwrapper.h @@ -0,0 +1,81 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_OSTREAMWRAPPER_H_ +#define RAPIDJSON_OSTREAMWRAPPER_H_ + +#include "stream.h" +#include + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Wrapper of \c std::basic_ostream into RapidJSON's Stream concept. +/*! + The classes can be wrapped including but not limited to: + + - \c std::ostringstream + - \c std::stringstream + - \c std::wpstringstream + - \c std::wstringstream + - \c std::ifstream + - \c std::fstream + - \c std::wofstream + - \c std::wfstream + + \tparam StreamType Class derived from \c std::basic_ostream. +*/ + +template +class BasicOStreamWrapper { +public: + typedef typename StreamType::char_type Ch; + BasicOStreamWrapper(StreamType& stream) : stream_(stream) {} + + void Put(Ch c) { + stream_.put(c); + } + + void Flush() { + stream_.flush(); + } + + // Not implemented + char Peek() const { RAPIDJSON_ASSERT(false); return 0; } + char Take() { RAPIDJSON_ASSERT(false); return 0; } + size_t Tell() const { RAPIDJSON_ASSERT(false); return 0; } + char* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + size_t PutEnd(char*) { RAPIDJSON_ASSERT(false); return 0; } + +private: + BasicOStreamWrapper(const BasicOStreamWrapper&); + BasicOStreamWrapper& operator=(const BasicOStreamWrapper&); + + StreamType& stream_; +}; + +typedef BasicOStreamWrapper OStreamWrapper; +typedef BasicOStreamWrapper WOStreamWrapper; + +#ifdef __clang__ +RAPIDJSON_DIAG_POP +#endif + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_OSTREAMWRAPPER_H_ diff --git a/include/rapidjson/pointer.h b/include/rapidjson/pointer.h new file mode 100644 index 0000000..355929e --- /dev/null +++ b/include/rapidjson/pointer.h @@ -0,0 +1,1482 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_POINTER_H_ +#define RAPIDJSON_POINTER_H_ + +#include "document.h" +#include "uri.h" +#include "internal/itoa.h" +#include "error/error.h" // PointerParseErrorCode + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(switch-enum) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +#if defined(RAPIDJSON_CPLUSPLUS) && RAPIDJSON_CPLUSPLUS >= 201703L +#define RAPIDJSON_IF_CONSTEXPR if constexpr +#else +#define RAPIDJSON_IF_CONSTEXPR if +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +static const SizeType kPointerInvalidIndex = ~SizeType(0); //!< Represents an invalid index in GenericPointer::Token + +/////////////////////////////////////////////////////////////////////////////// +// GenericPointer + +//! Represents a JSON Pointer. Use Pointer for UTF8 encoding and default allocator. +/*! + This class implements RFC 6901 "JavaScript Object Notation (JSON) Pointer" + (https://tools.ietf.org/html/rfc6901). + + A JSON pointer is for identifying a specific value in a JSON document + (GenericDocument). It can simplify coding of DOM tree manipulation, because it + can access multiple-level depth of DOM tree with single API call. + + After it parses a string representation (e.g. "/foo/0" or URI fragment + representation (e.g. "#/foo/0") into its internal representation (tokens), + it can be used to resolve a specific value in multiple documents, or sub-tree + of documents. + + Contrary to GenericValue, Pointer can be copy constructed and copy assigned. + Apart from assignment, a Pointer cannot be modified after construction. + + Although Pointer is very convenient, please aware that constructing Pointer + involves parsing and dynamic memory allocation. A special constructor with user- + supplied tokens eliminates these. + + GenericPointer depends on GenericDocument and GenericValue. + + \tparam ValueType The value type of the DOM tree. E.g. GenericValue > + \tparam Allocator The allocator type for allocating memory for internal representation. + + \note GenericPointer uses same encoding of ValueType. + However, Allocator of GenericPointer is independent of Allocator of Value. +*/ +template +class GenericPointer { +public: + typedef typename ValueType::EncodingType EncodingType; //!< Encoding type from Value + typedef typename ValueType::Ch Ch; //!< Character type from Value + typedef GenericUri UriType; + + + //! A token is the basic units of internal representation. + /*! + A JSON pointer string representation "/foo/123" is parsed to two tokens: + "foo" and 123. 123 will be represented in both numeric form and string form. + They are resolved according to the actual value type (object or array). + + For token that are not numbers, or the numeric value is out of bound + (greater than limits of SizeType), they are only treated as string form + (i.e. the token's index will be equal to kPointerInvalidIndex). + + This struct is public so that user can create a Pointer without parsing and + allocation, using a special constructor. + */ + struct Token { + const Ch* name; //!< Name of the token. It has null character at the end but it can contain null character. + SizeType length; //!< Length of the name. + SizeType index; //!< A valid array index, if it is not equal to kPointerInvalidIndex. + }; + + //!@name Constructors and destructor. + //@{ + + //! Default constructor. + GenericPointer(Allocator* allocator = 0) : allocator_(allocator), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) {} + + //! Constructor that parses a string or URI fragment representation. + /*! + \param source A null-terminated, string or URI fragment representation of JSON pointer. + \param allocator User supplied allocator for this pointer. If no allocator is provided, it creates a self-owned one. + */ + explicit GenericPointer(const Ch* source, Allocator* allocator = 0) : allocator_(allocator), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) { + Parse(source, internal::StrLen(source)); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Constructor that parses a string or URI fragment representation. + /*! + \param source A string or URI fragment representation of JSON pointer. + \param allocator User supplied allocator for this pointer. If no allocator is provided, it creates a self-owned one. + \note Requires the definition of the preprocessor symbol \ref RAPIDJSON_HAS_STDSTRING. + */ + explicit GenericPointer(const std::basic_string& source, Allocator* allocator = 0) : allocator_(allocator), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) { + Parse(source.c_str(), source.size()); + } +#endif + + //! Constructor that parses a string or URI fragment representation, with length of the source string. + /*! + \param source A string or URI fragment representation of JSON pointer. + \param length Length of source. + \param allocator User supplied allocator for this pointer. If no allocator is provided, it creates a self-owned one. + \note Slightly faster than the overload without length. + */ + GenericPointer(const Ch* source, size_t length, Allocator* allocator = 0) : allocator_(allocator), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) { + Parse(source, length); + } + + //! Constructor with user-supplied tokens. + /*! + This constructor let user supplies const array of tokens. + This prevents the parsing process and eliminates allocation. + This is preferred for memory constrained environments. + + \param tokens An constant array of tokens representing the JSON pointer. + \param tokenCount Number of tokens. + + \b Example + \code + #define NAME(s) { s, sizeof(s) / sizeof(s[0]) - 1, kPointerInvalidIndex } + #define INDEX(i) { #i, sizeof(#i) - 1, i } + + static const Pointer::Token kTokens[] = { NAME("foo"), INDEX(123) }; + static const Pointer p(kTokens, sizeof(kTokens) / sizeof(kTokens[0])); + // Equivalent to static const Pointer p("/foo/123"); + + #undef NAME + #undef INDEX + \endcode + */ + GenericPointer(const Token* tokens, size_t tokenCount) : allocator_(), ownAllocator_(), nameBuffer_(), tokens_(const_cast(tokens)), tokenCount_(tokenCount), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) {} + + //! Copy constructor. + GenericPointer(const GenericPointer& rhs) : allocator_(), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) { + *this = rhs; + } + + //! Copy constructor. + GenericPointer(const GenericPointer& rhs, Allocator* allocator) : allocator_(allocator), ownAllocator_(), nameBuffer_(), tokens_(), tokenCount_(), parseErrorOffset_(), parseErrorCode_(kPointerParseErrorNone) { + *this = rhs; + } + + //! Destructor. + ~GenericPointer() { + if (nameBuffer_) // If user-supplied tokens constructor is used, nameBuffer_ is nullptr and tokens_ are not deallocated. + Allocator::Free(tokens_); + RAPIDJSON_DELETE(ownAllocator_); + } + + //! Assignment operator. + GenericPointer& operator=(const GenericPointer& rhs) { + if (this != &rhs) { + // Do not delete ownAllcator + if (nameBuffer_) + Allocator::Free(tokens_); + + tokenCount_ = rhs.tokenCount_; + parseErrorOffset_ = rhs.parseErrorOffset_; + parseErrorCode_ = rhs.parseErrorCode_; + + if (rhs.nameBuffer_) + CopyFromRaw(rhs); // Normally parsed tokens. + else { + tokens_ = rhs.tokens_; // User supplied const tokens. + nameBuffer_ = 0; + } + } + return *this; + } + + //! Swap the content of this pointer with an other. + /*! + \param other The pointer to swap with. + \note Constant complexity. + */ + GenericPointer& Swap(GenericPointer& other) RAPIDJSON_NOEXCEPT { + internal::Swap(allocator_, other.allocator_); + internal::Swap(ownAllocator_, other.ownAllocator_); + internal::Swap(nameBuffer_, other.nameBuffer_); + internal::Swap(tokens_, other.tokens_); + internal::Swap(tokenCount_, other.tokenCount_); + internal::Swap(parseErrorOffset_, other.parseErrorOffset_); + internal::Swap(parseErrorCode_, other.parseErrorCode_); + return *this; + } + + //! free-standing swap function helper + /*! + Helper function to enable support for common swap implementation pattern based on \c std::swap: + \code + void swap(MyClass& a, MyClass& b) { + using std::swap; + swap(a.pointer, b.pointer); + // ... + } + \endcode + \see Swap() + */ + friend inline void swap(GenericPointer& a, GenericPointer& b) RAPIDJSON_NOEXCEPT { a.Swap(b); } + + //@} + + //!@name Append token + //@{ + + //! Append a token and return a new Pointer + /*! + \param token Token to be appended. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + GenericPointer Append(const Token& token, Allocator* allocator = 0) const { + GenericPointer r; + r.allocator_ = allocator; + Ch *p = r.CopyFromRaw(*this, 1, token.length + 1); + std::memcpy(p, token.name, (token.length + 1) * sizeof(Ch)); + r.tokens_[tokenCount_].name = p; + r.tokens_[tokenCount_].length = token.length; + r.tokens_[tokenCount_].index = token.index; + return r; + } + + //! Append a name token with length, and return a new Pointer + /*! + \param name Name to be appended. + \param length Length of name. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + GenericPointer Append(const Ch* name, SizeType length, Allocator* allocator = 0) const { + Token token = { name, length, kPointerInvalidIndex }; + return Append(token, allocator); + } + + //! Append a name token without length, and return a new Pointer + /*! + \param name Name (const Ch*) to be appended. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::NotExpr::Type, Ch> >), (GenericPointer)) + Append(T* name, Allocator* allocator = 0) const { + return Append(name, internal::StrLen(name), allocator); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Append a name token, and return a new Pointer + /*! + \param name Name to be appended. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + GenericPointer Append(const std::basic_string& name, Allocator* allocator = 0) const { + return Append(name.c_str(), static_cast(name.size()), allocator); + } +#endif + + //! Append a index token, and return a new Pointer + /*! + \param index Index to be appended. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + GenericPointer Append(SizeType index, Allocator* allocator = 0) const { + char buffer[21]; + char* end = sizeof(SizeType) == 4 ? internal::u32toa(index, buffer) : internal::u64toa(index, buffer); + SizeType length = static_cast(end - buffer); + buffer[length] = '\0'; + + RAPIDJSON_IF_CONSTEXPR (sizeof(Ch) == 1) { + Token token = { reinterpret_cast(buffer), length, index }; + return Append(token, allocator); + } + else { + Ch name[21]; + for (size_t i = 0; i <= length; i++) + name[i] = static_cast(buffer[i]); + Token token = { name, length, index }; + return Append(token, allocator); + } + } + + //! Append a token by value, and return a new Pointer + /*! + \param token token to be appended. + \param allocator Allocator for the newly return Pointer. + \return A new Pointer with appended token. + */ + GenericPointer Append(const ValueType& token, Allocator* allocator = 0) const { + if (token.IsString()) + return Append(token.GetString(), token.GetStringLength(), allocator); + else { + RAPIDJSON_ASSERT(token.IsUint64()); + RAPIDJSON_ASSERT(token.GetUint64() <= SizeType(~0)); + return Append(static_cast(token.GetUint64()), allocator); + } + } + + //!@name Handling Parse Error + //@{ + + //! Check whether this is a valid pointer. + bool IsValid() const { return parseErrorCode_ == kPointerParseErrorNone; } + + //! Get the parsing error offset in code unit. + size_t GetParseErrorOffset() const { return parseErrorOffset_; } + + //! Get the parsing error code. + PointerParseErrorCode GetParseErrorCode() const { return parseErrorCode_; } + + //@} + + //! Get the allocator of this pointer. + Allocator& GetAllocator() { return *allocator_; } + + //!@name Tokens + //@{ + + //! Get the token array (const version only). + const Token* GetTokens() const { return tokens_; } + + //! Get the number of tokens. + size_t GetTokenCount() const { return tokenCount_; } + + //@} + + //!@name Equality/inequality operators + //@{ + + //! Equality operator. + /*! + \note When any pointers are invalid, always returns false. + */ + bool operator==(const GenericPointer& rhs) const { + if (!IsValid() || !rhs.IsValid() || tokenCount_ != rhs.tokenCount_) + return false; + + for (size_t i = 0; i < tokenCount_; i++) { + if (tokens_[i].index != rhs.tokens_[i].index || + tokens_[i].length != rhs.tokens_[i].length || + (tokens_[i].length != 0 && std::memcmp(tokens_[i].name, rhs.tokens_[i].name, sizeof(Ch)* tokens_[i].length) != 0)) + { + return false; + } + } + + return true; + } + + //! Inequality operator. + /*! + \note When any pointers are invalid, always returns true. + */ + bool operator!=(const GenericPointer& rhs) const { return !(*this == rhs); } + + //! Less than operator. + /*! + \note Invalid pointers are always greater than valid ones. + */ + bool operator<(const GenericPointer& rhs) const { + if (!IsValid()) + return false; + if (!rhs.IsValid()) + return true; + + if (tokenCount_ != rhs.tokenCount_) + return tokenCount_ < rhs.tokenCount_; + + for (size_t i = 0; i < tokenCount_; i++) { + if (tokens_[i].index != rhs.tokens_[i].index) + return tokens_[i].index < rhs.tokens_[i].index; + + if (tokens_[i].length != rhs.tokens_[i].length) + return tokens_[i].length < rhs.tokens_[i].length; + + if (int cmp = std::memcmp(tokens_[i].name, rhs.tokens_[i].name, sizeof(Ch) * tokens_[i].length)) + return cmp < 0; + } + + return false; + } + + //@} + + //!@name Stringify + //@{ + + //! Stringify the pointer into string representation. + /*! + \tparam OutputStream Type of output stream. + \param os The output stream. + */ + template + bool Stringify(OutputStream& os) const { + return Stringify(os); + } + + //! Stringify the pointer into URI fragment representation. + /*! + \tparam OutputStream Type of output stream. + \param os The output stream. + */ + template + bool StringifyUriFragment(OutputStream& os) const { + return Stringify(os); + } + + //@} + + //!@name Create value + //@{ + + //! Create a value in a subtree. + /*! + If the value is not exist, it creates all parent values and a JSON Null value. + So it always succeed and return the newly created or existing value. + + Remind that it may change types of parents according to tokens, so it + potentially removes previously stored values. For example, if a document + was an array, and "/foo" is used to create a value, then the document + will be changed to an object, and all existing array elements are lost. + + \param root Root value of a DOM subtree to be resolved. It can be any value other than document root. + \param allocator Allocator for creating the values if the specified value or its parents are not exist. + \param alreadyExist If non-null, it stores whether the resolved value is already exist. + \return The resolved newly created (a JSON Null value), or already exists value. + */ + ValueType& Create(ValueType& root, typename ValueType::AllocatorType& allocator, bool* alreadyExist = 0) const { + RAPIDJSON_ASSERT(IsValid()); + ValueType* v = &root; + bool exist = true; + for (const Token *t = tokens_; t != tokens_ + tokenCount_; ++t) { + if (v->IsArray() && t->name[0] == '-' && t->length == 1) { + v->PushBack(ValueType().Move(), allocator); + v = &((*v)[v->Size() - 1]); + exist = false; + } + else { + if (t->index == kPointerInvalidIndex) { // must be object name + if (!v->IsObject()) + v->SetObject(); // Change to Object + } + else { // object name or array index + if (!v->IsArray() && !v->IsObject()) + v->SetArray(); // Change to Array + } + + if (v->IsArray()) { + if (t->index >= v->Size()) { + v->Reserve(t->index + 1, allocator); + while (t->index >= v->Size()) + v->PushBack(ValueType().Move(), allocator); + exist = false; + } + v = &((*v)[t->index]); + } + else { + typename ValueType::MemberIterator m = v->FindMember(GenericValue(GenericStringRef(t->name, t->length))); + if (m == v->MemberEnd()) { + v->AddMember(ValueType(t->name, t->length, allocator).Move(), ValueType().Move(), allocator); + m = v->MemberEnd(); + v = &(--m)->value; // Assumes AddMember() appends at the end + exist = false; + } + else + v = &m->value; + } + } + } + + if (alreadyExist) + *alreadyExist = exist; + + return *v; + } + + //! Creates a value in a document. + /*! + \param document A document to be resolved. + \param alreadyExist If non-null, it stores whether the resolved value is already exist. + \return The resolved newly created, or already exists value. + */ + template + ValueType& Create(GenericDocument& document, bool* alreadyExist = 0) const { + return Create(document, document.GetAllocator(), alreadyExist); + } + + //@} + + //!@name Compute URI + //@{ + + //! Compute the in-scope URI for a subtree. + // For use with JSON pointers into JSON schema documents. + /*! + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \param rootUri Root URI + \param unresolvedTokenIndex If the pointer cannot resolve a token in the pointer, this parameter can obtain the index of unresolved token. + \param allocator Allocator for Uris + \return Uri if it can be resolved. Otherwise null. + + \note + There are only 3 situations when a URI cannot be resolved: + 1. A value in the path is not an array nor object. + 2. An object value does not contain the token. + 3. A token is out of range of an array value. + + Use unresolvedTokenIndex to retrieve the token index. + */ + UriType GetUri(ValueType& root, const UriType& rootUri, size_t* unresolvedTokenIndex = 0, Allocator* allocator = 0) const { + static const Ch kIdString[] = { 'i', 'd', '\0' }; + static const ValueType kIdValue(kIdString, 2); + UriType base = UriType(rootUri, allocator); + RAPIDJSON_ASSERT(IsValid()); + ValueType* v = &root; + for (const Token *t = tokens_; t != tokens_ + tokenCount_; ++t) { + switch (v->GetType()) { + case kObjectType: + { + // See if we have an id, and if so resolve with the current base + typename ValueType::MemberIterator m = v->FindMember(kIdValue); + if (m != v->MemberEnd() && (m->value).IsString()) { + UriType here = UriType(m->value, allocator).Resolve(base, allocator); + base = here; + } + m = v->FindMember(GenericValue(GenericStringRef(t->name, t->length))); + if (m == v->MemberEnd()) + break; + v = &m->value; + } + continue; + case kArrayType: + if (t->index == kPointerInvalidIndex || t->index >= v->Size()) + break; + v = &((*v)[t->index]); + continue; + default: + break; + } + + // Error: unresolved token + if (unresolvedTokenIndex) + *unresolvedTokenIndex = static_cast(t - tokens_); + return UriType(allocator); + } + return base; + } + + UriType GetUri(const ValueType& root, const UriType& rootUri, size_t* unresolvedTokenIndex = 0, Allocator* allocator = 0) const { + return GetUri(const_cast(root), rootUri, unresolvedTokenIndex, allocator); + } + + + //!@name Query value + //@{ + + //! Query a value in a subtree. + /*! + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \param unresolvedTokenIndex If the pointer cannot resolve a token in the pointer, this parameter can obtain the index of unresolved token. + \return Pointer to the value if it can be resolved. Otherwise null. + + \note + There are only 3 situations when a value cannot be resolved: + 1. A value in the path is not an array nor object. + 2. An object value does not contain the token. + 3. A token is out of range of an array value. + + Use unresolvedTokenIndex to retrieve the token index. + */ + ValueType* Get(ValueType& root, size_t* unresolvedTokenIndex = 0) const { + RAPIDJSON_ASSERT(IsValid()); + ValueType* v = &root; + for (const Token *t = tokens_; t != tokens_ + tokenCount_; ++t) { + switch (v->GetType()) { + case kObjectType: + { + typename ValueType::MemberIterator m = v->FindMember(GenericValue(GenericStringRef(t->name, t->length))); + if (m == v->MemberEnd()) + break; + v = &m->value; + } + continue; + case kArrayType: + if (t->index == kPointerInvalidIndex || t->index >= v->Size()) + break; + v = &((*v)[t->index]); + continue; + default: + break; + } + + // Error: unresolved token + if (unresolvedTokenIndex) + *unresolvedTokenIndex = static_cast(t - tokens_); + return 0; + } + return v; + } + + //! Query a const value in a const subtree. + /*! + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \return Pointer to the value if it can be resolved. Otherwise null. + */ + const ValueType* Get(const ValueType& root, size_t* unresolvedTokenIndex = 0) const { + return Get(const_cast(root), unresolvedTokenIndex); + } + + //@} + + //!@name Query a value with default + //@{ + + //! Query a value in a subtree with default value. + /*! + Similar to Get(), but if the specified value do not exists, it creates all parents and clone the default value. + So that this function always succeed. + + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \param defaultValue Default value to be cloned if the value was not exists. + \param allocator Allocator for creating the values if the specified value or its parents are not exist. + \see Create() + */ + ValueType& GetWithDefault(ValueType& root, const ValueType& defaultValue, typename ValueType::AllocatorType& allocator) const { + bool alreadyExist; + ValueType& v = Create(root, allocator, &alreadyExist); + return alreadyExist ? v : v.CopyFrom(defaultValue, allocator); + } + + //! Query a value in a subtree with default null-terminated string. + ValueType& GetWithDefault(ValueType& root, const Ch* defaultValue, typename ValueType::AllocatorType& allocator) const { + bool alreadyExist; + ValueType& v = Create(root, allocator, &alreadyExist); + return alreadyExist ? v : v.SetString(defaultValue, allocator); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Query a value in a subtree with default std::basic_string. + ValueType& GetWithDefault(ValueType& root, const std::basic_string& defaultValue, typename ValueType::AllocatorType& allocator) const { + bool alreadyExist; + ValueType& v = Create(root, allocator, &alreadyExist); + return alreadyExist ? v : v.SetString(defaultValue, allocator); + } +#endif + + //! Query a value in a subtree with default primitive value. + /*! + \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t, \c bool + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (ValueType&)) + GetWithDefault(ValueType& root, T defaultValue, typename ValueType::AllocatorType& allocator) const { + return GetWithDefault(root, ValueType(defaultValue).Move(), allocator); + } + + //! Query a value in a document with default value. + template + ValueType& GetWithDefault(GenericDocument& document, const ValueType& defaultValue) const { + return GetWithDefault(document, defaultValue, document.GetAllocator()); + } + + //! Query a value in a document with default null-terminated string. + template + ValueType& GetWithDefault(GenericDocument& document, const Ch* defaultValue) const { + return GetWithDefault(document, defaultValue, document.GetAllocator()); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Query a value in a document with default std::basic_string. + template + ValueType& GetWithDefault(GenericDocument& document, const std::basic_string& defaultValue) const { + return GetWithDefault(document, defaultValue, document.GetAllocator()); + } +#endif + + //! Query a value in a document with default primitive value. + /*! + \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t, \c bool + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (ValueType&)) + GetWithDefault(GenericDocument& document, T defaultValue) const { + return GetWithDefault(document, defaultValue, document.GetAllocator()); + } + + //@} + + //!@name Set a value + //@{ + + //! Set a value in a subtree, with move semantics. + /*! + It creates all parents if they are not exist or types are different to the tokens. + So this function always succeeds but potentially remove existing values. + + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \param value Value to be set. + \param allocator Allocator for creating the values if the specified value or its parents are not exist. + \see Create() + */ + ValueType& Set(ValueType& root, ValueType& value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator) = value; + } + + //! Set a value in a subtree, with copy semantics. + ValueType& Set(ValueType& root, const ValueType& value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator).CopyFrom(value, allocator); + } + + //! Set a null-terminated string in a subtree. + ValueType& Set(ValueType& root, const Ch* value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator) = ValueType(value, allocator).Move(); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Set a std::basic_string in a subtree. + ValueType& Set(ValueType& root, const std::basic_string& value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator) = ValueType(value, allocator).Move(); + } +#endif + + //! Set a primitive value in a subtree. + /*! + \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t, \c bool + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (ValueType&)) + Set(ValueType& root, T value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator) = ValueType(value).Move(); + } + + //! Set a value in a document, with move semantics. + template + ValueType& Set(GenericDocument& document, ValueType& value) const { + return Create(document) = value; + } + + //! Set a value in a document, with copy semantics. + template + ValueType& Set(GenericDocument& document, const ValueType& value) const { + return Create(document).CopyFrom(value, document.GetAllocator()); + } + + //! Set a null-terminated string in a document. + template + ValueType& Set(GenericDocument& document, const Ch* value) const { + return Create(document) = ValueType(value, document.GetAllocator()).Move(); + } + +#if RAPIDJSON_HAS_STDSTRING + //! Sets a std::basic_string in a document. + template + ValueType& Set(GenericDocument& document, const std::basic_string& value) const { + return Create(document) = ValueType(value, document.GetAllocator()).Move(); + } +#endif + + //! Set a primitive value in a document. + /*! + \tparam T Either \ref Type, \c int, \c unsigned, \c int64_t, \c uint64_t, \c bool + */ + template + RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (ValueType&)) + Set(GenericDocument& document, T value) const { + return Create(document) = value; + } + + //@} + + //!@name Swap a value + //@{ + + //! Swap a value with a value in a subtree. + /*! + It creates all parents if they are not exist or types are different to the tokens. + So this function always succeeds but potentially remove existing values. + + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \param value Value to be swapped. + \param allocator Allocator for creating the values if the specified value or its parents are not exist. + \see Create() + */ + ValueType& Swap(ValueType& root, ValueType& value, typename ValueType::AllocatorType& allocator) const { + return Create(root, allocator).Swap(value); + } + + //! Swap a value with a value in a document. + template + ValueType& Swap(GenericDocument& document, ValueType& value) const { + return Create(document).Swap(value); + } + + //@} + + //! Erase a value in a subtree. + /*! + \param root Root value of a DOM sub-tree to be resolved. It can be any value other than document root. + \return Whether the resolved value is found and erased. + + \note Erasing with an empty pointer \c Pointer(""), i.e. the root, always fail and return false. + */ + bool Erase(ValueType& root) const { + RAPIDJSON_ASSERT(IsValid()); + if (tokenCount_ == 0) // Cannot erase the root + return false; + + ValueType* v = &root; + const Token* last = tokens_ + (tokenCount_ - 1); + for (const Token *t = tokens_; t != last; ++t) { + switch (v->GetType()) { + case kObjectType: + { + typename ValueType::MemberIterator m = v->FindMember(GenericValue(GenericStringRef(t->name, t->length))); + if (m == v->MemberEnd()) + return false; + v = &m->value; + } + break; + case kArrayType: + if (t->index == kPointerInvalidIndex || t->index >= v->Size()) + return false; + v = &((*v)[t->index]); + break; + default: + return false; + } + } + + switch (v->GetType()) { + case kObjectType: + return v->EraseMember(GenericStringRef(last->name, last->length)); + case kArrayType: + if (last->index == kPointerInvalidIndex || last->index >= v->Size()) + return false; + v->Erase(v->Begin() + last->index); + return true; + default: + return false; + } + } + +private: + //! Clone the content from rhs to this. + /*! + \param rhs Source pointer. + \param extraToken Extra tokens to be allocated. + \param extraNameBufferSize Extra name buffer size (in number of Ch) to be allocated. + \return Start of non-occupied name buffer, for storing extra names. + */ + Ch* CopyFromRaw(const GenericPointer& rhs, size_t extraToken = 0, size_t extraNameBufferSize = 0) { + if (!allocator_) // allocator is independently owned. + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + + size_t nameBufferSize = rhs.tokenCount_; // null terminators for tokens + for (Token *t = rhs.tokens_; t != rhs.tokens_ + rhs.tokenCount_; ++t) + nameBufferSize += t->length; + + tokenCount_ = rhs.tokenCount_ + extraToken; + tokens_ = static_cast(allocator_->Malloc(tokenCount_ * sizeof(Token) + (nameBufferSize + extraNameBufferSize) * sizeof(Ch))); + nameBuffer_ = reinterpret_cast(tokens_ + tokenCount_); + if (rhs.tokenCount_ > 0) { + std::memcpy(tokens_, rhs.tokens_, rhs.tokenCount_ * sizeof(Token)); + } + if (nameBufferSize > 0) { + std::memcpy(nameBuffer_, rhs.nameBuffer_, nameBufferSize * sizeof(Ch)); + } + + // The names of each token point to a string in the nameBuffer_. The + // previous memcpy copied over string pointers into the rhs.nameBuffer_, + // but they should point to the strings in the new nameBuffer_. + for (size_t i = 0; i < rhs.tokenCount_; ++i) { + // The offset between the string address and the name buffer should + // still be constant, so we can just get this offset and set each new + // token name according the new buffer start + the known offset. + std::ptrdiff_t name_offset = rhs.tokens_[i].name - rhs.nameBuffer_; + tokens_[i].name = nameBuffer_ + name_offset; + } + + return nameBuffer_ + nameBufferSize; + } + + //! Check whether a character should be percent-encoded. + /*! + According to RFC 3986 2.3 Unreserved Characters. + \param c The character (code unit) to be tested. + */ + bool NeedPercentEncode(Ch c) const { + return !((c >= '0' && c <= '9') || (c >= 'A' && c <='Z') || (c >= 'a' && c <= 'z') || c == '-' || c == '.' || c == '_' || c =='~'); + } + + //! Parse a JSON String or its URI fragment representation into tokens. +#ifndef __clang__ // -Wdocumentation + /*! + \param source Either a JSON Pointer string, or its URI fragment representation. Not need to be null terminated. + \param length Length of the source string. + \note Source cannot be JSON String Representation of JSON Pointer, e.g. In "/\u0000", \u0000 will not be unescaped. + */ +#endif + void Parse(const Ch* source, size_t length) { + RAPIDJSON_ASSERT(source != NULL); + RAPIDJSON_ASSERT(nameBuffer_ == 0); + RAPIDJSON_ASSERT(tokens_ == 0); + + // Create own allocator if user did not supply. + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + + // Count number of '/' as tokenCount + tokenCount_ = 0; + for (const Ch* s = source; s != source + length; s++) + if (*s == '/') + tokenCount_++; + + Token* token = tokens_ = static_cast(allocator_->Malloc(tokenCount_ * sizeof(Token) + length * sizeof(Ch))); + Ch* name = nameBuffer_ = reinterpret_cast(tokens_ + tokenCount_); + size_t i = 0; + + // Detect if it is a URI fragment + bool uriFragment = false; + if (source[i] == '#') { + uriFragment = true; + i++; + } + + if (i != length && source[i] != '/') { + parseErrorCode_ = kPointerParseErrorTokenMustBeginWithSolidus; + goto error; + } + + while (i < length) { + RAPIDJSON_ASSERT(source[i] == '/'); + i++; // consumes '/' + + token->name = name; + bool isNumber = true; + + while (i < length && source[i] != '/') { + Ch c = source[i]; + if (uriFragment) { + // Decoding percent-encoding for URI fragment + if (c == '%') { + PercentDecodeStream is(&source[i], source + length); + GenericInsituStringStream os(name); + Ch* begin = os.PutBegin(); + if (!Transcoder, EncodingType>().Validate(is, os) || !is.IsValid()) { + parseErrorCode_ = kPointerParseErrorInvalidPercentEncoding; + goto error; + } + size_t len = os.PutEnd(begin); + i += is.Tell() - 1; + if (len == 1) + c = *name; + else { + name += len; + isNumber = false; + i++; + continue; + } + } + else if (NeedPercentEncode(c)) { + parseErrorCode_ = kPointerParseErrorCharacterMustPercentEncode; + goto error; + } + } + + i++; + + // Escaping "~0" -> '~', "~1" -> '/' + if (c == '~') { + if (i < length) { + c = source[i]; + if (c == '0') c = '~'; + else if (c == '1') c = '/'; + else { + parseErrorCode_ = kPointerParseErrorInvalidEscape; + goto error; + } + i++; + } + else { + parseErrorCode_ = kPointerParseErrorInvalidEscape; + goto error; + } + } + + // First check for index: all of characters are digit + if (c < '0' || c > '9') + isNumber = false; + + *name++ = c; + } + token->length = static_cast(name - token->name); + if (token->length == 0) + isNumber = false; + *name++ = '\0'; // Null terminator + + // Second check for index: more than one digit cannot have leading zero + if (isNumber && token->length > 1 && token->name[0] == '0') + isNumber = false; + + // String to SizeType conversion + SizeType n = 0; + if (isNumber) { + for (size_t j = 0; j < token->length; j++) { + SizeType m = n * 10 + static_cast(token->name[j] - '0'); + if (m < n) { // overflow detection + isNumber = false; + break; + } + n = m; + } + } + + token->index = isNumber ? n : kPointerInvalidIndex; + token++; + } + + RAPIDJSON_ASSERT(name <= nameBuffer_ + length); // Should not overflow buffer + parseErrorCode_ = kPointerParseErrorNone; + return; + + error: + Allocator::Free(tokens_); + nameBuffer_ = 0; + tokens_ = 0; + tokenCount_ = 0; + parseErrorOffset_ = i; + return; + } + + //! Stringify to string or URI fragment representation. + /*! + \tparam uriFragment True for stringifying to URI fragment representation. False for string representation. + \tparam OutputStream type of output stream. + \param os The output stream. + */ + template + bool Stringify(OutputStream& os) const { + RAPIDJSON_ASSERT(IsValid()); + + if (uriFragment) + os.Put('#'); + + for (Token *t = tokens_; t != tokens_ + tokenCount_; ++t) { + os.Put('/'); + for (size_t j = 0; j < t->length; j++) { + Ch c = t->name[j]; + if (c == '~') { + os.Put('~'); + os.Put('0'); + } + else if (c == '/') { + os.Put('~'); + os.Put('1'); + } + else if (uriFragment && NeedPercentEncode(c)) { + // Transcode to UTF8 sequence + GenericStringStream source(&t->name[j]); + PercentEncodeStream target(os); + if (!Transcoder >().Validate(source, target)) + return false; + j += source.Tell() - 1; + } + else + os.Put(c); + } + } + return true; + } + + //! A helper stream for decoding a percent-encoded sequence into code unit. + /*! + This stream decodes %XY triplet into code unit (0-255). + If it encounters invalid characters, it sets output code unit as 0 and + mark invalid, and to be checked by IsValid(). + */ + class PercentDecodeStream { + public: + typedef typename ValueType::Ch Ch; + + //! Constructor + /*! + \param source Start of the stream + \param end Past-the-end of the stream. + */ + PercentDecodeStream(const Ch* source, const Ch* end) : src_(source), head_(source), end_(end), valid_(true) {} + + Ch Take() { + if (*src_ != '%' || src_ + 3 > end_) { // %XY triplet + valid_ = false; + return 0; + } + src_++; + Ch c = 0; + for (int j = 0; j < 2; j++) { + c = static_cast(c << 4); + Ch h = *src_; + if (h >= '0' && h <= '9') c = static_cast(c + h - '0'); + else if (h >= 'A' && h <= 'F') c = static_cast(c + h - 'A' + 10); + else if (h >= 'a' && h <= 'f') c = static_cast(c + h - 'a' + 10); + else { + valid_ = false; + return 0; + } + src_++; + } + return c; + } + + size_t Tell() const { return static_cast(src_ - head_); } + bool IsValid() const { return valid_; } + + private: + const Ch* src_; //!< Current read position. + const Ch* head_; //!< Original head of the string. + const Ch* end_; //!< Past-the-end position. + bool valid_; //!< Whether the parsing is valid. + }; + + //! A helper stream to encode character (UTF-8 code unit) into percent-encoded sequence. + template + class PercentEncodeStream { + public: + PercentEncodeStream(OutputStream& os) : os_(os) {} + void Put(char c) { // UTF-8 must be byte + unsigned char u = static_cast(c); + static const char hexDigits[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' }; + os_.Put('%'); + os_.Put(static_cast(hexDigits[u >> 4])); + os_.Put(static_cast(hexDigits[u & 15])); + } + private: + OutputStream& os_; + }; + + Allocator* allocator_; //!< The current allocator. It is either user-supplied or equal to ownAllocator_. + Allocator* ownAllocator_; //!< Allocator owned by this Pointer. + Ch* nameBuffer_; //!< A buffer containing all names in tokens. + Token* tokens_; //!< A list of tokens. + size_t tokenCount_; //!< Number of tokens in tokens_. + size_t parseErrorOffset_; //!< Offset in code unit when parsing fail. + PointerParseErrorCode parseErrorCode_; //!< Parsing error code. +}; + +//! GenericPointer for Value (UTF-8, default allocator). +typedef GenericPointer Pointer; + +//!@name Helper functions for GenericPointer +//@{ + +////////////////////////////////////////////////////////////////////////////// + +template +typename T::ValueType& CreateValueByPointer(T& root, const GenericPointer& pointer, typename T::AllocatorType& a) { + return pointer.Create(root, a); +} + +template +typename T::ValueType& CreateValueByPointer(T& root, const CharType(&source)[N], typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Create(root, a); +} + +// No allocator parameter + +template +typename DocumentType::ValueType& CreateValueByPointer(DocumentType& document, const GenericPointer& pointer) { + return pointer.Create(document); +} + +template +typename DocumentType::ValueType& CreateValueByPointer(DocumentType& document, const CharType(&source)[N]) { + return GenericPointer(source, N - 1).Create(document); +} + +////////////////////////////////////////////////////////////////////////////// + +template +typename T::ValueType* GetValueByPointer(T& root, const GenericPointer& pointer, size_t* unresolvedTokenIndex = 0) { + return pointer.Get(root, unresolvedTokenIndex); +} + +template +const typename T::ValueType* GetValueByPointer(const T& root, const GenericPointer& pointer, size_t* unresolvedTokenIndex = 0) { + return pointer.Get(root, unresolvedTokenIndex); +} + +template +typename T::ValueType* GetValueByPointer(T& root, const CharType (&source)[N], size_t* unresolvedTokenIndex = 0) { + return GenericPointer(source, N - 1).Get(root, unresolvedTokenIndex); +} + +template +const typename T::ValueType* GetValueByPointer(const T& root, const CharType(&source)[N], size_t* unresolvedTokenIndex = 0) { + return GenericPointer(source, N - 1).Get(root, unresolvedTokenIndex); +} + +////////////////////////////////////////////////////////////////////////////// + +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const GenericPointer& pointer, const typename T::ValueType& defaultValue, typename T::AllocatorType& a) { + return pointer.GetWithDefault(root, defaultValue, a); +} + +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const GenericPointer& pointer, const typename T::Ch* defaultValue, typename T::AllocatorType& a) { + return pointer.GetWithDefault(root, defaultValue, a); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const GenericPointer& pointer, const std::basic_string& defaultValue, typename T::AllocatorType& a) { + return pointer.GetWithDefault(root, defaultValue, a); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename T::ValueType&)) +GetValueByPointerWithDefault(T& root, const GenericPointer& pointer, T2 defaultValue, typename T::AllocatorType& a) { + return pointer.GetWithDefault(root, defaultValue, a); +} + +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const CharType(&source)[N], const typename T::ValueType& defaultValue, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).GetWithDefault(root, defaultValue, a); +} + +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const CharType(&source)[N], const typename T::Ch* defaultValue, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).GetWithDefault(root, defaultValue, a); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename T::ValueType& GetValueByPointerWithDefault(T& root, const CharType(&source)[N], const std::basic_string& defaultValue, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).GetWithDefault(root, defaultValue, a); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename T::ValueType&)) +GetValueByPointerWithDefault(T& root, const CharType(&source)[N], T2 defaultValue, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).GetWithDefault(root, defaultValue, a); +} + +// No allocator parameter + +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const GenericPointer& pointer, const typename DocumentType::ValueType& defaultValue) { + return pointer.GetWithDefault(document, defaultValue); +} + +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const GenericPointer& pointer, const typename DocumentType::Ch* defaultValue) { + return pointer.GetWithDefault(document, defaultValue); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const GenericPointer& pointer, const std::basic_string& defaultValue) { + return pointer.GetWithDefault(document, defaultValue); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename DocumentType::ValueType&)) +GetValueByPointerWithDefault(DocumentType& document, const GenericPointer& pointer, T2 defaultValue) { + return pointer.GetWithDefault(document, defaultValue); +} + +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const CharType(&source)[N], const typename DocumentType::ValueType& defaultValue) { + return GenericPointer(source, N - 1).GetWithDefault(document, defaultValue); +} + +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const CharType(&source)[N], const typename DocumentType::Ch* defaultValue) { + return GenericPointer(source, N - 1).GetWithDefault(document, defaultValue); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename DocumentType::ValueType& GetValueByPointerWithDefault(DocumentType& document, const CharType(&source)[N], const std::basic_string& defaultValue) { + return GenericPointer(source, N - 1).GetWithDefault(document, defaultValue); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename DocumentType::ValueType&)) +GetValueByPointerWithDefault(DocumentType& document, const CharType(&source)[N], T2 defaultValue) { + return GenericPointer(source, N - 1).GetWithDefault(document, defaultValue); +} + +////////////////////////////////////////////////////////////////////////////// + +template +typename T::ValueType& SetValueByPointer(T& root, const GenericPointer& pointer, typename T::ValueType& value, typename T::AllocatorType& a) { + return pointer.Set(root, value, a); +} + +template +typename T::ValueType& SetValueByPointer(T& root, const GenericPointer& pointer, const typename T::ValueType& value, typename T::AllocatorType& a) { + return pointer.Set(root, value, a); +} + +template +typename T::ValueType& SetValueByPointer(T& root, const GenericPointer& pointer, const typename T::Ch* value, typename T::AllocatorType& a) { + return pointer.Set(root, value, a); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename T::ValueType& SetValueByPointer(T& root, const GenericPointer& pointer, const std::basic_string& value, typename T::AllocatorType& a) { + return pointer.Set(root, value, a); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename T::ValueType&)) +SetValueByPointer(T& root, const GenericPointer& pointer, T2 value, typename T::AllocatorType& a) { + return pointer.Set(root, value, a); +} + +template +typename T::ValueType& SetValueByPointer(T& root, const CharType(&source)[N], typename T::ValueType& value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Set(root, value, a); +} + +template +typename T::ValueType& SetValueByPointer(T& root, const CharType(&source)[N], const typename T::ValueType& value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Set(root, value, a); +} + +template +typename T::ValueType& SetValueByPointer(T& root, const CharType(&source)[N], const typename T::Ch* value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Set(root, value, a); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename T::ValueType& SetValueByPointer(T& root, const CharType(&source)[N], const std::basic_string& value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Set(root, value, a); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename T::ValueType&)) +SetValueByPointer(T& root, const CharType(&source)[N], T2 value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Set(root, value, a); +} + +// No allocator parameter + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const GenericPointer& pointer, typename DocumentType::ValueType& value) { + return pointer.Set(document, value); +} + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const GenericPointer& pointer, const typename DocumentType::ValueType& value) { + return pointer.Set(document, value); +} + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const GenericPointer& pointer, const typename DocumentType::Ch* value) { + return pointer.Set(document, value); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const GenericPointer& pointer, const std::basic_string& value) { + return pointer.Set(document, value); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename DocumentType::ValueType&)) +SetValueByPointer(DocumentType& document, const GenericPointer& pointer, T2 value) { + return pointer.Set(document, value); +} + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const CharType(&source)[N], typename DocumentType::ValueType& value) { + return GenericPointer(source, N - 1).Set(document, value); +} + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const CharType(&source)[N], const typename DocumentType::ValueType& value) { + return GenericPointer(source, N - 1).Set(document, value); +} + +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const CharType(&source)[N], const typename DocumentType::Ch* value) { + return GenericPointer(source, N - 1).Set(document, value); +} + +#if RAPIDJSON_HAS_STDSTRING +template +typename DocumentType::ValueType& SetValueByPointer(DocumentType& document, const CharType(&source)[N], const std::basic_string& value) { + return GenericPointer(source, N - 1).Set(document, value); +} +#endif + +template +RAPIDJSON_DISABLEIF_RETURN((internal::OrExpr, internal::IsGenericValue >), (typename DocumentType::ValueType&)) +SetValueByPointer(DocumentType& document, const CharType(&source)[N], T2 value) { + return GenericPointer(source, N - 1).Set(document, value); +} + +////////////////////////////////////////////////////////////////////////////// + +template +typename T::ValueType& SwapValueByPointer(T& root, const GenericPointer& pointer, typename T::ValueType& value, typename T::AllocatorType& a) { + return pointer.Swap(root, value, a); +} + +template +typename T::ValueType& SwapValueByPointer(T& root, const CharType(&source)[N], typename T::ValueType& value, typename T::AllocatorType& a) { + return GenericPointer(source, N - 1).Swap(root, value, a); +} + +template +typename DocumentType::ValueType& SwapValueByPointer(DocumentType& document, const GenericPointer& pointer, typename DocumentType::ValueType& value) { + return pointer.Swap(document, value); +} + +template +typename DocumentType::ValueType& SwapValueByPointer(DocumentType& document, const CharType(&source)[N], typename DocumentType::ValueType& value) { + return GenericPointer(source, N - 1).Swap(document, value); +} + +////////////////////////////////////////////////////////////////////////////// + +template +bool EraseValueByPointer(T& root, const GenericPointer& pointer) { + return pointer.Erase(root); +} + +template +bool EraseValueByPointer(T& root, const CharType(&source)[N]) { + return GenericPointer(source, N - 1).Erase(root); +} + +//@} + +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) || defined(_MSC_VER) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_POINTER_H_ diff --git a/include/rapidjson/prettywriter.h b/include/rapidjson/prettywriter.h new file mode 100644 index 0000000..fe45df1 --- /dev/null +++ b/include/rapidjson/prettywriter.h @@ -0,0 +1,277 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_PRETTYWRITER_H_ +#define RAPIDJSON_PRETTYWRITER_H_ + +#include "writer.h" + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#if defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(c++98-compat) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Combination of PrettyWriter format flags. +/*! \see PrettyWriter::SetFormatOptions + */ +enum PrettyFormatOptions { + kFormatDefault = 0, //!< Default pretty formatting. + kFormatSingleLineArray = 1 //!< Format arrays on a single line. +}; + +//! Writer with indentation and spacing. +/*! + \tparam OutputStream Type of output os. + \tparam SourceEncoding Encoding of source string. + \tparam TargetEncoding Encoding of output stream. + \tparam StackAllocator Type of allocator for allocating memory of stack. +*/ +template, typename TargetEncoding = UTF8<>, typename StackAllocator = CrtAllocator, unsigned writeFlags = kWriteDefaultFlags> +class PrettyWriter : public Writer { +public: + typedef Writer Base; + typedef typename Base::Ch Ch; + + //! Constructor + /*! \param os Output stream. + \param allocator User supplied allocator. If it is null, it will create a private one. + \param levelDepth Initial capacity of stack. + */ + explicit PrettyWriter(OutputStream& os, StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : + Base(os, allocator, levelDepth), indentChar_(' '), indentCharCount_(4), formatOptions_(kFormatDefault) {} + + + explicit PrettyWriter(StackAllocator* allocator = 0, size_t levelDepth = Base::kDefaultLevelDepth) : + Base(allocator, levelDepth), indentChar_(' '), indentCharCount_(4), formatOptions_(kFormatDefault) {} + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + PrettyWriter(PrettyWriter&& rhs) : + Base(std::forward(rhs)), indentChar_(rhs.indentChar_), indentCharCount_(rhs.indentCharCount_), formatOptions_(rhs.formatOptions_) {} +#endif + + //! Set custom indentation. + /*! \param indentChar Character for indentation. Must be whitespace character (' ', '\\t', '\\n', '\\r'). + \param indentCharCount Number of indent characters for each indentation level. + \note The default indentation is 4 spaces. + */ + PrettyWriter& SetIndent(Ch indentChar, unsigned indentCharCount) { + RAPIDJSON_ASSERT(indentChar == ' ' || indentChar == '\t' || indentChar == '\n' || indentChar == '\r'); + indentChar_ = indentChar; + indentCharCount_ = indentCharCount; + return *this; + } + + //! Set pretty writer formatting options. + /*! \param options Formatting options. + */ + PrettyWriter& SetFormatOptions(PrettyFormatOptions options) { + formatOptions_ = options; + return *this; + } + + /*! @name Implementation of Handler + \see Handler + */ + //@{ + + bool Null() { PrettyPrefix(kNullType); return Base::EndValue(Base::WriteNull()); } + bool Bool(bool b) { PrettyPrefix(b ? kTrueType : kFalseType); return Base::EndValue(Base::WriteBool(b)); } + bool Int(int i) { PrettyPrefix(kNumberType); return Base::EndValue(Base::WriteInt(i)); } + bool Uint(unsigned u) { PrettyPrefix(kNumberType); return Base::EndValue(Base::WriteUint(u)); } + bool Int64(int64_t i64) { PrettyPrefix(kNumberType); return Base::EndValue(Base::WriteInt64(i64)); } + bool Uint64(uint64_t u64) { PrettyPrefix(kNumberType); return Base::EndValue(Base::WriteUint64(u64)); } + bool Double(double d) { PrettyPrefix(kNumberType); return Base::EndValue(Base::WriteDouble(d)); } + + bool RawNumber(const Ch* str, SizeType length, bool copy = false) { + RAPIDJSON_ASSERT(str != 0); + (void)copy; + PrettyPrefix(kNumberType); + return Base::EndValue(Base::WriteString(str, length)); + } + + bool String(const Ch* str, SizeType length, bool copy = false) { + RAPIDJSON_ASSERT(str != 0); + (void)copy; + PrettyPrefix(kStringType); + return Base::EndValue(Base::WriteString(str, length)); + } + +#if RAPIDJSON_HAS_STDSTRING + bool String(const std::basic_string& str) { + return String(str.data(), SizeType(str.size())); + } +#endif + + bool StartObject() { + PrettyPrefix(kObjectType); + new (Base::level_stack_.template Push()) typename Base::Level(false); + return Base::WriteStartObject(); + } + + bool Key(const Ch* str, SizeType length, bool copy = false) { return String(str, length, copy); } + +#if RAPIDJSON_HAS_STDSTRING + bool Key(const std::basic_string& str) { + return Key(str.data(), SizeType(str.size())); + } +#endif + + bool EndObject(SizeType memberCount = 0) { + (void)memberCount; + RAPIDJSON_ASSERT(Base::level_stack_.GetSize() >= sizeof(typename Base::Level)); // not inside an Object + RAPIDJSON_ASSERT(!Base::level_stack_.template Top()->inArray); // currently inside an Array, not Object + RAPIDJSON_ASSERT(0 == Base::level_stack_.template Top()->valueCount % 2); // Object has a Key without a Value + + bool empty = Base::level_stack_.template Pop(1)->valueCount == 0; + + if (!empty) { + Base::os_->Put('\n'); + WriteIndent(); + } + bool ret = Base::EndValue(Base::WriteEndObject()); + (void)ret; + RAPIDJSON_ASSERT(ret == true); + if (Base::level_stack_.Empty()) // end of json text + Base::Flush(); + return true; + } + + bool StartArray() { + PrettyPrefix(kArrayType); + new (Base::level_stack_.template Push()) typename Base::Level(true); + return Base::WriteStartArray(); + } + + bool EndArray(SizeType memberCount = 0) { + (void)memberCount; + RAPIDJSON_ASSERT(Base::level_stack_.GetSize() >= sizeof(typename Base::Level)); + RAPIDJSON_ASSERT(Base::level_stack_.template Top()->inArray); + bool empty = Base::level_stack_.template Pop(1)->valueCount == 0; + + if (!empty && !(formatOptions_ & kFormatSingleLineArray)) { + Base::os_->Put('\n'); + WriteIndent(); + } + bool ret = Base::EndValue(Base::WriteEndArray()); + (void)ret; + RAPIDJSON_ASSERT(ret == true); + if (Base::level_stack_.Empty()) // end of json text + Base::Flush(); + return true; + } + + //@} + + /*! @name Convenience extensions */ + //@{ + + //! Simpler but slower overload. + bool String(const Ch* str) { return String(str, internal::StrLen(str)); } + bool Key(const Ch* str) { return Key(str, internal::StrLen(str)); } + + //@} + + //! Write a raw JSON value. + /*! + For user to write a stringified JSON as a value. + + \param json A well-formed JSON value. It should not contain null character within [0, length - 1] range. + \param length Length of the json. + \param type Type of the root of json. + \note When using PrettyWriter::RawValue(), the result json may not be indented correctly. + */ + bool RawValue(const Ch* json, size_t length, Type type) { + RAPIDJSON_ASSERT(json != 0); + PrettyPrefix(type); + return Base::EndValue(Base::WriteRawValue(json, length)); + } + +protected: + void PrettyPrefix(Type type) { + (void)type; + if (Base::level_stack_.GetSize() != 0) { // this value is not at root + typename Base::Level* level = Base::level_stack_.template Top(); + + if (level->inArray) { + if (level->valueCount > 0) { + Base::os_->Put(','); // add comma if it is not the first element in array + if (formatOptions_ & kFormatSingleLineArray) + Base::os_->Put(' '); + } + + if (!(formatOptions_ & kFormatSingleLineArray)) { + Base::os_->Put('\n'); + WriteIndent(); + } + } + else { // in object + if (level->valueCount > 0) { + if (level->valueCount % 2 == 0) { + Base::os_->Put(','); + Base::os_->Put('\n'); + } + else { + Base::os_->Put(':'); + Base::os_->Put(' '); + } + } + else + Base::os_->Put('\n'); + + if (level->valueCount % 2 == 0) + WriteIndent(); + } + if (!level->inArray && level->valueCount % 2 == 0) + RAPIDJSON_ASSERT(type == kStringType); // if it's in object, then even number should be a name + level->valueCount++; + } + else { + RAPIDJSON_ASSERT(!Base::hasRoot_); // Should only has one and only one root. + Base::hasRoot_ = true; + } + } + + void WriteIndent() { + size_t count = (Base::level_stack_.GetSize() / sizeof(typename Base::Level)) * indentCharCount_; + PutN(*Base::os_, static_cast(indentChar_), count); + } + + Ch indentChar_; + unsigned indentCharCount_; + PrettyFormatOptions formatOptions_; + +private: + // Prohibit copy constructor & assignment operator. + PrettyWriter(const PrettyWriter&); + PrettyWriter& operator=(const PrettyWriter&); +}; + +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_RAPIDJSON_H_ diff --git a/include/rapidjson/rapidjson.h b/include/rapidjson/rapidjson.h new file mode 100644 index 0000000..247b8e6 --- /dev/null +++ b/include/rapidjson/rapidjson.h @@ -0,0 +1,741 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_RAPIDJSON_H_ +#define RAPIDJSON_RAPIDJSON_H_ + +/*!\file rapidjson.h + \brief common definitions and configuration + + \see RAPIDJSON_CONFIG + */ + +/*! \defgroup RAPIDJSON_CONFIG RapidJSON configuration + \brief Configuration macros for library features + + Some RapidJSON features are configurable to adapt the library to a wide + variety of platforms, environments and usage scenarios. Most of the + features can be configured in terms of overridden or predefined + preprocessor macros at compile-time. + + Some additional customization is available in the \ref RAPIDJSON_ERRORS APIs. + + \note These macros should be given on the compiler command-line + (where applicable) to avoid inconsistent values when compiling + different translation units of a single application. + */ + +#include // malloc(), realloc(), free(), size_t +#include // memset(), memcpy(), memmove(), memcmp() + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_VERSION_STRING +// +// ALWAYS synchronize the following 3 macros with corresponding variables in /CMakeLists.txt. +// + +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +// token stringification +#define RAPIDJSON_STRINGIFY(x) RAPIDJSON_DO_STRINGIFY(x) +#define RAPIDJSON_DO_STRINGIFY(x) #x + +// token concatenation +#define RAPIDJSON_JOIN(X, Y) RAPIDJSON_DO_JOIN(X, Y) +#define RAPIDJSON_DO_JOIN(X, Y) RAPIDJSON_DO_JOIN2(X, Y) +#define RAPIDJSON_DO_JOIN2(X, Y) X##Y +//!@endcond + +/*! \def RAPIDJSON_MAJOR_VERSION + \ingroup RAPIDJSON_CONFIG + \brief Major version of RapidJSON in integer. +*/ +/*! \def RAPIDJSON_MINOR_VERSION + \ingroup RAPIDJSON_CONFIG + \brief Minor version of RapidJSON in integer. +*/ +/*! \def RAPIDJSON_PATCH_VERSION + \ingroup RAPIDJSON_CONFIG + \brief Patch version of RapidJSON in integer. +*/ +/*! \def RAPIDJSON_VERSION_STRING + \ingroup RAPIDJSON_CONFIG + \brief Version of RapidJSON in ".." string format. +*/ +#define RAPIDJSON_MAJOR_VERSION 1 +#define RAPIDJSON_MINOR_VERSION 1 +#define RAPIDJSON_PATCH_VERSION 0 +#define RAPIDJSON_VERSION_STRING \ + RAPIDJSON_STRINGIFY(RAPIDJSON_MAJOR_VERSION.RAPIDJSON_MINOR_VERSION.RAPIDJSON_PATCH_VERSION) + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_NAMESPACE_(BEGIN|END) +/*! \def RAPIDJSON_NAMESPACE + \ingroup RAPIDJSON_CONFIG + \brief provide custom rapidjson namespace + + In order to avoid symbol clashes and/or "One Definition Rule" errors + between multiple inclusions of (different versions of) RapidJSON in + a single binary, users can customize the name of the main RapidJSON + namespace. + + In case of a single nesting level, defining \c RAPIDJSON_NAMESPACE + to a custom name (e.g. \c MyRapidJSON) is sufficient. If multiple + levels are needed, both \ref RAPIDJSON_NAMESPACE_BEGIN and \ref + RAPIDJSON_NAMESPACE_END need to be defined as well: + + \code + // in some .cpp file + #define RAPIDJSON_NAMESPACE my::rapidjson + #define RAPIDJSON_NAMESPACE_BEGIN namespace my { namespace rapidjson { + #define RAPIDJSON_NAMESPACE_END } } + #include "rapidjson/..." + \endcode + + \see rapidjson + */ +/*! \def RAPIDJSON_NAMESPACE_BEGIN + \ingroup RAPIDJSON_CONFIG + \brief provide custom rapidjson namespace (opening expression) + \see RAPIDJSON_NAMESPACE +*/ +/*! \def RAPIDJSON_NAMESPACE_END + \ingroup RAPIDJSON_CONFIG + \brief provide custom rapidjson namespace (closing expression) + \see RAPIDJSON_NAMESPACE +*/ +#ifndef RAPIDJSON_NAMESPACE +#define RAPIDJSON_NAMESPACE rapidjson +#endif +#ifndef RAPIDJSON_NAMESPACE_BEGIN +#define RAPIDJSON_NAMESPACE_BEGIN namespace RAPIDJSON_NAMESPACE { +#endif +#ifndef RAPIDJSON_NAMESPACE_END +#define RAPIDJSON_NAMESPACE_END } +#endif + +/////////////////////////////////////////////////////////////////////////////// +// __cplusplus macro + +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN + +#if defined(_MSC_VER) +#define RAPIDJSON_CPLUSPLUS _MSVC_LANG +#else +#define RAPIDJSON_CPLUSPLUS __cplusplus +#endif + +//!@endcond + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_HAS_STDSTRING + +#ifndef RAPIDJSON_HAS_STDSTRING +#ifdef RAPIDJSON_DOXYGEN_RUNNING +#define RAPIDJSON_HAS_STDSTRING 1 // force generation of documentation +#else +#define RAPIDJSON_HAS_STDSTRING 0 // no std::string support by default +#endif +/*! \def RAPIDJSON_HAS_STDSTRING + \ingroup RAPIDJSON_CONFIG + \brief Enable RapidJSON support for \c std::string + + By defining this preprocessor symbol to \c 1, several convenience functions for using + \ref rapidjson::GenericValue with \c std::string are enabled, especially + for construction and comparison. + + \hideinitializer +*/ +#endif // !defined(RAPIDJSON_HAS_STDSTRING) + +#if RAPIDJSON_HAS_STDSTRING +#include +#endif // RAPIDJSON_HAS_STDSTRING + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_USE_MEMBERSMAP + +/*! \def RAPIDJSON_USE_MEMBERSMAP + \ingroup RAPIDJSON_CONFIG + \brief Enable RapidJSON support for object members handling in a \c std::multimap + + By defining this preprocessor symbol to \c 1, \ref rapidjson::GenericValue object + members are stored in a \c std::multimap for faster lookup and deletion times, a + trade off with a slightly slower insertion time and a small object allocat(or)ed + memory overhead. + + \hideinitializer +*/ +#ifndef RAPIDJSON_USE_MEMBERSMAP +#define RAPIDJSON_USE_MEMBERSMAP 0 // not by default +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_NO_INT64DEFINE + +/*! \def RAPIDJSON_NO_INT64DEFINE + \ingroup RAPIDJSON_CONFIG + \brief Use external 64-bit integer types. + + RapidJSON requires the 64-bit integer types \c int64_t and \c uint64_t types + to be available at global scope. + + If users have their own definition, define RAPIDJSON_NO_INT64DEFINE to + prevent RapidJSON from defining its own types. +*/ +#ifndef RAPIDJSON_NO_INT64DEFINE +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#if defined(_MSC_VER) && (_MSC_VER < 1800) // Visual Studio 2013 +#include "msinttypes/stdint.h" +#include "msinttypes/inttypes.h" +#else +// Other compilers should have this. +#include +#include +#endif +//!@endcond +#ifdef RAPIDJSON_DOXYGEN_RUNNING +#define RAPIDJSON_NO_INT64DEFINE +#endif +#endif // RAPIDJSON_NO_INT64TYPEDEF + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_FORCEINLINE + +#ifndef RAPIDJSON_FORCEINLINE +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#if defined(_MSC_VER) && defined(NDEBUG) +#define RAPIDJSON_FORCEINLINE __forceinline +#elif defined(__GNUC__) && __GNUC__ >= 4 && defined(NDEBUG) +#define RAPIDJSON_FORCEINLINE __attribute__((always_inline)) +#else +#define RAPIDJSON_FORCEINLINE +#endif +//!@endcond +#endif // RAPIDJSON_FORCEINLINE + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_ENDIAN +#define RAPIDJSON_LITTLEENDIAN 0 //!< Little endian machine +#define RAPIDJSON_BIGENDIAN 1 //!< Big endian machine + +//! Endianness of the machine. +/*! + \def RAPIDJSON_ENDIAN + \ingroup RAPIDJSON_CONFIG + + GCC 4.6 provided macro for detecting endianness of the target machine. But other + compilers may not have this. User can define RAPIDJSON_ENDIAN to either + \ref RAPIDJSON_LITTLEENDIAN or \ref RAPIDJSON_BIGENDIAN. + + Default detection implemented with reference to + \li https://gcc.gnu.org/onlinedocs/gcc-4.6.0/cpp/Common-Predefined-Macros.html + \li http://www.boost.org/doc/libs/1_42_0/boost/detail/endian.hpp +*/ +#ifndef RAPIDJSON_ENDIAN +// Detect with GCC 4.6's macro +# ifdef __BYTE_ORDER__ +# if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__ +# define RAPIDJSON_ENDIAN RAPIDJSON_LITTLEENDIAN +# elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__ +# define RAPIDJSON_ENDIAN RAPIDJSON_BIGENDIAN +# else +# error Unknown machine endianness detected. User needs to define RAPIDJSON_ENDIAN. +# endif // __BYTE_ORDER__ +// Detect with GLIBC's endian.h +# elif defined(__GLIBC__) +# include +# if (__BYTE_ORDER == __LITTLE_ENDIAN) +# define RAPIDJSON_ENDIAN RAPIDJSON_LITTLEENDIAN +# elif (__BYTE_ORDER == __BIG_ENDIAN) +# define RAPIDJSON_ENDIAN RAPIDJSON_BIGENDIAN +# else +# error Unknown machine endianness detected. User needs to define RAPIDJSON_ENDIAN. +# endif // __GLIBC__ +// Detect with _LITTLE_ENDIAN and _BIG_ENDIAN macro +# elif defined(_LITTLE_ENDIAN) && !defined(_BIG_ENDIAN) +# define RAPIDJSON_ENDIAN RAPIDJSON_LITTLEENDIAN +# elif defined(_BIG_ENDIAN) && !defined(_LITTLE_ENDIAN) +# define RAPIDJSON_ENDIAN RAPIDJSON_BIGENDIAN +// Detect with architecture macros +# elif defined(__sparc) || defined(__sparc__) || defined(_POWER) || defined(__powerpc__) || defined(__ppc__) || defined(__ppc64__) || defined(__hpux) || defined(__hppa) || defined(_MIPSEB) || defined(_POWER) || defined(__s390__) +# define RAPIDJSON_ENDIAN RAPIDJSON_BIGENDIAN +# elif defined(__i386__) || defined(__alpha__) || defined(__ia64) || defined(__ia64__) || defined(_M_IX86) || defined(_M_IA64) || defined(_M_ALPHA) || defined(__amd64) || defined(__amd64__) || defined(_M_AMD64) || defined(__x86_64) || defined(__x86_64__) || defined(_M_X64) || defined(__bfin__) +# define RAPIDJSON_ENDIAN RAPIDJSON_LITTLEENDIAN +# elif defined(_MSC_VER) && (defined(_M_ARM) || defined(_M_ARM64)) +# define RAPIDJSON_ENDIAN RAPIDJSON_LITTLEENDIAN +# elif defined(RAPIDJSON_DOXYGEN_RUNNING) +# define RAPIDJSON_ENDIAN +# else +# error Unknown machine endianness detected. User needs to define RAPIDJSON_ENDIAN. +# endif +#endif // RAPIDJSON_ENDIAN + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_64BIT + +//! Whether using 64-bit architecture +#ifndef RAPIDJSON_64BIT +#if defined(__LP64__) || (defined(__x86_64__) && defined(__ILP32__)) || defined(_WIN64) || defined(__EMSCRIPTEN__) +#define RAPIDJSON_64BIT 1 +#else +#define RAPIDJSON_64BIT 0 +#endif +#endif // RAPIDJSON_64BIT + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_ALIGN + +//! Data alignment of the machine. +/*! \ingroup RAPIDJSON_CONFIG + \param x pointer to align + + Some machines require strict data alignment. The default is 8 bytes. + User can customize by defining the RAPIDJSON_ALIGN function macro. +*/ +#ifndef RAPIDJSON_ALIGN +#define RAPIDJSON_ALIGN(x) (((x) + static_cast(7u)) & ~static_cast(7u)) +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_UINT64_C2 + +//! Construct a 64-bit literal by a pair of 32-bit integer. +/*! + 64-bit literal with or without ULL suffix is prone to compiler warnings. + UINT64_C() is C macro which cause compilation problems. + Use this macro to define 64-bit constants by a pair of 32-bit integer. +*/ +#ifndef RAPIDJSON_UINT64_C2 +#define RAPIDJSON_UINT64_C2(high32, low32) ((static_cast(high32) << 32) | static_cast(low32)) +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_48BITPOINTER_OPTIMIZATION + +//! Use only lower 48-bit address for some pointers. +/*! + \ingroup RAPIDJSON_CONFIG + + This optimization uses the fact that current X86-64 architecture only implement lower 48-bit virtual address. + The higher 16-bit can be used for storing other data. + \c GenericValue uses this optimization to reduce its size form 24 bytes to 16 bytes in 64-bit architecture. +*/ +#ifndef RAPIDJSON_48BITPOINTER_OPTIMIZATION +#if defined(__amd64__) || defined(__amd64) || defined(__x86_64__) || defined(__x86_64) || defined(_M_X64) || defined(_M_AMD64) +#define RAPIDJSON_48BITPOINTER_OPTIMIZATION 1 +#else +#define RAPIDJSON_48BITPOINTER_OPTIMIZATION 0 +#endif +#endif // RAPIDJSON_48BITPOINTER_OPTIMIZATION + +#if RAPIDJSON_48BITPOINTER_OPTIMIZATION == 1 +#if RAPIDJSON_64BIT != 1 +#error RAPIDJSON_48BITPOINTER_OPTIMIZATION can only be set to 1 when RAPIDJSON_64BIT=1 +#endif +#define RAPIDJSON_SETPOINTER(type, p, x) (p = reinterpret_cast((reinterpret_cast(p) & static_cast(RAPIDJSON_UINT64_C2(0xFFFF0000, 0x00000000))) | reinterpret_cast(reinterpret_cast(x)))) +#define RAPIDJSON_GETPOINTER(type, p) (reinterpret_cast(reinterpret_cast(p) & static_cast(RAPIDJSON_UINT64_C2(0x0000FFFF, 0xFFFFFFFF)))) +#else +#define RAPIDJSON_SETPOINTER(type, p, x) (p = (x)) +#define RAPIDJSON_GETPOINTER(type, p) (p) +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_SSE2/RAPIDJSON_SSE42/RAPIDJSON_NEON/RAPIDJSON_SIMD + +/*! \def RAPIDJSON_SIMD + \ingroup RAPIDJSON_CONFIG + \brief Enable SSE2/SSE4.2/Neon optimization. + + RapidJSON supports optimized implementations for some parsing operations + based on the SSE2, SSE4.2 or NEon SIMD extensions on modern Intel + or ARM compatible processors. + + To enable these optimizations, three different symbols can be defined; + \code + // Enable SSE2 optimization. + #define RAPIDJSON_SSE2 + + // Enable SSE4.2 optimization. + #define RAPIDJSON_SSE42 + \endcode + + // Enable ARM Neon optimization. + #define RAPIDJSON_NEON + \endcode + + \c RAPIDJSON_SSE42 takes precedence over SSE2, if both are defined. + + If any of these symbols is defined, RapidJSON defines the macro + \c RAPIDJSON_SIMD to indicate the availability of the optimized code. +*/ +#if defined(RAPIDJSON_SSE2) || defined(RAPIDJSON_SSE42) \ + || defined(RAPIDJSON_NEON) || defined(RAPIDJSON_DOXYGEN_RUNNING) +#define RAPIDJSON_SIMD +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_NO_SIZETYPEDEFINE + +#ifndef RAPIDJSON_NO_SIZETYPEDEFINE +/*! \def RAPIDJSON_NO_SIZETYPEDEFINE + \ingroup RAPIDJSON_CONFIG + \brief User-provided \c SizeType definition. + + In order to avoid using 32-bit size types for indexing strings and arrays, + define this preprocessor symbol and provide the type rapidjson::SizeType + before including RapidJSON: + \code + #define RAPIDJSON_NO_SIZETYPEDEFINE + namespace rapidjson { typedef ::std::size_t SizeType; } + #include "rapidjson/..." + \endcode + + \see rapidjson::SizeType +*/ +#ifdef RAPIDJSON_DOXYGEN_RUNNING +#define RAPIDJSON_NO_SIZETYPEDEFINE +#endif +RAPIDJSON_NAMESPACE_BEGIN +//! Size type (for string lengths, array sizes, etc.) +/*! RapidJSON uses 32-bit array/string indices even on 64-bit platforms, + instead of using \c size_t. Users may override the SizeType by defining + \ref RAPIDJSON_NO_SIZETYPEDEFINE. +*/ +typedef unsigned SizeType; +RAPIDJSON_NAMESPACE_END +#endif + +// always import std::size_t to rapidjson namespace +RAPIDJSON_NAMESPACE_BEGIN +using std::size_t; +RAPIDJSON_NAMESPACE_END + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_ASSERT + +//! Assertion. +/*! \ingroup RAPIDJSON_CONFIG + By default, rapidjson uses C \c assert() for internal assertions. + User can override it by defining RAPIDJSON_ASSERT(x) macro. + + \note Parsing errors are handled and can be customized by the + \ref RAPIDJSON_ERRORS APIs. +*/ +#ifndef RAPIDJSON_ASSERT +#include +#define RAPIDJSON_ASSERT(x) assert(x) +#endif // RAPIDJSON_ASSERT + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_STATIC_ASSERT + +// Prefer C++11 static_assert, if available +#ifndef RAPIDJSON_STATIC_ASSERT +#if RAPIDJSON_CPLUSPLUS >= 201103L || ( defined(_MSC_VER) && _MSC_VER >= 1800 ) +#define RAPIDJSON_STATIC_ASSERT(x) \ + static_assert(x, RAPIDJSON_STRINGIFY(x)) +#endif // C++11 +#endif // RAPIDJSON_STATIC_ASSERT + +// Adopt C++03 implementation from boost +#ifndef RAPIDJSON_STATIC_ASSERT +#ifndef __clang__ +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#endif +RAPIDJSON_NAMESPACE_BEGIN +template struct STATIC_ASSERTION_FAILURE; +template <> struct STATIC_ASSERTION_FAILURE { enum { value = 1 }; }; +template struct StaticAssertTest {}; +RAPIDJSON_NAMESPACE_END + +#if defined(__GNUC__) || defined(__clang__) +#define RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE __attribute__((unused)) +#else +#define RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE +#endif +#ifndef __clang__ +//!@endcond +#endif + +/*! \def RAPIDJSON_STATIC_ASSERT + \brief (Internal) macro to check for conditions at compile-time + \param x compile-time condition + \hideinitializer + */ +#define RAPIDJSON_STATIC_ASSERT(x) \ + typedef ::RAPIDJSON_NAMESPACE::StaticAssertTest< \ + sizeof(::RAPIDJSON_NAMESPACE::STATIC_ASSERTION_FAILURE)> \ + RAPIDJSON_JOIN(StaticAssertTypedef, __LINE__) RAPIDJSON_STATIC_ASSERT_UNUSED_ATTRIBUTE +#endif // RAPIDJSON_STATIC_ASSERT + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_LIKELY, RAPIDJSON_UNLIKELY + +//! Compiler branching hint for expression with high probability to be true. +/*! + \ingroup RAPIDJSON_CONFIG + \param x Boolean expression likely to be true. +*/ +#ifndef RAPIDJSON_LIKELY +#if defined(__GNUC__) || defined(__clang__) +#define RAPIDJSON_LIKELY(x) __builtin_expect(!!(x), 1) +#else +#define RAPIDJSON_LIKELY(x) (x) +#endif +#endif + +//! Compiler branching hint for expression with low probability to be true. +/*! + \ingroup RAPIDJSON_CONFIG + \param x Boolean expression unlikely to be true. +*/ +#ifndef RAPIDJSON_UNLIKELY +#if defined(__GNUC__) || defined(__clang__) +#define RAPIDJSON_UNLIKELY(x) __builtin_expect(!!(x), 0) +#else +#define RAPIDJSON_UNLIKELY(x) (x) +#endif +#endif + +/////////////////////////////////////////////////////////////////////////////// +// Helpers + +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN + +#define RAPIDJSON_MULTILINEMACRO_BEGIN do { +#define RAPIDJSON_MULTILINEMACRO_END \ +} while((void)0, 0) + +// adopted from Boost +#define RAPIDJSON_VERSION_CODE(x,y,z) \ + (((x)*100000) + ((y)*100) + (z)) + +#if defined(__has_builtin) +#define RAPIDJSON_HAS_BUILTIN(x) __has_builtin(x) +#else +#define RAPIDJSON_HAS_BUILTIN(x) 0 +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_DIAG_PUSH/POP, RAPIDJSON_DIAG_OFF + +#if defined(__GNUC__) +#define RAPIDJSON_GNUC \ + RAPIDJSON_VERSION_CODE(__GNUC__,__GNUC_MINOR__,__GNUC_PATCHLEVEL__) +#endif + +#if defined(__clang__) || (defined(RAPIDJSON_GNUC) && RAPIDJSON_GNUC >= RAPIDJSON_VERSION_CODE(4,2,0)) + +#define RAPIDJSON_PRAGMA(x) _Pragma(RAPIDJSON_STRINGIFY(x)) +#define RAPIDJSON_DIAG_PRAGMA(x) RAPIDJSON_PRAGMA(GCC diagnostic x) +#define RAPIDJSON_DIAG_OFF(x) \ + RAPIDJSON_DIAG_PRAGMA(ignored RAPIDJSON_STRINGIFY(RAPIDJSON_JOIN(-W,x))) + +// push/pop support in Clang and GCC>=4.6 +#if defined(__clang__) || (defined(RAPIDJSON_GNUC) && RAPIDJSON_GNUC >= RAPIDJSON_VERSION_CODE(4,6,0)) +#define RAPIDJSON_DIAG_PUSH RAPIDJSON_DIAG_PRAGMA(push) +#define RAPIDJSON_DIAG_POP RAPIDJSON_DIAG_PRAGMA(pop) +#else // GCC >= 4.2, < 4.6 +#define RAPIDJSON_DIAG_PUSH /* ignored */ +#define RAPIDJSON_DIAG_POP /* ignored */ +#endif + +#elif defined(_MSC_VER) + +// pragma (MSVC specific) +#define RAPIDJSON_PRAGMA(x) __pragma(x) +#define RAPIDJSON_DIAG_PRAGMA(x) RAPIDJSON_PRAGMA(warning(x)) + +#define RAPIDJSON_DIAG_OFF(x) RAPIDJSON_DIAG_PRAGMA(disable: x) +#define RAPIDJSON_DIAG_PUSH RAPIDJSON_DIAG_PRAGMA(push) +#define RAPIDJSON_DIAG_POP RAPIDJSON_DIAG_PRAGMA(pop) + +#else + +#define RAPIDJSON_DIAG_OFF(x) /* ignored */ +#define RAPIDJSON_DIAG_PUSH /* ignored */ +#define RAPIDJSON_DIAG_POP /* ignored */ + +#endif // RAPIDJSON_DIAG_* + +/////////////////////////////////////////////////////////////////////////////// +// C++11 features + +#ifndef RAPIDJSON_HAS_CXX11 +#define RAPIDJSON_HAS_CXX11 (RAPIDJSON_CPLUSPLUS >= 201103L) +#endif + +#ifndef RAPIDJSON_HAS_CXX11_RVALUE_REFS +#if RAPIDJSON_HAS_CXX11 +#define RAPIDJSON_HAS_CXX11_RVALUE_REFS 1 +#elif defined(__clang__) +#if __has_feature(cxx_rvalue_references) && \ + (defined(_MSC_VER) || defined(_LIBCPP_VERSION) || defined(__GLIBCXX__) && __GLIBCXX__ >= 20080306) +#define RAPIDJSON_HAS_CXX11_RVALUE_REFS 1 +#else +#define RAPIDJSON_HAS_CXX11_RVALUE_REFS 0 +#endif +#elif (defined(RAPIDJSON_GNUC) && (RAPIDJSON_GNUC >= RAPIDJSON_VERSION_CODE(4,3,0)) && defined(__GXX_EXPERIMENTAL_CXX0X__)) || \ + (defined(_MSC_VER) && _MSC_VER >= 1600) || \ + (defined(__SUNPRO_CC) && __SUNPRO_CC >= 0x5140 && defined(__GXX_EXPERIMENTAL_CXX0X__)) + +#define RAPIDJSON_HAS_CXX11_RVALUE_REFS 1 +#else +#define RAPIDJSON_HAS_CXX11_RVALUE_REFS 0 +#endif +#endif // RAPIDJSON_HAS_CXX11_RVALUE_REFS + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS +#include // std::move +#endif + +#ifndef RAPIDJSON_HAS_CXX11_NOEXCEPT +#if RAPIDJSON_HAS_CXX11 +#define RAPIDJSON_HAS_CXX11_NOEXCEPT 1 +#elif defined(__clang__) +#define RAPIDJSON_HAS_CXX11_NOEXCEPT __has_feature(cxx_noexcept) +#elif (defined(RAPIDJSON_GNUC) && (RAPIDJSON_GNUC >= RAPIDJSON_VERSION_CODE(4,6,0)) && defined(__GXX_EXPERIMENTAL_CXX0X__)) || \ + (defined(_MSC_VER) && _MSC_VER >= 1900) || \ + (defined(__SUNPRO_CC) && __SUNPRO_CC >= 0x5140 && defined(__GXX_EXPERIMENTAL_CXX0X__)) +#define RAPIDJSON_HAS_CXX11_NOEXCEPT 1 +#else +#define RAPIDJSON_HAS_CXX11_NOEXCEPT 0 +#endif +#endif +#ifndef RAPIDJSON_NOEXCEPT +#if RAPIDJSON_HAS_CXX11_NOEXCEPT +#define RAPIDJSON_NOEXCEPT noexcept +#else +#define RAPIDJSON_NOEXCEPT throw() +#endif // RAPIDJSON_HAS_CXX11_NOEXCEPT +#endif + +// no automatic detection, yet +#ifndef RAPIDJSON_HAS_CXX11_TYPETRAITS +#if (defined(_MSC_VER) && _MSC_VER >= 1700) +#define RAPIDJSON_HAS_CXX11_TYPETRAITS 1 +#else +#define RAPIDJSON_HAS_CXX11_TYPETRAITS 0 +#endif +#endif + +#ifndef RAPIDJSON_HAS_CXX11_RANGE_FOR +#if defined(__clang__) +#define RAPIDJSON_HAS_CXX11_RANGE_FOR __has_feature(cxx_range_for) +#elif (defined(RAPIDJSON_GNUC) && (RAPIDJSON_GNUC >= RAPIDJSON_VERSION_CODE(4,6,0)) && defined(__GXX_EXPERIMENTAL_CXX0X__)) || \ + (defined(_MSC_VER) && _MSC_VER >= 1700) || \ + (defined(__SUNPRO_CC) && __SUNPRO_CC >= 0x5140 && defined(__GXX_EXPERIMENTAL_CXX0X__)) +#define RAPIDJSON_HAS_CXX11_RANGE_FOR 1 +#else +#define RAPIDJSON_HAS_CXX11_RANGE_FOR 0 +#endif +#endif // RAPIDJSON_HAS_CXX11_RANGE_FOR + +/////////////////////////////////////////////////////////////////////////////// +// C++17 features + +#ifndef RAPIDJSON_HAS_CXX17 +#define RAPIDJSON_HAS_CXX17 (RAPIDJSON_CPLUSPLUS >= 201703L) +#endif + +#if RAPIDJSON_HAS_CXX17 +# define RAPIDJSON_DELIBERATE_FALLTHROUGH [[fallthrough]] +#elif defined(__has_cpp_attribute) +# if __has_cpp_attribute(clang::fallthrough) +# define RAPIDJSON_DELIBERATE_FALLTHROUGH [[clang::fallthrough]] +# elif __has_cpp_attribute(fallthrough) +# define RAPIDJSON_DELIBERATE_FALLTHROUGH __attribute__((fallthrough)) +# else +# define RAPIDJSON_DELIBERATE_FALLTHROUGH +# endif +#else +# define RAPIDJSON_DELIBERATE_FALLTHROUGH +#endif + +//!@endcond + +//! Assertion (in non-throwing contexts). + /*! \ingroup RAPIDJSON_CONFIG + Some functions provide a \c noexcept guarantee, if the compiler supports it. + In these cases, the \ref RAPIDJSON_ASSERT macro cannot be overridden to + throw an exception. This macro adds a separate customization point for + such cases. + + Defaults to C \c assert() (as \ref RAPIDJSON_ASSERT), if \c noexcept is + supported, and to \ref RAPIDJSON_ASSERT otherwise. + */ + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_NOEXCEPT_ASSERT + +#ifndef RAPIDJSON_NOEXCEPT_ASSERT +#ifdef RAPIDJSON_ASSERT_THROWS +#include +#define RAPIDJSON_NOEXCEPT_ASSERT(x) assert(x) +#else +#define RAPIDJSON_NOEXCEPT_ASSERT(x) RAPIDJSON_ASSERT(x) +#endif // RAPIDJSON_ASSERT_THROWS +#endif // RAPIDJSON_NOEXCEPT_ASSERT + +/////////////////////////////////////////////////////////////////////////////// +// malloc/realloc/free + +#ifndef RAPIDJSON_MALLOC +///! customization point for global \c malloc +#define RAPIDJSON_MALLOC(size) std::malloc(size) +#endif +#ifndef RAPIDJSON_REALLOC +///! customization point for global \c realloc +#define RAPIDJSON_REALLOC(ptr, new_size) std::realloc(ptr, new_size) +#endif +#ifndef RAPIDJSON_FREE +///! customization point for global \c free +#define RAPIDJSON_FREE(ptr) std::free(ptr) +#endif + +/////////////////////////////////////////////////////////////////////////////// +// new/delete + +#ifndef RAPIDJSON_NEW +///! customization point for global \c new +#define RAPIDJSON_NEW(TypeName) new TypeName +#endif +#ifndef RAPIDJSON_DELETE +///! customization point for global \c delete +#define RAPIDJSON_DELETE(x) delete x +#endif + +/////////////////////////////////////////////////////////////////////////////// +// Type + +/*! \namespace rapidjson + \brief main RapidJSON namespace + \see RAPIDJSON_NAMESPACE +*/ +RAPIDJSON_NAMESPACE_BEGIN + +//! Type of JSON value +enum Type { + kNullType = 0, //!< null + kFalseType = 1, //!< false + kTrueType = 2, //!< true + kObjectType = 3, //!< object + kArrayType = 4, //!< array + kStringType = 5, //!< string + kNumberType = 6 //!< number +}; + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_RAPIDJSON_H_ diff --git a/include/rapidjson/reader.h b/include/rapidjson/reader.h new file mode 100644 index 0000000..f7ef610 --- /dev/null +++ b/include/rapidjson/reader.h @@ -0,0 +1,2246 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_READER_H_ +#define RAPIDJSON_READER_H_ + +/*! \file reader.h */ + +#include "allocators.h" +#include "stream.h" +#include "encodedstream.h" +#include "internal/clzll.h" +#include "internal/meta.h" +#include "internal/stack.h" +#include "internal/strtod.h" +#include + +#if defined(RAPIDJSON_SIMD) && defined(_MSC_VER) +#include +#pragma intrinsic(_BitScanForward) +#endif +#ifdef RAPIDJSON_SSE42 +#include +#elif defined(RAPIDJSON_SSE2) +#include +#elif defined(RAPIDJSON_NEON) +#include +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(old-style-cast) +RAPIDJSON_DIAG_OFF(padded) +RAPIDJSON_DIAG_OFF(switch-enum) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4127) // conditional expression is constant +RAPIDJSON_DIAG_OFF(4702) // unreachable code +#endif + +#ifdef __GNUC__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(effc++) +#endif + +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#define RAPIDJSON_NOTHING /* deliberately empty */ +#ifndef RAPIDJSON_PARSE_ERROR_EARLY_RETURN +#define RAPIDJSON_PARSE_ERROR_EARLY_RETURN(value) \ + RAPIDJSON_MULTILINEMACRO_BEGIN \ + if (RAPIDJSON_UNLIKELY(HasParseError())) { return value; } \ + RAPIDJSON_MULTILINEMACRO_END +#endif +#define RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID \ + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(RAPIDJSON_NOTHING) +//!@endcond + +/*! \def RAPIDJSON_PARSE_ERROR_NORETURN + \ingroup RAPIDJSON_ERRORS + \brief Macro to indicate a parse error. + \param parseErrorCode \ref rapidjson::ParseErrorCode of the error + \param offset position of the error in JSON input (\c size_t) + + This macros can be used as a customization point for the internal + error handling mechanism of RapidJSON. + + A common usage model is to throw an exception instead of requiring the + caller to explicitly check the \ref rapidjson::GenericReader::Parse's + return value: + + \code + #define RAPIDJSON_PARSE_ERROR_NORETURN(parseErrorCode,offset) \ + throw ParseException(parseErrorCode, #parseErrorCode, offset) + + #include // std::runtime_error + #include "rapidjson/error/error.h" // rapidjson::ParseResult + + struct ParseException : std::runtime_error, rapidjson::ParseResult { + ParseException(rapidjson::ParseErrorCode code, const char* msg, size_t offset) + : std::runtime_error(msg), ParseResult(code, offset) {} + }; + + #include "rapidjson/reader.h" + \endcode + + \see RAPIDJSON_PARSE_ERROR, rapidjson::GenericReader::Parse + */ +#ifndef RAPIDJSON_PARSE_ERROR_NORETURN +#define RAPIDJSON_PARSE_ERROR_NORETURN(parseErrorCode, offset) \ + RAPIDJSON_MULTILINEMACRO_BEGIN \ + RAPIDJSON_ASSERT(!HasParseError()); /* Error can only be assigned once */ \ + SetParseError(parseErrorCode, offset); \ + RAPIDJSON_MULTILINEMACRO_END +#endif + +/*! \def RAPIDJSON_PARSE_ERROR + \ingroup RAPIDJSON_ERRORS + \brief (Internal) macro to indicate and handle a parse error. + \param parseErrorCode \ref rapidjson::ParseErrorCode of the error + \param offset position of the error in JSON input (\c size_t) + + Invokes RAPIDJSON_PARSE_ERROR_NORETURN and stops the parsing. + + \see RAPIDJSON_PARSE_ERROR_NORETURN + \hideinitializer + */ +#ifndef RAPIDJSON_PARSE_ERROR +#define RAPIDJSON_PARSE_ERROR(parseErrorCode, offset) \ + RAPIDJSON_MULTILINEMACRO_BEGIN \ + RAPIDJSON_PARSE_ERROR_NORETURN(parseErrorCode, offset); \ + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; \ + RAPIDJSON_MULTILINEMACRO_END +#endif + +#include "error/error.h" // ParseErrorCode, ParseResult + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// ParseFlag + +/*! \def RAPIDJSON_PARSE_DEFAULT_FLAGS + \ingroup RAPIDJSON_CONFIG + \brief User-defined kParseDefaultFlags definition. + + User can define this as any \c ParseFlag combinations. +*/ +#ifndef RAPIDJSON_PARSE_DEFAULT_FLAGS +#define RAPIDJSON_PARSE_DEFAULT_FLAGS kParseNoFlags +#endif + +//! Combination of parseFlags +/*! \see Reader::Parse, Document::Parse, Document::ParseInsitu, Document::ParseStream + */ +enum ParseFlag { + kParseNoFlags = 0, //!< No flags are set. + kParseInsituFlag = 1, //!< In-situ(destructive) parsing. + kParseValidateEncodingFlag = 2, //!< Validate encoding of JSON strings. + kParseIterativeFlag = 4, //!< Iterative(constant complexity in terms of function call stack size) parsing. + kParseStopWhenDoneFlag = 8, //!< After parsing a complete JSON root from stream, stop further processing the rest of stream. When this flag is used, parser will not generate kParseErrorDocumentRootNotSingular error. + kParseFullPrecisionFlag = 16, //!< Parse number in full precision (but slower). + kParseCommentsFlag = 32, //!< Allow one-line (//) and multi-line (/**/) comments. + kParseNumbersAsStringsFlag = 64, //!< Parse all numbers (ints/doubles) as strings. + kParseTrailingCommasFlag = 128, //!< Allow trailing commas at the end of objects and arrays. + kParseNanAndInfFlag = 256, //!< Allow parsing NaN, Inf, Infinity, -Inf and -Infinity as doubles. + kParseEscapedApostropheFlag = 512, //!< Allow escaped apostrophe in strings. + kParseDefaultFlags = RAPIDJSON_PARSE_DEFAULT_FLAGS //!< Default parse flags. Can be customized by defining RAPIDJSON_PARSE_DEFAULT_FLAGS +}; + +/////////////////////////////////////////////////////////////////////////////// +// Handler + +/*! \class rapidjson::Handler + \brief Concept for receiving events from GenericReader upon parsing. + The functions return true if no error occurs. If they return false, + the event publisher should terminate the process. +\code +concept Handler { + typename Ch; + + bool Null(); + bool Bool(bool b); + bool Int(int i); + bool Uint(unsigned i); + bool Int64(int64_t i); + bool Uint64(uint64_t i); + bool Double(double d); + /// enabled via kParseNumbersAsStringsFlag, string is not null-terminated (use length) + bool RawNumber(const Ch* str, SizeType length, bool copy); + bool String(const Ch* str, SizeType length, bool copy); + bool StartObject(); + bool Key(const Ch* str, SizeType length, bool copy); + bool EndObject(SizeType memberCount); + bool StartArray(); + bool EndArray(SizeType elementCount); +}; +\endcode +*/ +/////////////////////////////////////////////////////////////////////////////// +// BaseReaderHandler + +//! Default implementation of Handler. +/*! This can be used as base class of any reader handler. + \note implements Handler concept +*/ +template, typename Derived = void> +struct BaseReaderHandler { + typedef typename Encoding::Ch Ch; + + typedef typename internal::SelectIf, BaseReaderHandler, Derived>::Type Override; + + bool Default() { return true; } + bool Null() { return static_cast(*this).Default(); } + bool Bool(bool) { return static_cast(*this).Default(); } + bool Int(int) { return static_cast(*this).Default(); } + bool Uint(unsigned) { return static_cast(*this).Default(); } + bool Int64(int64_t) { return static_cast(*this).Default(); } + bool Uint64(uint64_t) { return static_cast(*this).Default(); } + bool Double(double) { return static_cast(*this).Default(); } + /// enabled via kParseNumbersAsStringsFlag, string is not null-terminated (use length) + bool RawNumber(const Ch* str, SizeType len, bool copy) { return static_cast(*this).String(str, len, copy); } + bool String(const Ch*, SizeType, bool) { return static_cast(*this).Default(); } + bool StartObject() { return static_cast(*this).Default(); } + bool Key(const Ch* str, SizeType len, bool copy) { return static_cast(*this).String(str, len, copy); } + bool EndObject(SizeType) { return static_cast(*this).Default(); } + bool StartArray() { return static_cast(*this).Default(); } + bool EndArray(SizeType) { return static_cast(*this).Default(); } +}; + +/////////////////////////////////////////////////////////////////////////////// +// StreamLocalCopy + +namespace internal { + +template::copyOptimization> +class StreamLocalCopy; + +//! Do copy optimization. +template +class StreamLocalCopy { +public: + StreamLocalCopy(Stream& original) : s(original), original_(original) {} + ~StreamLocalCopy() { original_ = s; } + + Stream s; + +private: + StreamLocalCopy& operator=(const StreamLocalCopy&) /* = delete */; + + Stream& original_; +}; + +//! Keep reference. +template +class StreamLocalCopy { +public: + StreamLocalCopy(Stream& original) : s(original) {} + + Stream& s; + +private: + StreamLocalCopy& operator=(const StreamLocalCopy&) /* = delete */; +}; + +} // namespace internal + +/////////////////////////////////////////////////////////////////////////////// +// SkipWhitespace + +//! Skip the JSON white spaces in a stream. +/*! \param is A input stream for skipping white spaces. + \note This function has SSE2/SSE4.2 specialization. +*/ +template +void SkipWhitespace(InputStream& is) { + internal::StreamLocalCopy copy(is); + InputStream& s(copy.s); + + typename InputStream::Ch c; + while ((c = s.Peek()) == ' ' || c == '\n' || c == '\r' || c == '\t') + s.Take(); +} + +inline const char* SkipWhitespace(const char* p, const char* end) { + while (p != end && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')) + ++p; + return p; +} + +#ifdef RAPIDJSON_SSE42 +//! Skip whitespace with SSE 4.2 pcmpistrm instruction, testing 16 8-byte characters at once. +inline const char *SkipWhitespace_SIMD(const char* p) { + // Fast return for single non-whitespace + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + // 16-byte align to the next boundary + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + // The rest of string using SIMD + static const char whitespace[16] = " \n\r\t"; + const __m128i w = _mm_loadu_si128(reinterpret_cast(&whitespace[0])); + + for (;; p += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + const int r = _mm_cmpistri(w, s, _SIDD_UBYTE_OPS | _SIDD_CMP_EQUAL_ANY | _SIDD_LEAST_SIGNIFICANT | _SIDD_NEGATIVE_POLARITY); + if (r != 16) // some of characters is non-whitespace + return p + r; + } +} + +inline const char *SkipWhitespace_SIMD(const char* p, const char* end) { + // Fast return for single non-whitespace + if (p != end && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')) + ++p; + else + return p; + + // The middle of string using SIMD + static const char whitespace[16] = " \n\r\t"; + const __m128i w = _mm_loadu_si128(reinterpret_cast(&whitespace[0])); + + for (; p <= end - 16; p += 16) { + const __m128i s = _mm_loadu_si128(reinterpret_cast(p)); + const int r = _mm_cmpistri(w, s, _SIDD_UBYTE_OPS | _SIDD_CMP_EQUAL_ANY | _SIDD_LEAST_SIGNIFICANT | _SIDD_NEGATIVE_POLARITY); + if (r != 16) // some of characters is non-whitespace + return p + r; + } + + return SkipWhitespace(p, end); +} + +#elif defined(RAPIDJSON_SSE2) + +//! Skip whitespace with SSE2 instructions, testing 16 8-byte characters at once. +inline const char *SkipWhitespace_SIMD(const char* p) { + // Fast return for single non-whitespace + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + // 16-byte align to the next boundary + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + // The rest of string + #define C16(c) { c, c, c, c, c, c, c, c, c, c, c, c, c, c, c, c } + static const char whitespaces[4][16] = { C16(' '), C16('\n'), C16('\r'), C16('\t') }; + #undef C16 + + const __m128i w0 = _mm_loadu_si128(reinterpret_cast(&whitespaces[0][0])); + const __m128i w1 = _mm_loadu_si128(reinterpret_cast(&whitespaces[1][0])); + const __m128i w2 = _mm_loadu_si128(reinterpret_cast(&whitespaces[2][0])); + const __m128i w3 = _mm_loadu_si128(reinterpret_cast(&whitespaces[3][0])); + + for (;; p += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + __m128i x = _mm_cmpeq_epi8(s, w0); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w1)); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w2)); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w3)); + unsigned short r = static_cast(~_mm_movemask_epi8(x)); + if (r != 0) { // some of characters may be non-whitespace +#ifdef _MSC_VER // Find the index of first non-whitespace + unsigned long offset; + _BitScanForward(&offset, r); + return p + offset; +#else + return p + __builtin_ffs(r) - 1; +#endif + } + } +} + +inline const char *SkipWhitespace_SIMD(const char* p, const char* end) { + // Fast return for single non-whitespace + if (p != end && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')) + ++p; + else + return p; + + // The rest of string + #define C16(c) { c, c, c, c, c, c, c, c, c, c, c, c, c, c, c, c } + static const char whitespaces[4][16] = { C16(' '), C16('\n'), C16('\r'), C16('\t') }; + #undef C16 + + const __m128i w0 = _mm_loadu_si128(reinterpret_cast(&whitespaces[0][0])); + const __m128i w1 = _mm_loadu_si128(reinterpret_cast(&whitespaces[1][0])); + const __m128i w2 = _mm_loadu_si128(reinterpret_cast(&whitespaces[2][0])); + const __m128i w3 = _mm_loadu_si128(reinterpret_cast(&whitespaces[3][0])); + + for (; p <= end - 16; p += 16) { + const __m128i s = _mm_loadu_si128(reinterpret_cast(p)); + __m128i x = _mm_cmpeq_epi8(s, w0); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w1)); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w2)); + x = _mm_or_si128(x, _mm_cmpeq_epi8(s, w3)); + unsigned short r = static_cast(~_mm_movemask_epi8(x)); + if (r != 0) { // some of characters may be non-whitespace +#ifdef _MSC_VER // Find the index of first non-whitespace + unsigned long offset; + _BitScanForward(&offset, r); + return p + offset; +#else + return p + __builtin_ffs(r) - 1; +#endif + } + } + + return SkipWhitespace(p, end); +} + +#elif defined(RAPIDJSON_NEON) + +//! Skip whitespace with ARM Neon instructions, testing 16 8-byte characters at once. +inline const char *SkipWhitespace_SIMD(const char* p) { + // Fast return for single non-whitespace + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + // 16-byte align to the next boundary + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t') + ++p; + else + return p; + + const uint8x16_t w0 = vmovq_n_u8(' '); + const uint8x16_t w1 = vmovq_n_u8('\n'); + const uint8x16_t w2 = vmovq_n_u8('\r'); + const uint8x16_t w3 = vmovq_n_u8('\t'); + + for (;; p += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, w0); + x = vorrq_u8(x, vceqq_u8(s, w1)); + x = vorrq_u8(x, vceqq_u8(s, w2)); + x = vorrq_u8(x, vceqq_u8(s, w3)); + + x = vmvnq_u8(x); // Negate + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + return p + 8 + (lz >> 3); + } + } else { + uint32_t lz = internal::clzll(low); + return p + (lz >> 3); + } + } +} + +inline const char *SkipWhitespace_SIMD(const char* p, const char* end) { + // Fast return for single non-whitespace + if (p != end && (*p == ' ' || *p == '\n' || *p == '\r' || *p == '\t')) + ++p; + else + return p; + + const uint8x16_t w0 = vmovq_n_u8(' '); + const uint8x16_t w1 = vmovq_n_u8('\n'); + const uint8x16_t w2 = vmovq_n_u8('\r'); + const uint8x16_t w3 = vmovq_n_u8('\t'); + + for (; p <= end - 16; p += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, w0); + x = vorrq_u8(x, vceqq_u8(s, w1)); + x = vorrq_u8(x, vceqq_u8(s, w2)); + x = vorrq_u8(x, vceqq_u8(s, w3)); + + x = vmvnq_u8(x); // Negate + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + return p + 8 + (lz >> 3); + } + } else { + uint32_t lz = internal::clzll(low); + return p + (lz >> 3); + } + } + + return SkipWhitespace(p, end); +} + +#endif // RAPIDJSON_NEON + +#ifdef RAPIDJSON_SIMD +//! Template function specialization for InsituStringStream +template<> inline void SkipWhitespace(InsituStringStream& is) { + is.src_ = const_cast(SkipWhitespace_SIMD(is.src_)); +} + +//! Template function specialization for StringStream +template<> inline void SkipWhitespace(StringStream& is) { + is.src_ = SkipWhitespace_SIMD(is.src_); +} + +template<> inline void SkipWhitespace(EncodedInputStream, MemoryStream>& is) { + is.is_.src_ = SkipWhitespace_SIMD(is.is_.src_, is.is_.end_); +} +#endif // RAPIDJSON_SIMD + +/////////////////////////////////////////////////////////////////////////////// +// GenericReader + +//! SAX-style JSON parser. Use \ref Reader for UTF8 encoding and default allocator. +/*! GenericReader parses JSON text from a stream, and send events synchronously to an + object implementing Handler concept. + + It needs to allocate a stack for storing a single decoded string during + non-destructive parsing. + + For in-situ parsing, the decoded string is directly written to the source + text string, no temporary buffer is required. + + A GenericReader object can be reused for parsing multiple JSON text. + + \tparam SourceEncoding Encoding of the input stream. + \tparam TargetEncoding Encoding of the parse output. + \tparam StackAllocator Allocator type for stack. +*/ +template +class GenericReader { +public: + typedef typename SourceEncoding::Ch Ch; //!< SourceEncoding character type + + //! Constructor. + /*! \param stackAllocator Optional allocator for allocating stack memory. (Only use for non-destructive parsing) + \param stackCapacity stack capacity in bytes for storing a single decoded string. (Only use for non-destructive parsing) + */ + GenericReader(StackAllocator* stackAllocator = 0, size_t stackCapacity = kDefaultStackCapacity) : + stack_(stackAllocator, stackCapacity), parseResult_(), state_(IterativeParsingStartState) {} + + //! Parse JSON text. + /*! \tparam parseFlags Combination of \ref ParseFlag. + \tparam InputStream Type of input stream, implementing Stream concept. + \tparam Handler Type of handler, implementing Handler concept. + \param is Input stream to be parsed. + \param handler The handler to receive events. + \return Whether the parsing is successful. + */ + template + ParseResult Parse(InputStream& is, Handler& handler) { + if (parseFlags & kParseIterativeFlag) + return IterativeParse(is, handler); + + parseResult_.Clear(); + + ClearStackOnExit scope(*this); + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + + if (RAPIDJSON_UNLIKELY(is.Peek() == '\0')) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorDocumentEmpty, is.Tell()); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + } + else { + ParseValue(is, handler); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + + if (!(parseFlags & kParseStopWhenDoneFlag)) { + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + + if (RAPIDJSON_UNLIKELY(is.Peek() != '\0')) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorDocumentRootNotSingular, is.Tell()); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + } + } + } + + return parseResult_; + } + + //! Parse JSON text (with \ref kParseDefaultFlags) + /*! \tparam InputStream Type of input stream, implementing Stream concept + \tparam Handler Type of handler, implementing Handler concept. + \param is Input stream to be parsed. + \param handler The handler to receive events. + \return Whether the parsing is successful. + */ + template + ParseResult Parse(InputStream& is, Handler& handler) { + return Parse(is, handler); + } + + //! Initialize JSON text token-by-token parsing + /*! + */ + void IterativeParseInit() { + parseResult_.Clear(); + state_ = IterativeParsingStartState; + } + + //! Parse one token from JSON text + /*! \tparam InputStream Type of input stream, implementing Stream concept + \tparam Handler Type of handler, implementing Handler concept. + \param is Input stream to be parsed. + \param handler The handler to receive events. + \return Whether the parsing is successful. + */ + template + bool IterativeParseNext(InputStream& is, Handler& handler) { + while (RAPIDJSON_LIKELY(is.Peek() != '\0')) { + SkipWhitespaceAndComments(is); + + Token t = Tokenize(is.Peek()); + IterativeParsingState n = Predict(state_, t); + IterativeParsingState d = Transit(state_, t, n, is, handler); + + // If we've finished or hit an error... + if (RAPIDJSON_UNLIKELY(IsIterativeParsingCompleteState(d))) { + // Report errors. + if (d == IterativeParsingErrorState) { + HandleError(state_, is); + return false; + } + + // Transition to the finish state. + RAPIDJSON_ASSERT(d == IterativeParsingFinishState); + state_ = d; + + // If StopWhenDone is not set... + if (!(parseFlags & kParseStopWhenDoneFlag)) { + // ... and extra non-whitespace data is found... + SkipWhitespaceAndComments(is); + if (is.Peek() != '\0') { + // ... this is considered an error. + HandleError(state_, is); + return false; + } + } + + // Success! We are done! + return true; + } + + // Transition to the new state. + state_ = d; + + // If we parsed anything other than a delimiter, we invoked the handler, so we can return true now. + if (!IsIterativeParsingDelimiterState(n)) + return true; + } + + // We reached the end of file. + stack_.Clear(); + + if (state_ != IterativeParsingFinishState) { + HandleError(state_, is); + return false; + } + + return true; + } + + //! Check if token-by-token parsing JSON text is complete + /*! \return Whether the JSON has been fully decoded. + */ + RAPIDJSON_FORCEINLINE bool IterativeParseComplete() const { + return IsIterativeParsingCompleteState(state_); + } + + //! Whether a parse error has occurred in the last parsing. + bool HasParseError() const { return parseResult_.IsError(); } + + //! Get the \ref ParseErrorCode of last parsing. + ParseErrorCode GetParseErrorCode() const { return parseResult_.Code(); } + + //! Get the position of last parsing error in input, 0 otherwise. + size_t GetErrorOffset() const { return parseResult_.Offset(); } + +protected: + void SetParseError(ParseErrorCode code, size_t offset) { parseResult_.Set(code, offset); } + +private: + // Prohibit copy constructor & assignment operator. + GenericReader(const GenericReader&); + GenericReader& operator=(const GenericReader&); + + void ClearStack() { stack_.Clear(); } + + // clear stack on any exit from ParseStream, e.g. due to exception + struct ClearStackOnExit { + explicit ClearStackOnExit(GenericReader& r) : r_(r) {} + ~ClearStackOnExit() { r_.ClearStack(); } + private: + GenericReader& r_; + ClearStackOnExit(const ClearStackOnExit&); + ClearStackOnExit& operator=(const ClearStackOnExit&); + }; + + template + void SkipWhitespaceAndComments(InputStream& is) { + SkipWhitespace(is); + + if (parseFlags & kParseCommentsFlag) { + while (RAPIDJSON_UNLIKELY(Consume(is, '/'))) { + if (Consume(is, '*')) { + while (true) { + if (RAPIDJSON_UNLIKELY(is.Peek() == '\0')) + RAPIDJSON_PARSE_ERROR(kParseErrorUnspecificSyntaxError, is.Tell()); + else if (Consume(is, '*')) { + if (Consume(is, '/')) + break; + } + else + is.Take(); + } + } + else if (RAPIDJSON_LIKELY(Consume(is, '/'))) + while (is.Peek() != '\0' && is.Take() != '\n') {} + else + RAPIDJSON_PARSE_ERROR(kParseErrorUnspecificSyntaxError, is.Tell()); + + SkipWhitespace(is); + } + } + } + + // Parse object: { string : value, ... } + template + void ParseObject(InputStream& is, Handler& handler) { + RAPIDJSON_ASSERT(is.Peek() == '{'); + is.Take(); // Skip '{' + + if (RAPIDJSON_UNLIKELY(!handler.StartObject())) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + if (Consume(is, '}')) { + if (RAPIDJSON_UNLIKELY(!handler.EndObject(0))) // empty object + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + return; + } + + for (SizeType memberCount = 0;;) { + if (RAPIDJSON_UNLIKELY(is.Peek() != '"')) + RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissName, is.Tell()); + + ParseString(is, handler, true); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + if (RAPIDJSON_UNLIKELY(!Consume(is, ':'))) + RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissColon, is.Tell()); + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + ParseValue(is, handler); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + ++memberCount; + + switch (is.Peek()) { + case ',': + is.Take(); + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + break; + case '}': + is.Take(); + if (RAPIDJSON_UNLIKELY(!handler.EndObject(memberCount))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + return; + default: + RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissCommaOrCurlyBracket, is.Tell()); break; // This useless break is only for making warning and coverage happy + } + + if (parseFlags & kParseTrailingCommasFlag) { + if (is.Peek() == '}') { + if (RAPIDJSON_UNLIKELY(!handler.EndObject(memberCount))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + is.Take(); + return; + } + } + } + } + + // Parse array: [ value, ... ] + template + void ParseArray(InputStream& is, Handler& handler) { + RAPIDJSON_ASSERT(is.Peek() == '['); + is.Take(); // Skip '[' + + if (RAPIDJSON_UNLIKELY(!handler.StartArray())) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + if (Consume(is, ']')) { + if (RAPIDJSON_UNLIKELY(!handler.EndArray(0))) // empty array + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + return; + } + + for (SizeType elementCount = 0;;) { + ParseValue(is, handler); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + ++elementCount; + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + + if (Consume(is, ',')) { + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + } + else if (Consume(is, ']')) { + if (RAPIDJSON_UNLIKELY(!handler.EndArray(elementCount))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + return; + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorArrayMissCommaOrSquareBracket, is.Tell()); + + if (parseFlags & kParseTrailingCommasFlag) { + if (is.Peek() == ']') { + if (RAPIDJSON_UNLIKELY(!handler.EndArray(elementCount))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + is.Take(); + return; + } + } + } + } + + template + void ParseNull(InputStream& is, Handler& handler) { + RAPIDJSON_ASSERT(is.Peek() == 'n'); + is.Take(); + + if (RAPIDJSON_LIKELY(Consume(is, 'u') && Consume(is, 'l') && Consume(is, 'l'))) { + if (RAPIDJSON_UNLIKELY(!handler.Null())) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, is.Tell()); + } + + template + void ParseTrue(InputStream& is, Handler& handler) { + RAPIDJSON_ASSERT(is.Peek() == 't'); + is.Take(); + + if (RAPIDJSON_LIKELY(Consume(is, 'r') && Consume(is, 'u') && Consume(is, 'e'))) { + if (RAPIDJSON_UNLIKELY(!handler.Bool(true))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, is.Tell()); + } + + template + void ParseFalse(InputStream& is, Handler& handler) { + RAPIDJSON_ASSERT(is.Peek() == 'f'); + is.Take(); + + if (RAPIDJSON_LIKELY(Consume(is, 'a') && Consume(is, 'l') && Consume(is, 's') && Consume(is, 'e'))) { + if (RAPIDJSON_UNLIKELY(!handler.Bool(false))) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, is.Tell()); + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, is.Tell()); + } + + template + RAPIDJSON_FORCEINLINE static bool Consume(InputStream& is, typename InputStream::Ch expect) { + if (RAPIDJSON_LIKELY(is.Peek() == expect)) { + is.Take(); + return true; + } + else + return false; + } + + // Helper function to parse four hexadecimal digits in \uXXXX in ParseString(). + template + unsigned ParseHex4(InputStream& is, size_t escapeOffset) { + unsigned codepoint = 0; + for (int i = 0; i < 4; i++) { + Ch c = is.Peek(); + codepoint <<= 4; + codepoint += static_cast(c); + if (c >= '0' && c <= '9') + codepoint -= '0'; + else if (c >= 'A' && c <= 'F') + codepoint -= 'A' - 10; + else if (c >= 'a' && c <= 'f') + codepoint -= 'a' - 10; + else { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorStringUnicodeEscapeInvalidHex, escapeOffset); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(0); + } + is.Take(); + } + return codepoint; + } + + template + class StackStream { + public: + typedef CharType Ch; + + StackStream(internal::Stack& stack) : stack_(stack), length_(0) {} + RAPIDJSON_FORCEINLINE void Put(Ch c) { + *stack_.template Push() = c; + ++length_; + } + + RAPIDJSON_FORCEINLINE void* Push(SizeType count) { + length_ += count; + return stack_.template Push(count); + } + + size_t Length() const { return length_; } + + Ch* Pop() { + return stack_.template Pop(length_); + } + + private: + StackStream(const StackStream&); + StackStream& operator=(const StackStream&); + + internal::Stack& stack_; + SizeType length_; + }; + + // Parse string and generate String event. Different code paths for kParseInsituFlag. + template + void ParseString(InputStream& is, Handler& handler, bool isKey = false) { + internal::StreamLocalCopy copy(is); + InputStream& s(copy.s); + + RAPIDJSON_ASSERT(s.Peek() == '\"'); + s.Take(); // Skip '\"' + + bool success = false; + if (parseFlags & kParseInsituFlag) { + typename InputStream::Ch *head = s.PutBegin(); + ParseStringToStream(s, s); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + size_t length = s.PutEnd(head) - 1; + RAPIDJSON_ASSERT(length <= 0xFFFFFFFF); + const typename TargetEncoding::Ch* const str = reinterpret_cast(head); + success = (isKey ? handler.Key(str, SizeType(length), false) : handler.String(str, SizeType(length), false)); + } + else { + StackStream stackStream(stack_); + ParseStringToStream(s, stackStream); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + SizeType length = static_cast(stackStream.Length()) - 1; + const typename TargetEncoding::Ch* const str = stackStream.Pop(); + success = (isKey ? handler.Key(str, length, true) : handler.String(str, length, true)); + } + if (RAPIDJSON_UNLIKELY(!success)) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, s.Tell()); + } + + // Parse string to an output is + // This function handles the prefix/suffix double quotes, escaping, and optional encoding validation. + template + RAPIDJSON_FORCEINLINE void ParseStringToStream(InputStream& is, OutputStream& os) { +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#define Z16 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 + static const char escape[256] = { + Z16, Z16, 0, 0,'\"', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, '/', + Z16, Z16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,'\\', 0, 0, 0, + 0, 0,'\b', 0, 0, 0,'\f', 0, 0, 0, 0, 0, 0, 0,'\n', 0, + 0, 0,'\r', 0,'\t', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + Z16, Z16, Z16, Z16, Z16, Z16, Z16, Z16 + }; +#undef Z16 +//!@endcond + + for (;;) { + // Scan and copy string before "\\\"" or < 0x20. This is an optional optimzation. + if (!(parseFlags & kParseValidateEncodingFlag)) + ScanCopyUnescapedString(is, os); + + Ch c = is.Peek(); + if (RAPIDJSON_UNLIKELY(c == '\\')) { // Escape + size_t escapeOffset = is.Tell(); // For invalid escaping, report the initial '\\' as error offset + is.Take(); + Ch e = is.Peek(); + if ((sizeof(Ch) == 1 || unsigned(e) < 256) && RAPIDJSON_LIKELY(escape[static_cast(e)])) { + is.Take(); + os.Put(static_cast(escape[static_cast(e)])); + } + else if ((parseFlags & kParseEscapedApostropheFlag) && RAPIDJSON_LIKELY(e == '\'')) { // Allow escaped apostrophe + is.Take(); + os.Put('\''); + } + else if (RAPIDJSON_LIKELY(e == 'u')) { // Unicode + is.Take(); + unsigned codepoint = ParseHex4(is, escapeOffset); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + if (RAPIDJSON_UNLIKELY(codepoint >= 0xD800 && codepoint <= 0xDFFF)) { + // high surrogate, check if followed by valid low surrogate + if (RAPIDJSON_LIKELY(codepoint <= 0xDBFF)) { + // Handle UTF-16 surrogate pair + if (RAPIDJSON_UNLIKELY(!Consume(is, '\\') || !Consume(is, 'u'))) + RAPIDJSON_PARSE_ERROR(kParseErrorStringUnicodeSurrogateInvalid, escapeOffset); + unsigned codepoint2 = ParseHex4(is, escapeOffset); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN_VOID; + if (RAPIDJSON_UNLIKELY(codepoint2 < 0xDC00 || codepoint2 > 0xDFFF)) + RAPIDJSON_PARSE_ERROR(kParseErrorStringUnicodeSurrogateInvalid, escapeOffset); + codepoint = (((codepoint - 0xD800) << 10) | (codepoint2 - 0xDC00)) + 0x10000; + } + // single low surrogate + else + { + RAPIDJSON_PARSE_ERROR(kParseErrorStringUnicodeSurrogateInvalid, escapeOffset); + } + } + TEncoding::Encode(os, codepoint); + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorStringEscapeInvalid, escapeOffset); + } + else if (RAPIDJSON_UNLIKELY(c == '"')) { // Closing double quote + is.Take(); + os.Put('\0'); // null-terminate the string + return; + } + else if (RAPIDJSON_UNLIKELY(static_cast(c) < 0x20)) { // RFC 4627: unescaped = %x20-21 / %x23-5B / %x5D-10FFFF + if (c == '\0') + RAPIDJSON_PARSE_ERROR(kParseErrorStringMissQuotationMark, is.Tell()); + else + RAPIDJSON_PARSE_ERROR(kParseErrorStringInvalidEncoding, is.Tell()); + } + else { + size_t offset = is.Tell(); + if (RAPIDJSON_UNLIKELY((parseFlags & kParseValidateEncodingFlag ? + !Transcoder::Validate(is, os) : + !Transcoder::Transcode(is, os)))) + RAPIDJSON_PARSE_ERROR(kParseErrorStringInvalidEncoding, offset); + } + } + } + + template + static RAPIDJSON_FORCEINLINE void ScanCopyUnescapedString(InputStream&, OutputStream&) { + // Do nothing for generic version + } + +#if defined(RAPIDJSON_SSE2) || defined(RAPIDJSON_SSE42) + // StringStream -> StackStream + static RAPIDJSON_FORCEINLINE void ScanCopyUnescapedString(StringStream& is, StackStream& os) { + const char* p = is.src_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = p; + return; + } + else + os.Put(*p++); + + // The rest of string using SIMD + static const char dquote[16] = { '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"' }; + static const char bslash[16] = { '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\' }; + static const char space[16] = { 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F }; + const __m128i dq = _mm_loadu_si128(reinterpret_cast(&dquote[0])); + const __m128i bs = _mm_loadu_si128(reinterpret_cast(&bslash[0])); + const __m128i sp = _mm_loadu_si128(reinterpret_cast(&space[0])); + + for (;; p += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + const __m128i t1 = _mm_cmpeq_epi8(s, dq); + const __m128i t2 = _mm_cmpeq_epi8(s, bs); + const __m128i t3 = _mm_cmpeq_epi8(_mm_max_epu8(s, sp), sp); // s < 0x20 <=> max(s, 0x1F) == 0x1F + const __m128i x = _mm_or_si128(_mm_or_si128(t1, t2), t3); + unsigned short r = static_cast(_mm_movemask_epi8(x)); + if (RAPIDJSON_UNLIKELY(r != 0)) { // some of characters is escaped + SizeType length; + #ifdef _MSC_VER // Find the index of first escaped + unsigned long offset; + _BitScanForward(&offset, r); + length = offset; + #else + length = static_cast(__builtin_ffs(r) - 1); + #endif + if (length != 0) { + char* q = reinterpret_cast(os.Push(length)); + for (size_t i = 0; i < length; i++) + q[i] = p[i]; + + p += length; + } + break; + } + _mm_storeu_si128(reinterpret_cast<__m128i *>(os.Push(16)), s); + } + + is.src_ = p; + } + + // InsituStringStream -> InsituStringStream + static RAPIDJSON_FORCEINLINE void ScanCopyUnescapedString(InsituStringStream& is, InsituStringStream& os) { + RAPIDJSON_ASSERT(&is == &os); + (void)os; + + if (is.src_ == is.dst_) { + SkipUnescapedString(is); + return; + } + + char* p = is.src_; + char *q = is.dst_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = p; + is.dst_ = q; + return; + } + else + *q++ = *p++; + + // The rest of string using SIMD + static const char dquote[16] = { '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"' }; + static const char bslash[16] = { '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\' }; + static const char space[16] = { 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F }; + const __m128i dq = _mm_loadu_si128(reinterpret_cast(&dquote[0])); + const __m128i bs = _mm_loadu_si128(reinterpret_cast(&bslash[0])); + const __m128i sp = _mm_loadu_si128(reinterpret_cast(&space[0])); + + for (;; p += 16, q += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + const __m128i t1 = _mm_cmpeq_epi8(s, dq); + const __m128i t2 = _mm_cmpeq_epi8(s, bs); + const __m128i t3 = _mm_cmpeq_epi8(_mm_max_epu8(s, sp), sp); // s < 0x20 <=> max(s, 0x1F) == 0x1F + const __m128i x = _mm_or_si128(_mm_or_si128(t1, t2), t3); + unsigned short r = static_cast(_mm_movemask_epi8(x)); + if (RAPIDJSON_UNLIKELY(r != 0)) { // some of characters is escaped + size_t length; +#ifdef _MSC_VER // Find the index of first escaped + unsigned long offset; + _BitScanForward(&offset, r); + length = offset; +#else + length = static_cast(__builtin_ffs(r) - 1); +#endif + for (const char* pend = p + length; p != pend; ) + *q++ = *p++; + break; + } + _mm_storeu_si128(reinterpret_cast<__m128i *>(q), s); + } + + is.src_ = p; + is.dst_ = q; + } + + // When read/write pointers are the same for insitu stream, just skip unescaped characters + static RAPIDJSON_FORCEINLINE void SkipUnescapedString(InsituStringStream& is) { + RAPIDJSON_ASSERT(is.src_ == is.dst_); + char* p = is.src_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + for (; p != nextAligned; p++) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = is.dst_ = p; + return; + } + + // The rest of string using SIMD + static const char dquote[16] = { '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"' }; + static const char bslash[16] = { '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\' }; + static const char space[16] = { 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F }; + const __m128i dq = _mm_loadu_si128(reinterpret_cast(&dquote[0])); + const __m128i bs = _mm_loadu_si128(reinterpret_cast(&bslash[0])); + const __m128i sp = _mm_loadu_si128(reinterpret_cast(&space[0])); + + for (;; p += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + const __m128i t1 = _mm_cmpeq_epi8(s, dq); + const __m128i t2 = _mm_cmpeq_epi8(s, bs); + const __m128i t3 = _mm_cmpeq_epi8(_mm_max_epu8(s, sp), sp); // s < 0x20 <=> max(s, 0x1F) == 0x1F + const __m128i x = _mm_or_si128(_mm_or_si128(t1, t2), t3); + unsigned short r = static_cast(_mm_movemask_epi8(x)); + if (RAPIDJSON_UNLIKELY(r != 0)) { // some of characters is escaped + size_t length; +#ifdef _MSC_VER // Find the index of first escaped + unsigned long offset; + _BitScanForward(&offset, r); + length = offset; +#else + length = static_cast(__builtin_ffs(r) - 1); +#endif + p += length; + break; + } + } + + is.src_ = is.dst_ = p; + } +#elif defined(RAPIDJSON_NEON) + // StringStream -> StackStream + static RAPIDJSON_FORCEINLINE void ScanCopyUnescapedString(StringStream& is, StackStream& os) { + const char* p = is.src_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = p; + return; + } + else + os.Put(*p++); + + // The rest of string using SIMD + const uint8x16_t s0 = vmovq_n_u8('"'); + const uint8x16_t s1 = vmovq_n_u8('\\'); + const uint8x16_t s2 = vmovq_n_u8('\b'); + const uint8x16_t s3 = vmovq_n_u8(32); + + for (;; p += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, s0); + x = vorrq_u8(x, vceqq_u8(s, s1)); + x = vorrq_u8(x, vceqq_u8(s, s2)); + x = vorrq_u8(x, vcltq_u8(s, s3)); + + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + SizeType length = 0; + bool escaped = false; + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + length = 8 + (lz >> 3); + escaped = true; + } + } else { + uint32_t lz = internal::clzll(low); + length = lz >> 3; + escaped = true; + } + if (RAPIDJSON_UNLIKELY(escaped)) { // some of characters is escaped + if (length != 0) { + char* q = reinterpret_cast(os.Push(length)); + for (size_t i = 0; i < length; i++) + q[i] = p[i]; + + p += length; + } + break; + } + vst1q_u8(reinterpret_cast(os.Push(16)), s); + } + + is.src_ = p; + } + + // InsituStringStream -> InsituStringStream + static RAPIDJSON_FORCEINLINE void ScanCopyUnescapedString(InsituStringStream& is, InsituStringStream& os) { + RAPIDJSON_ASSERT(&is == &os); + (void)os; + + if (is.src_ == is.dst_) { + SkipUnescapedString(is); + return; + } + + char* p = is.src_; + char *q = is.dst_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + while (p != nextAligned) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = p; + is.dst_ = q; + return; + } + else + *q++ = *p++; + + // The rest of string using SIMD + const uint8x16_t s0 = vmovq_n_u8('"'); + const uint8x16_t s1 = vmovq_n_u8('\\'); + const uint8x16_t s2 = vmovq_n_u8('\b'); + const uint8x16_t s3 = vmovq_n_u8(32); + + for (;; p += 16, q += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, s0); + x = vorrq_u8(x, vceqq_u8(s, s1)); + x = vorrq_u8(x, vceqq_u8(s, s2)); + x = vorrq_u8(x, vcltq_u8(s, s3)); + + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + SizeType length = 0; + bool escaped = false; + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + length = 8 + (lz >> 3); + escaped = true; + } + } else { + uint32_t lz = internal::clzll(low); + length = lz >> 3; + escaped = true; + } + if (RAPIDJSON_UNLIKELY(escaped)) { // some of characters is escaped + for (const char* pend = p + length; p != pend; ) { + *q++ = *p++; + } + break; + } + vst1q_u8(reinterpret_cast(q), s); + } + + is.src_ = p; + is.dst_ = q; + } + + // When read/write pointers are the same for insitu stream, just skip unescaped characters + static RAPIDJSON_FORCEINLINE void SkipUnescapedString(InsituStringStream& is) { + RAPIDJSON_ASSERT(is.src_ == is.dst_); + char* p = is.src_; + + // Scan one by one until alignment (unaligned load may cross page boundary and cause crash) + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + for (; p != nextAligned; p++) + if (RAPIDJSON_UNLIKELY(*p == '\"') || RAPIDJSON_UNLIKELY(*p == '\\') || RAPIDJSON_UNLIKELY(static_cast(*p) < 0x20)) { + is.src_ = is.dst_ = p; + return; + } + + // The rest of string using SIMD + const uint8x16_t s0 = vmovq_n_u8('"'); + const uint8x16_t s1 = vmovq_n_u8('\\'); + const uint8x16_t s2 = vmovq_n_u8('\b'); + const uint8x16_t s3 = vmovq_n_u8(32); + + for (;; p += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, s0); + x = vorrq_u8(x, vceqq_u8(s, s1)); + x = vorrq_u8(x, vceqq_u8(s, s2)); + x = vorrq_u8(x, vcltq_u8(s, s3)); + + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + p += 8 + (lz >> 3); + break; + } + } else { + uint32_t lz = internal::clzll(low); + p += lz >> 3; + break; + } + } + + is.src_ = is.dst_ = p; + } +#endif // RAPIDJSON_NEON + + template + class NumberStream; + + template + class NumberStream { + public: + typedef typename InputStream::Ch Ch; + + NumberStream(GenericReader& reader, InputStream& s) : is(s) { (void)reader; } + + RAPIDJSON_FORCEINLINE Ch Peek() const { return is.Peek(); } + RAPIDJSON_FORCEINLINE Ch TakePush() { return is.Take(); } + RAPIDJSON_FORCEINLINE Ch Take() { return is.Take(); } + RAPIDJSON_FORCEINLINE void Push(char) {} + + size_t Tell() { return is.Tell(); } + size_t Length() { return 0; } + const StackCharacter* Pop() { return 0; } + + protected: + NumberStream& operator=(const NumberStream&); + + InputStream& is; + }; + + template + class NumberStream : public NumberStream { + typedef NumberStream Base; + public: + NumberStream(GenericReader& reader, InputStream& s) : Base(reader, s), stackStream(reader.stack_) {} + + RAPIDJSON_FORCEINLINE Ch TakePush() { + stackStream.Put(static_cast(Base::is.Peek())); + return Base::is.Take(); + } + + RAPIDJSON_FORCEINLINE void Push(StackCharacter c) { + stackStream.Put(c); + } + + size_t Length() { return stackStream.Length(); } + + const StackCharacter* Pop() { + stackStream.Put('\0'); + return stackStream.Pop(); + } + + private: + StackStream stackStream; + }; + + template + class NumberStream : public NumberStream { + typedef NumberStream Base; + public: + NumberStream(GenericReader& reader, InputStream& s) : Base(reader, s) {} + + RAPIDJSON_FORCEINLINE Ch Take() { return Base::TakePush(); } + }; + + template + void ParseNumber(InputStream& is, Handler& handler) { + typedef typename internal::SelectIf, typename TargetEncoding::Ch, char>::Type NumberCharacter; + + internal::StreamLocalCopy copy(is); + NumberStream s(*this, copy.s); + + size_t startOffset = s.Tell(); + double d = 0.0; + bool useNanOrInf = false; + + // Parse minus + bool minus = Consume(s, '-'); + + // Parse int: zero / ( digit1-9 *DIGIT ) + unsigned i = 0; + uint64_t i64 = 0; + bool use64bit = false; + int significandDigit = 0; + if (RAPIDJSON_UNLIKELY(s.Peek() == '0')) { + i = 0; + s.TakePush(); + } + else if (RAPIDJSON_LIKELY(s.Peek() >= '1' && s.Peek() <= '9')) { + i = static_cast(s.TakePush() - '0'); + + if (minus) + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (RAPIDJSON_UNLIKELY(i >= 214748364)) { // 2^31 = 2147483648 + if (RAPIDJSON_LIKELY(i != 214748364 || s.Peek() > '8')) { + i64 = i; + use64bit = true; + break; + } + } + i = i * 10 + static_cast(s.TakePush() - '0'); + significandDigit++; + } + else + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (RAPIDJSON_UNLIKELY(i >= 429496729)) { // 2^32 - 1 = 4294967295 + if (RAPIDJSON_LIKELY(i != 429496729 || s.Peek() > '5')) { + i64 = i; + use64bit = true; + break; + } + } + i = i * 10 + static_cast(s.TakePush() - '0'); + significandDigit++; + } + } + // Parse NaN or Infinity here + else if ((parseFlags & kParseNanAndInfFlag) && RAPIDJSON_LIKELY((s.Peek() == 'I' || s.Peek() == 'N'))) { + if (Consume(s, 'N')) { + if (Consume(s, 'a') && Consume(s, 'N')) { + d = std::numeric_limits::quiet_NaN(); + useNanOrInf = true; + } + } + else if (RAPIDJSON_LIKELY(Consume(s, 'I'))) { + if (Consume(s, 'n') && Consume(s, 'f')) { + d = (minus ? -std::numeric_limits::infinity() : std::numeric_limits::infinity()); + useNanOrInf = true; + + if (RAPIDJSON_UNLIKELY(s.Peek() == 'i' && !(Consume(s, 'i') && Consume(s, 'n') + && Consume(s, 'i') && Consume(s, 't') && Consume(s, 'y')))) { + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, s.Tell()); + } + } + } + + if (RAPIDJSON_UNLIKELY(!useNanOrInf)) { + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, s.Tell()); + } + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, s.Tell()); + + // Parse 64bit int + bool useDouble = false; + if (use64bit) { + if (minus) + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (RAPIDJSON_UNLIKELY(i64 >= RAPIDJSON_UINT64_C2(0x0CCCCCCC, 0xCCCCCCCC))) // 2^63 = 9223372036854775808 + if (RAPIDJSON_LIKELY(i64 != RAPIDJSON_UINT64_C2(0x0CCCCCCC, 0xCCCCCCCC) || s.Peek() > '8')) { + d = static_cast(i64); + useDouble = true; + break; + } + i64 = i64 * 10 + static_cast(s.TakePush() - '0'); + significandDigit++; + } + else + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (RAPIDJSON_UNLIKELY(i64 >= RAPIDJSON_UINT64_C2(0x19999999, 0x99999999))) // 2^64 - 1 = 18446744073709551615 + if (RAPIDJSON_LIKELY(i64 != RAPIDJSON_UINT64_C2(0x19999999, 0x99999999) || s.Peek() > '5')) { + d = static_cast(i64); + useDouble = true; + break; + } + i64 = i64 * 10 + static_cast(s.TakePush() - '0'); + significandDigit++; + } + } + + // Force double for big integer + if (useDouble) { + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + d = d * 10 + (s.TakePush() - '0'); + } + } + + // Parse frac = decimal-point 1*DIGIT + int expFrac = 0; + size_t decimalPosition; + if (!useNanOrInf && Consume(s, '.')) { + decimalPosition = s.Length(); + + if (RAPIDJSON_UNLIKELY(!(s.Peek() >= '0' && s.Peek() <= '9'))) + RAPIDJSON_PARSE_ERROR(kParseErrorNumberMissFraction, s.Tell()); + + if (!useDouble) { +#if RAPIDJSON_64BIT + // Use i64 to store significand in 64-bit architecture + if (!use64bit) + i64 = i; + + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (i64 > RAPIDJSON_UINT64_C2(0x1FFFFF, 0xFFFFFFFF)) // 2^53 - 1 for fast path + break; + else { + i64 = i64 * 10 + static_cast(s.TakePush() - '0'); + --expFrac; + if (i64 != 0) + significandDigit++; + } + } + + d = static_cast(i64); +#else + // Use double to store significand in 32-bit architecture + d = static_cast(use64bit ? i64 : i); +#endif + useDouble = true; + } + + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + if (significandDigit < 17) { + d = d * 10.0 + (s.TakePush() - '0'); + --expFrac; + if (RAPIDJSON_LIKELY(d > 0.0)) + significandDigit++; + } + else + s.TakePush(); + } + } + else + decimalPosition = s.Length(); // decimal position at the end of integer. + + // Parse exp = e [ minus / plus ] 1*DIGIT + int exp = 0; + if (!useNanOrInf && (Consume(s, 'e') || Consume(s, 'E'))) { + if (!useDouble) { + d = static_cast(use64bit ? i64 : i); + useDouble = true; + } + + bool expMinus = false; + if (Consume(s, '+')) + ; + else if (Consume(s, '-')) + expMinus = true; + + if (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + exp = static_cast(s.Take() - '0'); + if (expMinus) { + // (exp + expFrac) must not underflow int => we're detecting when -exp gets + // dangerously close to INT_MIN (a pessimistic next digit 9 would push it into + // underflow territory): + // + // -(exp * 10 + 9) + expFrac >= INT_MIN + // <=> exp <= (expFrac - INT_MIN - 9) / 10 + RAPIDJSON_ASSERT(expFrac <= 0); + int maxExp = (expFrac + 2147483639) / 10; + + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + exp = exp * 10 + static_cast(s.Take() - '0'); + if (RAPIDJSON_UNLIKELY(exp > maxExp)) { + while (RAPIDJSON_UNLIKELY(s.Peek() >= '0' && s.Peek() <= '9')) // Consume the rest of exponent + s.Take(); + } + } + } + else { // positive exp + int maxExp = 308 - expFrac; + while (RAPIDJSON_LIKELY(s.Peek() >= '0' && s.Peek() <= '9')) { + exp = exp * 10 + static_cast(s.Take() - '0'); + if (RAPIDJSON_UNLIKELY(exp > maxExp)) + RAPIDJSON_PARSE_ERROR(kParseErrorNumberTooBig, startOffset); + } + } + } + else + RAPIDJSON_PARSE_ERROR(kParseErrorNumberMissExponent, s.Tell()); + + if (expMinus) + exp = -exp; + } + + // Finish parsing, call event according to the type of number. + bool cont = true; + + if (parseFlags & kParseNumbersAsStringsFlag) { + if (parseFlags & kParseInsituFlag) { + s.Pop(); // Pop stack no matter if it will be used or not. + typename InputStream::Ch* head = is.PutBegin(); + const size_t length = s.Tell() - startOffset; + RAPIDJSON_ASSERT(length <= 0xFFFFFFFF); + // unable to insert the \0 character here, it will erase the comma after this number + const typename TargetEncoding::Ch* const str = reinterpret_cast(head); + cont = handler.RawNumber(str, SizeType(length), false); + } + else { + SizeType numCharsToCopy = static_cast(s.Length()); + GenericStringStream > srcStream(s.Pop()); + StackStream dstStream(stack_); + while (numCharsToCopy--) { + Transcoder, TargetEncoding>::Transcode(srcStream, dstStream); + } + dstStream.Put('\0'); + const typename TargetEncoding::Ch* str = dstStream.Pop(); + const SizeType length = static_cast(dstStream.Length()) - 1; + cont = handler.RawNumber(str, SizeType(length), true); + } + } + else { + size_t length = s.Length(); + const NumberCharacter* decimal = s.Pop(); // Pop stack no matter if it will be used or not. + + if (useDouble) { + int p = exp + expFrac; + if (parseFlags & kParseFullPrecisionFlag) + d = internal::StrtodFullPrecision(d, p, decimal, length, decimalPosition, exp); + else + d = internal::StrtodNormalPrecision(d, p); + + // Use > max, instead of == inf, to fix bogus warning -Wfloat-equal + if (d > (std::numeric_limits::max)()) { + // Overflow + // TODO: internal::StrtodX should report overflow (or underflow) + RAPIDJSON_PARSE_ERROR(kParseErrorNumberTooBig, startOffset); + } + + cont = handler.Double(minus ? -d : d); + } + else if (useNanOrInf) { + cont = handler.Double(d); + } + else { + if (use64bit) { + if (minus) + cont = handler.Int64(static_cast(~i64 + 1)); + else + cont = handler.Uint64(i64); + } + else { + if (minus) + cont = handler.Int(static_cast(~i + 1)); + else + cont = handler.Uint(i); + } + } + } + if (RAPIDJSON_UNLIKELY(!cont)) + RAPIDJSON_PARSE_ERROR(kParseErrorTermination, startOffset); + } + + // Parse any JSON value + template + void ParseValue(InputStream& is, Handler& handler) { + switch (is.Peek()) { + case 'n': ParseNull (is, handler); break; + case 't': ParseTrue (is, handler); break; + case 'f': ParseFalse (is, handler); break; + case '"': ParseString(is, handler); break; + case '{': ParseObject(is, handler); break; + case '[': ParseArray (is, handler); break; + default : + ParseNumber(is, handler); + break; + + } + } + + // Iterative Parsing + + // States + enum IterativeParsingState { + IterativeParsingFinishState = 0, // sink states at top + IterativeParsingErrorState, // sink states at top + IterativeParsingStartState, + + // Object states + IterativeParsingObjectInitialState, + IterativeParsingMemberKeyState, + IterativeParsingMemberValueState, + IterativeParsingObjectFinishState, + + // Array states + IterativeParsingArrayInitialState, + IterativeParsingElementState, + IterativeParsingArrayFinishState, + + // Single value state + IterativeParsingValueState, + + // Delimiter states (at bottom) + IterativeParsingElementDelimiterState, + IterativeParsingMemberDelimiterState, + IterativeParsingKeyValueDelimiterState, + + cIterativeParsingStateCount + }; + + // Tokens + enum Token { + LeftBracketToken = 0, + RightBracketToken, + + LeftCurlyBracketToken, + RightCurlyBracketToken, + + CommaToken, + ColonToken, + + StringToken, + FalseToken, + TrueToken, + NullToken, + NumberToken, + + kTokenCount + }; + + RAPIDJSON_FORCEINLINE Token Tokenize(Ch c) const { + +//!@cond RAPIDJSON_HIDDEN_FROM_DOXYGEN +#define N NumberToken +#define N16 N,N,N,N,N,N,N,N,N,N,N,N,N,N,N,N + // Maps from ASCII to Token + static const unsigned char tokenMap[256] = { + N16, // 00~0F + N16, // 10~1F + N, N, StringToken, N, N, N, N, N, N, N, N, N, CommaToken, N, N, N, // 20~2F + N, N, N, N, N, N, N, N, N, N, ColonToken, N, N, N, N, N, // 30~3F + N16, // 40~4F + N, N, N, N, N, N, N, N, N, N, N, LeftBracketToken, N, RightBracketToken, N, N, // 50~5F + N, N, N, N, N, N, FalseToken, N, N, N, N, N, N, N, NullToken, N, // 60~6F + N, N, N, N, TrueToken, N, N, N, N, N, N, LeftCurlyBracketToken, N, RightCurlyBracketToken, N, N, // 70~7F + N16, N16, N16, N16, N16, N16, N16, N16 // 80~FF + }; +#undef N +#undef N16 +//!@endcond + + if (sizeof(Ch) == 1 || static_cast(c) < 256) + return static_cast(tokenMap[static_cast(c)]); + else + return NumberToken; + } + + RAPIDJSON_FORCEINLINE IterativeParsingState Predict(IterativeParsingState state, Token token) const { + // current state x one lookahead token -> new state + static const char G[cIterativeParsingStateCount][kTokenCount] = { + // Finish(sink state) + { + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState + }, + // Error(sink state) + { + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState + }, + // Start + { + IterativeParsingArrayInitialState, // Left bracket + IterativeParsingErrorState, // Right bracket + IterativeParsingObjectInitialState, // Left curly bracket + IterativeParsingErrorState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingValueState, // String + IterativeParsingValueState, // False + IterativeParsingValueState, // True + IterativeParsingValueState, // Null + IterativeParsingValueState // Number + }, + // ObjectInitial + { + IterativeParsingErrorState, // Left bracket + IterativeParsingErrorState, // Right bracket + IterativeParsingErrorState, // Left curly bracket + IterativeParsingObjectFinishState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingMemberKeyState, // String + IterativeParsingErrorState, // False + IterativeParsingErrorState, // True + IterativeParsingErrorState, // Null + IterativeParsingErrorState // Number + }, + // MemberKey + { + IterativeParsingErrorState, // Left bracket + IterativeParsingErrorState, // Right bracket + IterativeParsingErrorState, // Left curly bracket + IterativeParsingErrorState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingKeyValueDelimiterState, // Colon + IterativeParsingErrorState, // String + IterativeParsingErrorState, // False + IterativeParsingErrorState, // True + IterativeParsingErrorState, // Null + IterativeParsingErrorState // Number + }, + // MemberValue + { + IterativeParsingErrorState, // Left bracket + IterativeParsingErrorState, // Right bracket + IterativeParsingErrorState, // Left curly bracket + IterativeParsingObjectFinishState, // Right curly bracket + IterativeParsingMemberDelimiterState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingErrorState, // String + IterativeParsingErrorState, // False + IterativeParsingErrorState, // True + IterativeParsingErrorState, // Null + IterativeParsingErrorState // Number + }, + // ObjectFinish(sink state) + { + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState + }, + // ArrayInitial + { + IterativeParsingArrayInitialState, // Left bracket(push Element state) + IterativeParsingArrayFinishState, // Right bracket + IterativeParsingObjectInitialState, // Left curly bracket(push Element state) + IterativeParsingErrorState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingElementState, // String + IterativeParsingElementState, // False + IterativeParsingElementState, // True + IterativeParsingElementState, // Null + IterativeParsingElementState // Number + }, + // Element + { + IterativeParsingErrorState, // Left bracket + IterativeParsingArrayFinishState, // Right bracket + IterativeParsingErrorState, // Left curly bracket + IterativeParsingErrorState, // Right curly bracket + IterativeParsingElementDelimiterState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingErrorState, // String + IterativeParsingErrorState, // False + IterativeParsingErrorState, // True + IterativeParsingErrorState, // Null + IterativeParsingErrorState // Number + }, + // ArrayFinish(sink state) + { + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState + }, + // Single Value (sink state) + { + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, IterativeParsingErrorState, + IterativeParsingErrorState + }, + // ElementDelimiter + { + IterativeParsingArrayInitialState, // Left bracket(push Element state) + IterativeParsingArrayFinishState, // Right bracket + IterativeParsingObjectInitialState, // Left curly bracket(push Element state) + IterativeParsingErrorState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingElementState, // String + IterativeParsingElementState, // False + IterativeParsingElementState, // True + IterativeParsingElementState, // Null + IterativeParsingElementState // Number + }, + // MemberDelimiter + { + IterativeParsingErrorState, // Left bracket + IterativeParsingErrorState, // Right bracket + IterativeParsingErrorState, // Left curly bracket + IterativeParsingObjectFinishState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingMemberKeyState, // String + IterativeParsingErrorState, // False + IterativeParsingErrorState, // True + IterativeParsingErrorState, // Null + IterativeParsingErrorState // Number + }, + // KeyValueDelimiter + { + IterativeParsingArrayInitialState, // Left bracket(push MemberValue state) + IterativeParsingErrorState, // Right bracket + IterativeParsingObjectInitialState, // Left curly bracket(push MemberValue state) + IterativeParsingErrorState, // Right curly bracket + IterativeParsingErrorState, // Comma + IterativeParsingErrorState, // Colon + IterativeParsingMemberValueState, // String + IterativeParsingMemberValueState, // False + IterativeParsingMemberValueState, // True + IterativeParsingMemberValueState, // Null + IterativeParsingMemberValueState // Number + }, + }; // End of G + + return static_cast(G[state][token]); + } + + // Make an advance in the token stream and state based on the candidate destination state which was returned by Transit(). + // May return a new state on state pop. + template + RAPIDJSON_FORCEINLINE IterativeParsingState Transit(IterativeParsingState src, Token token, IterativeParsingState dst, InputStream& is, Handler& handler) { + (void)token; + + switch (dst) { + case IterativeParsingErrorState: + return dst; + + case IterativeParsingObjectInitialState: + case IterativeParsingArrayInitialState: + { + // Push the state(Element or MemeberValue) if we are nested in another array or value of member. + // In this way we can get the correct state on ObjectFinish or ArrayFinish by frame pop. + IterativeParsingState n = src; + if (src == IterativeParsingArrayInitialState || src == IterativeParsingElementDelimiterState) + n = IterativeParsingElementState; + else if (src == IterativeParsingKeyValueDelimiterState) + n = IterativeParsingMemberValueState; + // Push current state. + *stack_.template Push(1) = n; + // Initialize and push the member/element count. + *stack_.template Push(1) = 0; + // Call handler + bool hr = (dst == IterativeParsingObjectInitialState) ? handler.StartObject() : handler.StartArray(); + // On handler short circuits the parsing. + if (!hr) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorTermination, is.Tell()); + return IterativeParsingErrorState; + } + else { + is.Take(); + return dst; + } + } + + case IterativeParsingMemberKeyState: + ParseString(is, handler, true); + if (HasParseError()) + return IterativeParsingErrorState; + else + return dst; + + case IterativeParsingKeyValueDelimiterState: + RAPIDJSON_ASSERT(token == ColonToken); + is.Take(); + return dst; + + case IterativeParsingMemberValueState: + // Must be non-compound value. Or it would be ObjectInitial or ArrayInitial state. + ParseValue(is, handler); + if (HasParseError()) { + return IterativeParsingErrorState; + } + return dst; + + case IterativeParsingElementState: + // Must be non-compound value. Or it would be ObjectInitial or ArrayInitial state. + ParseValue(is, handler); + if (HasParseError()) { + return IterativeParsingErrorState; + } + return dst; + + case IterativeParsingMemberDelimiterState: + case IterativeParsingElementDelimiterState: + is.Take(); + // Update member/element count. + *stack_.template Top() = *stack_.template Top() + 1; + return dst; + + case IterativeParsingObjectFinishState: + { + // Transit from delimiter is only allowed when trailing commas are enabled + if (!(parseFlags & kParseTrailingCommasFlag) && src == IterativeParsingMemberDelimiterState) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorObjectMissName, is.Tell()); + return IterativeParsingErrorState; + } + // Get member count. + SizeType c = *stack_.template Pop(1); + // If the object is not empty, count the last member. + if (src == IterativeParsingMemberValueState) + ++c; + // Restore the state. + IterativeParsingState n = static_cast(*stack_.template Pop(1)); + // Transit to Finish state if this is the topmost scope. + if (n == IterativeParsingStartState) + n = IterativeParsingFinishState; + // Call handler + bool hr = handler.EndObject(c); + // On handler short circuits the parsing. + if (!hr) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorTermination, is.Tell()); + return IterativeParsingErrorState; + } + else { + is.Take(); + return n; + } + } + + case IterativeParsingArrayFinishState: + { + // Transit from delimiter is only allowed when trailing commas are enabled + if (!(parseFlags & kParseTrailingCommasFlag) && src == IterativeParsingElementDelimiterState) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorValueInvalid, is.Tell()); + return IterativeParsingErrorState; + } + // Get element count. + SizeType c = *stack_.template Pop(1); + // If the array is not empty, count the last element. + if (src == IterativeParsingElementState) + ++c; + // Restore the state. + IterativeParsingState n = static_cast(*stack_.template Pop(1)); + // Transit to Finish state if this is the topmost scope. + if (n == IterativeParsingStartState) + n = IterativeParsingFinishState; + // Call handler + bool hr = handler.EndArray(c); + // On handler short circuits the parsing. + if (!hr) { + RAPIDJSON_PARSE_ERROR_NORETURN(kParseErrorTermination, is.Tell()); + return IterativeParsingErrorState; + } + else { + is.Take(); + return n; + } + } + + default: + // This branch is for IterativeParsingValueState actually. + // Use `default:` rather than + // `case IterativeParsingValueState:` is for code coverage. + + // The IterativeParsingStartState is not enumerated in this switch-case. + // It is impossible for that case. And it can be caught by following assertion. + + // The IterativeParsingFinishState is not enumerated in this switch-case either. + // It is a "derivative" state which cannot triggered from Predict() directly. + // Therefore it cannot happen here. And it can be caught by following assertion. + RAPIDJSON_ASSERT(dst == IterativeParsingValueState); + + // Must be non-compound value. Or it would be ObjectInitial or ArrayInitial state. + ParseValue(is, handler); + if (HasParseError()) { + return IterativeParsingErrorState; + } + return IterativeParsingFinishState; + } + } + + template + void HandleError(IterativeParsingState src, InputStream& is) { + if (HasParseError()) { + // Error flag has been set. + return; + } + + switch (src) { + case IterativeParsingStartState: RAPIDJSON_PARSE_ERROR(kParseErrorDocumentEmpty, is.Tell()); return; + case IterativeParsingFinishState: RAPIDJSON_PARSE_ERROR(kParseErrorDocumentRootNotSingular, is.Tell()); return; + case IterativeParsingObjectInitialState: + case IterativeParsingMemberDelimiterState: RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissName, is.Tell()); return; + case IterativeParsingMemberKeyState: RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissColon, is.Tell()); return; + case IterativeParsingMemberValueState: RAPIDJSON_PARSE_ERROR(kParseErrorObjectMissCommaOrCurlyBracket, is.Tell()); return; + case IterativeParsingKeyValueDelimiterState: + case IterativeParsingArrayInitialState: + case IterativeParsingElementDelimiterState: RAPIDJSON_PARSE_ERROR(kParseErrorValueInvalid, is.Tell()); return; + default: RAPIDJSON_ASSERT(src == IterativeParsingElementState); RAPIDJSON_PARSE_ERROR(kParseErrorArrayMissCommaOrSquareBracket, is.Tell()); return; + } + } + + RAPIDJSON_FORCEINLINE bool IsIterativeParsingDelimiterState(IterativeParsingState s) const { + return s >= IterativeParsingElementDelimiterState; + } + + RAPIDJSON_FORCEINLINE bool IsIterativeParsingCompleteState(IterativeParsingState s) const { + return s <= IterativeParsingErrorState; + } + + template + ParseResult IterativeParse(InputStream& is, Handler& handler) { + parseResult_.Clear(); + ClearStackOnExit scope(*this); + IterativeParsingState state = IterativeParsingStartState; + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + while (is.Peek() != '\0') { + Token t = Tokenize(is.Peek()); + IterativeParsingState n = Predict(state, t); + IterativeParsingState d = Transit(state, t, n, is, handler); + + if (d == IterativeParsingErrorState) { + HandleError(state, is); + break; + } + + state = d; + + // Do not further consume streams if a root JSON has been parsed. + if ((parseFlags & kParseStopWhenDoneFlag) && state == IterativeParsingFinishState) + break; + + SkipWhitespaceAndComments(is); + RAPIDJSON_PARSE_ERROR_EARLY_RETURN(parseResult_); + } + + // Handle the end of file. + if (state != IterativeParsingFinishState) + HandleError(state, is); + + return parseResult_; + } + + static const size_t kDefaultStackCapacity = 256; //!< Default stack capacity in bytes for storing a single decoded string. + internal::Stack stack_; //!< A stack for storing decoded string temporarily during non-destructive parsing. + ParseResult parseResult_; + IterativeParsingState state_; +}; // class GenericReader + +//! Reader with UTF8 encoding and default allocator. +typedef GenericReader, UTF8<> > Reader; + +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) || defined(_MSC_VER) +RAPIDJSON_DIAG_POP +#endif + + +#ifdef __GNUC__ +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_READER_H_ diff --git a/include/rapidjson/schema.h b/include/rapidjson/schema.h new file mode 100644 index 0000000..f049285 --- /dev/null +++ b/include/rapidjson/schema.h @@ -0,0 +1,3261 @@ +// Tencent is pleased to support the open source community by making RapidJSON available-> +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip-> All rights reserved-> +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License-> You may obtain a copy of the License at +// +// http://opensource->org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied-> See the License for the +// specific language governing permissions and limitations under the License-> + +#ifndef RAPIDJSON_SCHEMA_H_ +#define RAPIDJSON_SCHEMA_H_ + +#include "document.h" +#include "pointer.h" +#include "stringbuffer.h" +#include "error/en.h" +#include "uri.h" +#include // abs, floor + +#if !defined(RAPIDJSON_SCHEMA_USE_INTERNALREGEX) +#define RAPIDJSON_SCHEMA_USE_INTERNALREGEX 1 +#endif + +#if !defined(RAPIDJSON_SCHEMA_USE_STDREGEX) || !(__cplusplus >=201103L || (defined(_MSC_VER) && _MSC_VER >= 1800)) +#define RAPIDJSON_SCHEMA_USE_STDREGEX 0 +#endif + +#if RAPIDJSON_SCHEMA_USE_INTERNALREGEX +#include "internal/regex.h" +#elif RAPIDJSON_SCHEMA_USE_STDREGEX +#include +#endif + +#if RAPIDJSON_SCHEMA_USE_INTERNALREGEX || RAPIDJSON_SCHEMA_USE_STDREGEX +#define RAPIDJSON_SCHEMA_HAS_REGEX 1 +#else +#define RAPIDJSON_SCHEMA_HAS_REGEX 0 +#endif + +#ifndef RAPIDJSON_SCHEMA_VERBOSE +#define RAPIDJSON_SCHEMA_VERBOSE 0 +#endif + +RAPIDJSON_DIAG_PUSH + +#if defined(__GNUC__) +RAPIDJSON_DIAG_OFF(effc++) +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_OFF(weak-vtables) +RAPIDJSON_DIAG_OFF(exit-time-destructors) +RAPIDJSON_DIAG_OFF(c++98-compat-pedantic) +RAPIDJSON_DIAG_OFF(variadic-macros) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// Verbose Utilities + +#if RAPIDJSON_SCHEMA_VERBOSE + +namespace internal { + +inline void PrintInvalidKeywordData(const char* keyword) { + printf(" Fail keyword: '%s'\n", keyword); +} + +inline void PrintInvalidKeywordData(const wchar_t* keyword) { + wprintf(L" Fail keyword: '%ls'\n", keyword); +} + +inline void PrintInvalidDocumentData(const char* document) { + printf(" Fail document: '%s'\n", document); +} + +inline void PrintInvalidDocumentData(const wchar_t* document) { + wprintf(L" Fail document: '%ls'\n", document); +} + +inline void PrintValidatorPointersData(const char* s, const char* d, unsigned depth) { + printf(" Sch: %*s'%s'\n Doc: %*s'%s'\n", depth * 4, " ", s, depth * 4, " ", d); +} + +inline void PrintValidatorPointersData(const wchar_t* s, const wchar_t* d, unsigned depth) { + wprintf(L" Sch: %*ls'%ls'\n Doc: %*ls'%ls'\n", depth * 4, L" ", s, depth * 4, L" ", d); +} + +inline void PrintSchemaIdsData(const char* base, const char* local, const char* resolved) { + printf(" Resolving id: Base: '%s', Local: '%s', Resolved: '%s'\n", base, local, resolved); +} + +inline void PrintSchemaIdsData(const wchar_t* base, const wchar_t* local, const wchar_t* resolved) { + wprintf(L" Resolving id: Base: '%ls', Local: '%ls', Resolved: '%ls'\n", base, local, resolved); +} + +inline void PrintMethodData(const char* method) { + printf("%s\n", method); +} + +inline void PrintMethodData(const char* method, bool b) { + printf("%s, Data: '%s'\n", method, b ? "true" : "false"); +} + +inline void PrintMethodData(const char* method, int64_t i) { + printf("%s, Data: '%" PRId64 "'\n", method, i); +} + +inline void PrintMethodData(const char* method, uint64_t u) { + printf("%s, Data: '%" PRIu64 "'\n", method, u); +} + +inline void PrintMethodData(const char* method, double d) { + printf("%s, Data: '%lf'\n", method, d); +} + +inline void PrintMethodData(const char* method, const char* s) { + printf("%s, Data: '%s'\n", method, s); +} + +inline void PrintMethodData(const char* method, const wchar_t* s) { + wprintf(L"%hs, Data: '%ls'\n", method, s); +} + +inline void PrintMethodData(const char* method, const char* s1, const char* s2) { + printf("%s, Data: '%s', '%s'\n", method, s1, s2); +} + +inline void PrintMethodData(const char* method, const wchar_t* s1, const wchar_t* s2) { + wprintf(L"%hs, Data: '%ls', '%ls'\n", method, s1, s2); +} + +} // namespace internal + +#endif // RAPIDJSON_SCHEMA_VERBOSE + +#ifndef RAPIDJSON_SCHEMA_PRINT +#if RAPIDJSON_SCHEMA_VERBOSE +#define RAPIDJSON_SCHEMA_PRINT(name, ...) internal::Print##name##Data(__VA_ARGS__) +#else +#define RAPIDJSON_SCHEMA_PRINT(name, ...) +#endif +#endif + +/////////////////////////////////////////////////////////////////////////////// +// RAPIDJSON_INVALID_KEYWORD_RETURN + +#define RAPIDJSON_INVALID_KEYWORD_RETURN(code)\ +RAPIDJSON_MULTILINEMACRO_BEGIN\ + context.invalidCode = code;\ + context.invalidKeyword = SchemaType::GetValidateErrorKeyword(code).GetString();\ + RAPIDJSON_SCHEMA_PRINT(InvalidKeyword, context.invalidKeyword);\ + return false;\ +RAPIDJSON_MULTILINEMACRO_END + +/////////////////////////////////////////////////////////////////////////////// +// ValidateFlag + +/*! \def RAPIDJSON_VALIDATE_DEFAULT_FLAGS + \ingroup RAPIDJSON_CONFIG + \brief User-defined kValidateDefaultFlags definition. + + User can define this as any \c ValidateFlag combinations. +*/ +#ifndef RAPIDJSON_VALIDATE_DEFAULT_FLAGS +#define RAPIDJSON_VALIDATE_DEFAULT_FLAGS kValidateNoFlags +#endif + +//! Combination of validate flags +enum ValidateFlag { + kValidateNoFlags = 0, //!< No flags are set. + kValidateContinueOnErrorFlag = 1, //!< Don't stop after first validation error. + kValidateReadFlag = 2, //!< Validation is for a read semantic. + kValidateWriteFlag = 4, //!< Validation is for a write semantic. + kValidateDefaultFlags = RAPIDJSON_VALIDATE_DEFAULT_FLAGS //!< Default validate flags. Can be customized by defining RAPIDJSON_VALIDATE_DEFAULT_FLAGS +}; + +/////////////////////////////////////////////////////////////////////////////// +// Specification +enum SchemaDraft { + kDraftUnknown = -1, + kDraftNone = 0, + kDraft03 = 3, + kDraftMin = 4, //!< Current minimum supported draft + kDraft04 = 4, + kDraft05 = 5, + kDraftMax = 5, //!< Current maximum supported draft + kDraft06 = 6, + kDraft07 = 7, + kDraft2019_09 = 8, + kDraft2020_12 = 9 +}; + +enum OpenApiVersion { + kVersionUnknown = -1, + kVersionNone = 0, + kVersionMin = 2, //!< Current minimum supported version + kVersion20 = 2, + kVersion30 = 3, + kVersionMax = 3, //!< Current maximum supported version + kVersion31 = 4, +}; + +struct Specification { + Specification(SchemaDraft d) : draft(d), oapi(kVersionNone) {} + Specification(OpenApiVersion o) : oapi(o) { + if (oapi == kVersion20) draft = kDraft04; + else if (oapi == kVersion30) draft = kDraft05; + else if (oapi == kVersion31) draft = kDraft2020_12; + else draft = kDraft04; + } + ~Specification() {} + bool IsSupported() const { + return ((draft >= kDraftMin && draft <= kDraftMax) && ((oapi == kVersionNone) || (oapi >= kVersionMin && oapi <= kVersionMax))); + } + SchemaDraft draft; + OpenApiVersion oapi; +}; + +/////////////////////////////////////////////////////////////////////////////// +// Forward declarations + +template +class GenericSchemaDocument; + +namespace internal { + +template +class Schema; + +/////////////////////////////////////////////////////////////////////////////// +// ISchemaValidator + +class ISchemaValidator { +public: + virtual ~ISchemaValidator() {} + virtual bool IsValid() const = 0; + virtual void SetValidateFlags(unsigned flags) = 0; + virtual unsigned GetValidateFlags() const = 0; +}; + +/////////////////////////////////////////////////////////////////////////////// +// ISchemaStateFactory + +template +class ISchemaStateFactory { +public: + virtual ~ISchemaStateFactory() {} + virtual ISchemaValidator* CreateSchemaValidator(const SchemaType&, const bool inheritContinueOnErrors) = 0; + virtual void DestroySchemaValidator(ISchemaValidator* validator) = 0; + virtual void* CreateHasher() = 0; + virtual uint64_t GetHashCode(void* hasher) = 0; + virtual void DestroryHasher(void* hasher) = 0; + virtual void* MallocState(size_t size) = 0; + virtual void FreeState(void* p) = 0; +}; + +/////////////////////////////////////////////////////////////////////////////// +// IValidationErrorHandler + +template +class IValidationErrorHandler { +public: + typedef typename SchemaType::Ch Ch; + typedef typename SchemaType::SValue SValue; + + virtual ~IValidationErrorHandler() {} + + virtual void NotMultipleOf(int64_t actual, const SValue& expected) = 0; + virtual void NotMultipleOf(uint64_t actual, const SValue& expected) = 0; + virtual void NotMultipleOf(double actual, const SValue& expected) = 0; + virtual void AboveMaximum(int64_t actual, const SValue& expected, bool exclusive) = 0; + virtual void AboveMaximum(uint64_t actual, const SValue& expected, bool exclusive) = 0; + virtual void AboveMaximum(double actual, const SValue& expected, bool exclusive) = 0; + virtual void BelowMinimum(int64_t actual, const SValue& expected, bool exclusive) = 0; + virtual void BelowMinimum(uint64_t actual, const SValue& expected, bool exclusive) = 0; + virtual void BelowMinimum(double actual, const SValue& expected, bool exclusive) = 0; + + virtual void TooLong(const Ch* str, SizeType length, SizeType expected) = 0; + virtual void TooShort(const Ch* str, SizeType length, SizeType expected) = 0; + virtual void DoesNotMatch(const Ch* str, SizeType length) = 0; + + virtual void DisallowedItem(SizeType index) = 0; + virtual void TooFewItems(SizeType actualCount, SizeType expectedCount) = 0; + virtual void TooManyItems(SizeType actualCount, SizeType expectedCount) = 0; + virtual void DuplicateItems(SizeType index1, SizeType index2) = 0; + + virtual void TooManyProperties(SizeType actualCount, SizeType expectedCount) = 0; + virtual void TooFewProperties(SizeType actualCount, SizeType expectedCount) = 0; + virtual void StartMissingProperties() = 0; + virtual void AddMissingProperty(const SValue& name) = 0; + virtual bool EndMissingProperties() = 0; + virtual void PropertyViolations(ISchemaValidator** subvalidators, SizeType count) = 0; + virtual void DisallowedProperty(const Ch* name, SizeType length) = 0; + + virtual void StartDependencyErrors() = 0; + virtual void StartMissingDependentProperties() = 0; + virtual void AddMissingDependentProperty(const SValue& targetName) = 0; + virtual void EndMissingDependentProperties(const SValue& sourceName) = 0; + virtual void AddDependencySchemaError(const SValue& souceName, ISchemaValidator* subvalidator) = 0; + virtual bool EndDependencyErrors() = 0; + + virtual void DisallowedValue(const ValidateErrorCode code) = 0; + virtual void StartDisallowedType() = 0; + virtual void AddExpectedType(const typename SchemaType::ValueType& expectedType) = 0; + virtual void EndDisallowedType(const typename SchemaType::ValueType& actualType) = 0; + virtual void NotAllOf(ISchemaValidator** subvalidators, SizeType count) = 0; + virtual void NoneOf(ISchemaValidator** subvalidators, SizeType count) = 0; + virtual void NotOneOf(ISchemaValidator** subvalidators, SizeType count) = 0; + virtual void MultipleOneOf(SizeType index1, SizeType index2) = 0; + virtual void Disallowed() = 0; + virtual void DisallowedWhenWriting() = 0; + virtual void DisallowedWhenReading() = 0; +}; + + +/////////////////////////////////////////////////////////////////////////////// +// Hasher + +// For comparison of compound value +template +class Hasher { +public: + typedef typename Encoding::Ch Ch; + + Hasher(Allocator* allocator = 0, size_t stackCapacity = kDefaultSize) : stack_(allocator, stackCapacity) {} + + bool Null() { return WriteType(kNullType); } + bool Bool(bool b) { return WriteType(b ? kTrueType : kFalseType); } + bool Int(int i) { Number n; n.u.i = i; n.d = static_cast(i); return WriteNumber(n); } + bool Uint(unsigned u) { Number n; n.u.u = u; n.d = static_cast(u); return WriteNumber(n); } + bool Int64(int64_t i) { Number n; n.u.i = i; n.d = static_cast(i); return WriteNumber(n); } + bool Uint64(uint64_t u) { Number n; n.u.u = u; n.d = static_cast(u); return WriteNumber(n); } + bool Double(double d) { + Number n; + if (d < 0) n.u.i = static_cast(d); + else n.u.u = static_cast(d); + n.d = d; + return WriteNumber(n); + } + + bool RawNumber(const Ch* str, SizeType len, bool) { + WriteBuffer(kNumberType, str, len * sizeof(Ch)); + return true; + } + + bool String(const Ch* str, SizeType len, bool) { + WriteBuffer(kStringType, str, len * sizeof(Ch)); + return true; + } + + bool StartObject() { return true; } + bool Key(const Ch* str, SizeType len, bool copy) { return String(str, len, copy); } + bool EndObject(SizeType memberCount) { + uint64_t h = Hash(0, kObjectType); + uint64_t* kv = stack_.template Pop(memberCount * 2); + for (SizeType i = 0; i < memberCount; i++) + // Issue #2205 + // Hasing the key to avoid key=value cases with bug-prone zero-value hash + h ^= Hash(Hash(0, kv[i * 2]), kv[i * 2 + 1]); // Use xor to achieve member order insensitive + *stack_.template Push() = h; + return true; + } + + bool StartArray() { return true; } + bool EndArray(SizeType elementCount) { + uint64_t h = Hash(0, kArrayType); + uint64_t* e = stack_.template Pop(elementCount); + for (SizeType i = 0; i < elementCount; i++) + h = Hash(h, e[i]); // Use hash to achieve element order sensitive + *stack_.template Push() = h; + return true; + } + + bool IsValid() const { return stack_.GetSize() == sizeof(uint64_t); } + + uint64_t GetHashCode() const { + RAPIDJSON_ASSERT(IsValid()); + return *stack_.template Top(); + } + +private: + static const size_t kDefaultSize = 256; + struct Number { + union U { + uint64_t u; + int64_t i; + }u; + double d; + }; + + bool WriteType(Type type) { return WriteBuffer(type, 0, 0); } + + bool WriteNumber(const Number& n) { return WriteBuffer(kNumberType, &n, sizeof(n)); } + + bool WriteBuffer(Type type, const void* data, size_t len) { + // FNV-1a from http://isthe.com/chongo/tech/comp/fnv/ + uint64_t h = Hash(RAPIDJSON_UINT64_C2(0xcbf29ce4, 0x84222325), type); + const unsigned char* d = static_cast(data); + for (size_t i = 0; i < len; i++) + h = Hash(h, d[i]); + *stack_.template Push() = h; + return true; + } + + static uint64_t Hash(uint64_t h, uint64_t d) { + static const uint64_t kPrime = RAPIDJSON_UINT64_C2(0x00000100, 0x000001b3); + h ^= d; + h *= kPrime; + return h; + } + + Stack stack_; +}; + +/////////////////////////////////////////////////////////////////////////////// +// SchemaValidationContext + +template +struct SchemaValidationContext { + typedef Schema SchemaType; + typedef ISchemaStateFactory SchemaValidatorFactoryType; + typedef IValidationErrorHandler ErrorHandlerType; + typedef typename SchemaType::ValueType ValueType; + typedef typename ValueType::Ch Ch; + + enum PatternValidatorType { + kPatternValidatorOnly, + kPatternValidatorWithProperty, + kPatternValidatorWithAdditionalProperty + }; + + SchemaValidationContext(SchemaValidatorFactoryType& f, ErrorHandlerType& eh, const SchemaType* s, unsigned fl = 0) : + factory(f), + error_handler(eh), + schema(s), + flags(fl), + valueSchema(), + invalidKeyword(), + invalidCode(), + hasher(), + arrayElementHashCodes(), + validators(), + validatorCount(), + patternPropertiesValidators(), + patternPropertiesValidatorCount(), + patternPropertiesSchemas(), + patternPropertiesSchemaCount(), + valuePatternValidatorType(kPatternValidatorOnly), + propertyExist(), + inArray(false), + valueUniqueness(false), + arrayUniqueness(false) + { + } + + ~SchemaValidationContext() { + if (hasher) + factory.DestroryHasher(hasher); + if (validators) { + for (SizeType i = 0; i < validatorCount; i++) { + if (validators[i]) { + factory.DestroySchemaValidator(validators[i]); + } + } + factory.FreeState(validators); + } + if (patternPropertiesValidators) { + for (SizeType i = 0; i < patternPropertiesValidatorCount; i++) { + if (patternPropertiesValidators[i]) { + factory.DestroySchemaValidator(patternPropertiesValidators[i]); + } + } + factory.FreeState(patternPropertiesValidators); + } + if (patternPropertiesSchemas) + factory.FreeState(patternPropertiesSchemas); + if (propertyExist) + factory.FreeState(propertyExist); + } + + SchemaValidatorFactoryType& factory; + ErrorHandlerType& error_handler; + const SchemaType* schema; + unsigned flags; + const SchemaType* valueSchema; + const Ch* invalidKeyword; + ValidateErrorCode invalidCode; + void* hasher; // Only validator access + void* arrayElementHashCodes; // Only validator access this + ISchemaValidator** validators; + SizeType validatorCount; + ISchemaValidator** patternPropertiesValidators; + SizeType patternPropertiesValidatorCount; + const SchemaType** patternPropertiesSchemas; + SizeType patternPropertiesSchemaCount; + PatternValidatorType valuePatternValidatorType; + PatternValidatorType objectPatternValidatorType; + SizeType arrayElementIndex; + bool* propertyExist; + bool inArray; + bool valueUniqueness; + bool arrayUniqueness; +}; + +/////////////////////////////////////////////////////////////////////////////// +// Schema + +template +class Schema { +public: + typedef typename SchemaDocumentType::ValueType ValueType; + typedef typename SchemaDocumentType::AllocatorType AllocatorType; + typedef typename SchemaDocumentType::PointerType PointerType; + typedef typename ValueType::EncodingType EncodingType; + typedef typename EncodingType::Ch Ch; + typedef SchemaValidationContext Context; + typedef Schema SchemaType; + typedef GenericValue SValue; + typedef IValidationErrorHandler ErrorHandler; + typedef GenericUri UriType; + friend class GenericSchemaDocument; + + Schema(SchemaDocumentType* schemaDocument, const PointerType& p, const ValueType& value, const ValueType& document, AllocatorType* allocator, const UriType& id = UriType()) : + allocator_(allocator), + uri_(schemaDocument->GetURI(), *allocator), + id_(id, allocator), + spec_(schemaDocument->GetSpecification()), + pointer_(p, allocator), + typeless_(schemaDocument->GetTypeless()), + enum_(), + enumCount_(), + not_(), + type_((1 << kTotalSchemaType) - 1), // typeless + validatorCount_(), + notValidatorIndex_(), + properties_(), + additionalPropertiesSchema_(), + patternProperties_(), + patternPropertyCount_(), + propertyCount_(), + minProperties_(), + maxProperties_(SizeType(~0)), + additionalProperties_(true), + hasDependencies_(), + hasRequired_(), + hasSchemaDependencies_(), + additionalItemsSchema_(), + itemsList_(), + itemsTuple_(), + itemsTupleCount_(), + minItems_(), + maxItems_(SizeType(~0)), + additionalItems_(true), + uniqueItems_(false), + pattern_(), + minLength_(0), + maxLength_(~SizeType(0)), + exclusiveMinimum_(false), + exclusiveMaximum_(false), + defaultValueLength_(0), + readOnly_(false), + writeOnly_(false), + nullable_(false) + { + GenericStringBuffer sb; + p.StringifyUriFragment(sb); + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Schema", sb.GetString(), id.GetString()); + + typedef typename ValueType::ConstValueIterator ConstValueIterator; + typedef typename ValueType::ConstMemberIterator ConstMemberIterator; + + // PR #1393 + // Early add this Schema and its $ref(s) in schemaDocument's map to avoid infinite + // recursion (with recursive schemas), since schemaDocument->getSchema() is always + // checked before creating a new one. Don't cache typeless_, though. + if (this != typeless_) { + typedef typename SchemaDocumentType::SchemaEntry SchemaEntry; + SchemaEntry *entry = schemaDocument->schemaMap_.template Push(); + new (entry) SchemaEntry(pointer_, this, true, allocator_); + schemaDocument->AddSchemaRefs(this); + } + + if (!value.IsObject()) + return; + + // If we have an id property, resolve it with the in-scope id + // Not supported for open api 2.0 or 3.0 + if (spec_.oapi != kVersion20 && spec_.oapi != kVersion30) + if (const ValueType* v = GetMember(value, GetIdString())) { + if (v->IsString()) { + UriType local(*v, allocator); + id_ = local.Resolve(id_, allocator); + RAPIDJSON_SCHEMA_PRINT(SchemaIds, id.GetString(), v->GetString(), id_.GetString()); + } + } + + if (const ValueType* v = GetMember(value, GetTypeString())) { + type_ = 0; + if (v->IsString()) + AddType(*v); + else if (v->IsArray()) + for (ConstValueIterator itr = v->Begin(); itr != v->End(); ++itr) + AddType(*itr); + } + + if (const ValueType* v = GetMember(value, GetEnumString())) { + if (v->IsArray() && v->Size() > 0) { + enum_ = static_cast(allocator_->Malloc(sizeof(uint64_t) * v->Size())); + for (ConstValueIterator itr = v->Begin(); itr != v->End(); ++itr) { + typedef Hasher > EnumHasherType; + char buffer[256u + 24]; + MemoryPoolAllocator hasherAllocator(buffer, sizeof(buffer)); + EnumHasherType h(&hasherAllocator, 256); + itr->Accept(h); + enum_[enumCount_++] = h.GetHashCode(); + } + } + } + + if (schemaDocument) + AssignIfExist(allOf_, *schemaDocument, p, value, GetAllOfString(), document); + + // AnyOf, OneOf, Not not supported for open api 2.0 + if (schemaDocument && spec_.oapi != kVersion20) { + AssignIfExist(anyOf_, *schemaDocument, p, value, GetAnyOfString(), document); + AssignIfExist(oneOf_, *schemaDocument, p, value, GetOneOfString(), document); + + if (const ValueType* v = GetMember(value, GetNotString())) { + schemaDocument->CreateSchema(¬_, p.Append(GetNotString(), allocator_), *v, document, id_); + notValidatorIndex_ = validatorCount_; + validatorCount_++; + } + } + + // Object + + const ValueType* properties = GetMember(value, GetPropertiesString()); + const ValueType* required = GetMember(value, GetRequiredString()); + const ValueType* dependencies = GetMember(value, GetDependenciesString()); + { + // Gather properties from properties/required/dependencies + SValue allProperties(kArrayType); + + if (properties && properties->IsObject()) + for (ConstMemberIterator itr = properties->MemberBegin(); itr != properties->MemberEnd(); ++itr) + AddUniqueElement(allProperties, itr->name); + + if (required && required->IsArray()) + for (ConstValueIterator itr = required->Begin(); itr != required->End(); ++itr) + if (itr->IsString()) + AddUniqueElement(allProperties, *itr); + + // Dependencies not supported for open api 2.0 and 3.0 + if (spec_.oapi != kVersion20 && spec_.oapi != kVersion30) + if (dependencies && dependencies->IsObject()) + for (ConstMemberIterator itr = dependencies->MemberBegin(); itr != dependencies->MemberEnd(); ++itr) { + AddUniqueElement(allProperties, itr->name); + if (itr->value.IsArray()) + for (ConstValueIterator i = itr->value.Begin(); i != itr->value.End(); ++i) + if (i->IsString()) + AddUniqueElement(allProperties, *i); + } + + if (allProperties.Size() > 0) { + propertyCount_ = allProperties.Size(); + properties_ = static_cast(allocator_->Malloc(sizeof(Property) * propertyCount_)); + for (SizeType i = 0; i < propertyCount_; i++) { + new (&properties_[i]) Property(); + properties_[i].name = allProperties[i]; + properties_[i].schema = typeless_; + } + } + } + + if (properties && properties->IsObject()) { + PointerType q = p.Append(GetPropertiesString(), allocator_); + for (ConstMemberIterator itr = properties->MemberBegin(); itr != properties->MemberEnd(); ++itr) { + SizeType index; + if (FindPropertyIndex(itr->name, &index)) + schemaDocument->CreateSchema(&properties_[index].schema, q.Append(itr->name, allocator_), itr->value, document, id_); + } + } + + // PatternProperties not supported for open api 2.0 and 3.0 + if (spec_.oapi != kVersion20 && spec_.oapi != kVersion30) + if (const ValueType* v = GetMember(value, GetPatternPropertiesString())) { + PointerType q = p.Append(GetPatternPropertiesString(), allocator_); + patternProperties_ = static_cast(allocator_->Malloc(sizeof(PatternProperty) * v->MemberCount())); + patternPropertyCount_ = 0; + + for (ConstMemberIterator itr = v->MemberBegin(); itr != v->MemberEnd(); ++itr) { + new (&patternProperties_[patternPropertyCount_]) PatternProperty(); + PointerType r = q.Append(itr->name, allocator_); + patternProperties_[patternPropertyCount_].pattern = CreatePattern(itr->name, schemaDocument, r); + schemaDocument->CreateSchema(&patternProperties_[patternPropertyCount_].schema, r, itr->value, document, id_); + patternPropertyCount_++; + } + } + + if (required && required->IsArray()) + for (ConstValueIterator itr = required->Begin(); itr != required->End(); ++itr) + if (itr->IsString()) { + SizeType index; + if (FindPropertyIndex(*itr, &index)) { + properties_[index].required = true; + hasRequired_ = true; + } + } + + // Dependencies not supported for open api 2.0 and 3.0 + if (spec_.oapi != kVersion20 && spec_.oapi != kVersion30) + if (dependencies && dependencies->IsObject()) { + PointerType q = p.Append(GetDependenciesString(), allocator_); + hasDependencies_ = true; + for (ConstMemberIterator itr = dependencies->MemberBegin(); itr != dependencies->MemberEnd(); ++itr) { + SizeType sourceIndex; + if (FindPropertyIndex(itr->name, &sourceIndex)) { + if (itr->value.IsArray()) { + properties_[sourceIndex].dependencies = static_cast(allocator_->Malloc(sizeof(bool) * propertyCount_)); + std::memset(properties_[sourceIndex].dependencies, 0, sizeof(bool)* propertyCount_); + for (ConstValueIterator targetItr = itr->value.Begin(); targetItr != itr->value.End(); ++targetItr) { + SizeType targetIndex; + if (FindPropertyIndex(*targetItr, &targetIndex)) + properties_[sourceIndex].dependencies[targetIndex] = true; + } + } + else if (itr->value.IsObject()) { + hasSchemaDependencies_ = true; + schemaDocument->CreateSchema(&properties_[sourceIndex].dependenciesSchema, q.Append(itr->name, allocator_), itr->value, document, id_); + properties_[sourceIndex].dependenciesValidatorIndex = validatorCount_; + validatorCount_++; + } + } + } + } + + if (const ValueType* v = GetMember(value, GetAdditionalPropertiesString())) { + if (v->IsBool()) + additionalProperties_ = v->GetBool(); + else if (v->IsObject()) + schemaDocument->CreateSchema(&additionalPropertiesSchema_, p.Append(GetAdditionalPropertiesString(), allocator_), *v, document, id_); + } + + AssignIfExist(minProperties_, value, GetMinPropertiesString()); + AssignIfExist(maxProperties_, value, GetMaxPropertiesString()); + + // Array + if (const ValueType* v = GetMember(value, GetItemsString())) { + PointerType q = p.Append(GetItemsString(), allocator_); + if (v->IsObject()) // List validation + schemaDocument->CreateSchema(&itemsList_, q, *v, document, id_); + else if (v->IsArray()) { // Tuple validation + itemsTuple_ = static_cast(allocator_->Malloc(sizeof(const Schema*) * v->Size())); + SizeType index = 0; + for (ConstValueIterator itr = v->Begin(); itr != v->End(); ++itr, index++) + schemaDocument->CreateSchema(&itemsTuple_[itemsTupleCount_++], q.Append(index, allocator_), *itr, document, id_); + } + } + + AssignIfExist(minItems_, value, GetMinItemsString()); + AssignIfExist(maxItems_, value, GetMaxItemsString()); + + // AdditionalItems not supported for openapi 2.0 and 3.0 + if (spec_.oapi != kVersion20 && spec_.oapi != kVersion30) + if (const ValueType* v = GetMember(value, GetAdditionalItemsString())) { + if (v->IsBool()) + additionalItems_ = v->GetBool(); + else if (v->IsObject()) + schemaDocument->CreateSchema(&additionalItemsSchema_, p.Append(GetAdditionalItemsString(), allocator_), *v, document, id_); + } + + AssignIfExist(uniqueItems_, value, GetUniqueItemsString()); + + // String + AssignIfExist(minLength_, value, GetMinLengthString()); + AssignIfExist(maxLength_, value, GetMaxLengthString()); + + if (const ValueType* v = GetMember(value, GetPatternString())) + pattern_ = CreatePattern(*v, schemaDocument, p.Append(GetPatternString(), allocator_)); + + // Number + if (const ValueType* v = GetMember(value, GetMinimumString())) + if (v->IsNumber()) + minimum_.CopyFrom(*v, *allocator_); + + if (const ValueType* v = GetMember(value, GetMaximumString())) + if (v->IsNumber()) + maximum_.CopyFrom(*v, *allocator_); + + AssignIfExist(exclusiveMinimum_, value, GetExclusiveMinimumString()); + AssignIfExist(exclusiveMaximum_, value, GetExclusiveMaximumString()); + + if (const ValueType* v = GetMember(value, GetMultipleOfString())) + if (v->IsNumber() && v->GetDouble() > 0.0) + multipleOf_.CopyFrom(*v, *allocator_); + + // Default + if (const ValueType* v = GetMember(value, GetDefaultValueString())) + if (v->IsString()) + defaultValueLength_ = v->GetStringLength(); + + // ReadOnly - open api only (until draft 7 supported) + // WriteOnly - open api 3 only (until draft 7 supported) + // Both can't be true + if (spec_.oapi != kVersionNone) + AssignIfExist(readOnly_, value, GetReadOnlyString()); + if (spec_.oapi >= kVersion30) + AssignIfExist(writeOnly_, value, GetWriteOnlyString()); + if (readOnly_ && writeOnly_) + schemaDocument->SchemaError(kSchemaErrorReadOnlyAndWriteOnly, p); + + // Nullable - open api 3 only + // If true add 'null' as allowable type + if (spec_.oapi >= kVersion30) { + AssignIfExist(nullable_, value, GetNullableString()); + if (nullable_) + AddType(GetNullString()); + } + } + + ~Schema() { + AllocatorType::Free(enum_); + if (properties_) { + for (SizeType i = 0; i < propertyCount_; i++) + properties_[i].~Property(); + AllocatorType::Free(properties_); + } + if (patternProperties_) { + for (SizeType i = 0; i < patternPropertyCount_; i++) + patternProperties_[i].~PatternProperty(); + AllocatorType::Free(patternProperties_); + } + AllocatorType::Free(itemsTuple_); +#if RAPIDJSON_SCHEMA_HAS_REGEX + if (pattern_) { + pattern_->~RegexType(); + AllocatorType::Free(pattern_); + } +#endif + } + + const SValue& GetURI() const { + return uri_; + } + + const UriType& GetId() const { + return id_; + } + + const Specification& GetSpecification() const { + return spec_; + } + + const PointerType& GetPointer() const { + return pointer_; + } + + bool BeginValue(Context& context) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::BeginValue"); + if (context.inArray) { + if (uniqueItems_) + context.valueUniqueness = true; + + if (itemsList_) + context.valueSchema = itemsList_; + else if (itemsTuple_) { + if (context.arrayElementIndex < itemsTupleCount_) + context.valueSchema = itemsTuple_[context.arrayElementIndex]; + else if (additionalItemsSchema_) + context.valueSchema = additionalItemsSchema_; + else if (additionalItems_) + context.valueSchema = typeless_; + else { + context.error_handler.DisallowedItem(context.arrayElementIndex); + // Must set valueSchema for when kValidateContinueOnErrorFlag is set, else reports spurious type error + context.valueSchema = typeless_; + // Must bump arrayElementIndex for when kValidateContinueOnErrorFlag is set + context.arrayElementIndex++; + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorAdditionalItems); + } + } + else + context.valueSchema = typeless_; + + context.arrayElementIndex++; + } + return true; + } + + RAPIDJSON_FORCEINLINE bool EndValue(Context& context) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::EndValue"); + // Only check pattern properties if we have validators + if (context.patternPropertiesValidatorCount > 0) { + bool otherValid = false; + SizeType count = context.patternPropertiesValidatorCount; + if (context.objectPatternValidatorType != Context::kPatternValidatorOnly) + otherValid = context.patternPropertiesValidators[--count]->IsValid(); + + bool patternValid = true; + for (SizeType i = 0; i < count; i++) + if (!context.patternPropertiesValidators[i]->IsValid()) { + patternValid = false; + break; + } + + if (context.objectPatternValidatorType == Context::kPatternValidatorOnly) { + if (!patternValid) { + context.error_handler.PropertyViolations(context.patternPropertiesValidators, count); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorPatternProperties); + } + } + else if (context.objectPatternValidatorType == Context::kPatternValidatorWithProperty) { + if (!patternValid || !otherValid) { + context.error_handler.PropertyViolations(context.patternPropertiesValidators, count + 1); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorPatternProperties); + } + } + else if (!patternValid && !otherValid) { // kPatternValidatorWithAdditionalProperty) + context.error_handler.PropertyViolations(context.patternPropertiesValidators, count + 1); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorPatternProperties); + } + } + + // For enums only check if we have a hasher + if (enum_ && context.hasher) { + const uint64_t h = context.factory.GetHashCode(context.hasher); + for (SizeType i = 0; i < enumCount_; i++) + if (enum_[i] == h) + goto foundEnum; + context.error_handler.DisallowedValue(kValidateErrorEnum); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorEnum); + foundEnum:; + } + + // Only check allOf etc if we have validators + if (context.validatorCount > 0) { + if (allOf_.schemas) + for (SizeType i = allOf_.begin; i < allOf_.begin + allOf_.count; i++) + if (!context.validators[i]->IsValid()) { + context.error_handler.NotAllOf(&context.validators[allOf_.begin], allOf_.count); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorAllOf); + } + + if (anyOf_.schemas) { + for (SizeType i = anyOf_.begin; i < anyOf_.begin + anyOf_.count; i++) + if (context.validators[i]->IsValid()) + goto foundAny; + context.error_handler.NoneOf(&context.validators[anyOf_.begin], anyOf_.count); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorAnyOf); + foundAny:; + } + + if (oneOf_.schemas) { + bool oneValid = false; + SizeType firstMatch = 0; + for (SizeType i = oneOf_.begin; i < oneOf_.begin + oneOf_.count; i++) + if (context.validators[i]->IsValid()) { + if (oneValid) { + context.error_handler.MultipleOneOf(firstMatch, i - oneOf_.begin); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorOneOfMatch); + } else { + oneValid = true; + firstMatch = i - oneOf_.begin; + } + } + if (!oneValid) { + context.error_handler.NotOneOf(&context.validators[oneOf_.begin], oneOf_.count); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorOneOf); + } + } + + if (not_ && context.validators[notValidatorIndex_]->IsValid()) { + context.error_handler.Disallowed(); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorNot); + } + } + + return true; + } + + bool Null(Context& context) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Null"); + if (!(type_ & (1 << kNullSchemaType))) { + DisallowedType(context, GetNullString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + return CreateParallelValidator(context); + } + + bool Bool(Context& context, bool b) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Bool", b); + if (!CheckBool(context, b)) + return false; + return CreateParallelValidator(context); + } + + bool Int(Context& context, int i) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Int", (int64_t)i); + if (!CheckInt(context, i)) + return false; + return CreateParallelValidator(context); + } + + bool Uint(Context& context, unsigned u) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Uint", (uint64_t)u); + if (!CheckUint(context, u)) + return false; + return CreateParallelValidator(context); + } + + bool Int64(Context& context, int64_t i) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Int64", i); + if (!CheckInt(context, i)) + return false; + return CreateParallelValidator(context); + } + + bool Uint64(Context& context, uint64_t u) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Uint64", u); + if (!CheckUint(context, u)) + return false; + return CreateParallelValidator(context); + } + + bool Double(Context& context, double d) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Double", d); + if (!(type_ & (1 << kNumberSchemaType))) { + DisallowedType(context, GetNumberString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + if (!minimum_.IsNull() && !CheckDoubleMinimum(context, d)) + return false; + + if (!maximum_.IsNull() && !CheckDoubleMaximum(context, d)) + return false; + + if (!multipleOf_.IsNull() && !CheckDoubleMultipleOf(context, d)) + return false; + + return CreateParallelValidator(context); + } + + bool String(Context& context, const Ch* str, SizeType length, bool) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::String", str); + if (!(type_ & (1 << kStringSchemaType))) { + DisallowedType(context, GetStringString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + if (minLength_ != 0 || maxLength_ != SizeType(~0)) { + SizeType count; + if (internal::CountStringCodePoint(str, length, &count)) { + if (count < minLength_) { + context.error_handler.TooShort(str, length, minLength_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMinLength); + } + if (count > maxLength_) { + context.error_handler.TooLong(str, length, maxLength_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMaxLength); + } + } + } + + if (pattern_ && !IsPatternMatch(pattern_, str, length)) { + context.error_handler.DoesNotMatch(str, length); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorPattern); + } + + return CreateParallelValidator(context); + } + + bool StartObject(Context& context) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::StartObject"); + if (!(type_ & (1 << kObjectSchemaType))) { + DisallowedType(context, GetObjectString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + if (hasDependencies_ || hasRequired_) { + context.propertyExist = static_cast(context.factory.MallocState(sizeof(bool) * propertyCount_)); + std::memset(context.propertyExist, 0, sizeof(bool) * propertyCount_); + } + + if (patternProperties_) { // pre-allocate schema array + SizeType count = patternPropertyCount_ + 1; // extra for valuePatternValidatorType + context.patternPropertiesSchemas = static_cast(context.factory.MallocState(sizeof(const SchemaType*) * count)); + context.patternPropertiesSchemaCount = 0; + std::memset(context.patternPropertiesSchemas, 0, sizeof(SchemaType*) * count); + } + + return CreateParallelValidator(context); + } + + bool Key(Context& context, const Ch* str, SizeType len, bool) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::Key", str); + + if (patternProperties_) { + context.patternPropertiesSchemaCount = 0; + for (SizeType i = 0; i < patternPropertyCount_; i++) + if (patternProperties_[i].pattern && IsPatternMatch(patternProperties_[i].pattern, str, len)) { + context.patternPropertiesSchemas[context.patternPropertiesSchemaCount++] = patternProperties_[i].schema; + context.valueSchema = typeless_; + } + } + + SizeType index = 0; + if (FindPropertyIndex(ValueType(str, len).Move(), &index)) { + if (context.patternPropertiesSchemaCount > 0) { + context.patternPropertiesSchemas[context.patternPropertiesSchemaCount++] = properties_[index].schema; + context.valueSchema = typeless_; + context.valuePatternValidatorType = Context::kPatternValidatorWithProperty; + } + else + context.valueSchema = properties_[index].schema; + + if (context.propertyExist) + context.propertyExist[index] = true; + + return true; + } + + if (additionalPropertiesSchema_) { + if (context.patternPropertiesSchemaCount > 0) { + context.patternPropertiesSchemas[context.patternPropertiesSchemaCount++] = additionalPropertiesSchema_; + context.valueSchema = typeless_; + context.valuePatternValidatorType = Context::kPatternValidatorWithAdditionalProperty; + } + else + context.valueSchema = additionalPropertiesSchema_; + return true; + } + else if (additionalProperties_) { + context.valueSchema = typeless_; + return true; + } + + if (context.patternPropertiesSchemaCount == 0) { // patternProperties are not additional properties + // Must set valueSchema for when kValidateContinueOnErrorFlag is set, else reports spurious type error + context.valueSchema = typeless_; + context.error_handler.DisallowedProperty(str, len); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorAdditionalProperties); + } + + return true; + } + + bool EndObject(Context& context, SizeType memberCount) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::EndObject"); + if (hasRequired_) { + context.error_handler.StartMissingProperties(); + for (SizeType index = 0; index < propertyCount_; index++) + if (properties_[index].required && !context.propertyExist[index]) + if (properties_[index].schema->defaultValueLength_ == 0 ) + context.error_handler.AddMissingProperty(properties_[index].name); + if (context.error_handler.EndMissingProperties()) + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorRequired); + } + + if (memberCount < minProperties_) { + context.error_handler.TooFewProperties(memberCount, minProperties_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMinProperties); + } + + if (memberCount > maxProperties_) { + context.error_handler.TooManyProperties(memberCount, maxProperties_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMaxProperties); + } + + if (hasDependencies_) { + context.error_handler.StartDependencyErrors(); + for (SizeType sourceIndex = 0; sourceIndex < propertyCount_; sourceIndex++) { + const Property& source = properties_[sourceIndex]; + if (context.propertyExist[sourceIndex]) { + if (source.dependencies) { + context.error_handler.StartMissingDependentProperties(); + for (SizeType targetIndex = 0; targetIndex < propertyCount_; targetIndex++) + if (source.dependencies[targetIndex] && !context.propertyExist[targetIndex]) + context.error_handler.AddMissingDependentProperty(properties_[targetIndex].name); + context.error_handler.EndMissingDependentProperties(source.name); + } + else if (source.dependenciesSchema) { + ISchemaValidator* dependenciesValidator = context.validators[source.dependenciesValidatorIndex]; + if (!dependenciesValidator->IsValid()) + context.error_handler.AddDependencySchemaError(source.name, dependenciesValidator); + } + } + } + if (context.error_handler.EndDependencyErrors()) + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorDependencies); + } + + return true; + } + + bool StartArray(Context& context) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::StartArray"); + context.arrayElementIndex = 0; + context.inArray = true; // Ensure we note that we are in an array + + if (!(type_ & (1 << kArraySchemaType))) { + DisallowedType(context, GetArrayString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + return CreateParallelValidator(context); + } + + bool EndArray(Context& context, SizeType elementCount) const { + RAPIDJSON_SCHEMA_PRINT(Method, "Schema::EndArray"); + context.inArray = false; + + if (elementCount < minItems_) { + context.error_handler.TooFewItems(elementCount, minItems_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMinItems); + } + + if (elementCount > maxItems_) { + context.error_handler.TooManyItems(elementCount, maxItems_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMaxItems); + } + + return true; + } + + static const ValueType& GetValidateErrorKeyword(ValidateErrorCode validateErrorCode) { + switch (validateErrorCode) { + case kValidateErrorMultipleOf: return GetMultipleOfString(); + case kValidateErrorMaximum: return GetMaximumString(); + case kValidateErrorExclusiveMaximum: return GetMaximumString(); // Same + case kValidateErrorMinimum: return GetMinimumString(); + case kValidateErrorExclusiveMinimum: return GetMinimumString(); // Same + + case kValidateErrorMaxLength: return GetMaxLengthString(); + case kValidateErrorMinLength: return GetMinLengthString(); + case kValidateErrorPattern: return GetPatternString(); + + case kValidateErrorMaxItems: return GetMaxItemsString(); + case kValidateErrorMinItems: return GetMinItemsString(); + case kValidateErrorUniqueItems: return GetUniqueItemsString(); + case kValidateErrorAdditionalItems: return GetAdditionalItemsString(); + + case kValidateErrorMaxProperties: return GetMaxPropertiesString(); + case kValidateErrorMinProperties: return GetMinPropertiesString(); + case kValidateErrorRequired: return GetRequiredString(); + case kValidateErrorAdditionalProperties: return GetAdditionalPropertiesString(); + case kValidateErrorPatternProperties: return GetPatternPropertiesString(); + case kValidateErrorDependencies: return GetDependenciesString(); + + case kValidateErrorEnum: return GetEnumString(); + case kValidateErrorType: return GetTypeString(); + + case kValidateErrorOneOf: return GetOneOfString(); + case kValidateErrorOneOfMatch: return GetOneOfString(); // Same + case kValidateErrorAllOf: return GetAllOfString(); + case kValidateErrorAnyOf: return GetAnyOfString(); + case kValidateErrorNot: return GetNotString(); + + case kValidateErrorReadOnly: return GetReadOnlyString(); + case kValidateErrorWriteOnly: return GetWriteOnlyString(); + + default: return GetNullString(); + } + } + + + // Generate functions for string literal according to Ch +#define RAPIDJSON_STRING_(name, ...) \ + static const ValueType& Get##name##String() {\ + static const Ch s[] = { __VA_ARGS__, '\0' };\ + static const ValueType v(s, static_cast(sizeof(s) / sizeof(Ch) - 1));\ + return v;\ + } + + RAPIDJSON_STRING_(Null, 'n', 'u', 'l', 'l') + RAPIDJSON_STRING_(Boolean, 'b', 'o', 'o', 'l', 'e', 'a', 'n') + RAPIDJSON_STRING_(Object, 'o', 'b', 'j', 'e', 'c', 't') + RAPIDJSON_STRING_(Array, 'a', 'r', 'r', 'a', 'y') + RAPIDJSON_STRING_(String, 's', 't', 'r', 'i', 'n', 'g') + RAPIDJSON_STRING_(Number, 'n', 'u', 'm', 'b', 'e', 'r') + RAPIDJSON_STRING_(Integer, 'i', 'n', 't', 'e', 'g', 'e', 'r') + RAPIDJSON_STRING_(Type, 't', 'y', 'p', 'e') + RAPIDJSON_STRING_(Enum, 'e', 'n', 'u', 'm') + RAPIDJSON_STRING_(AllOf, 'a', 'l', 'l', 'O', 'f') + RAPIDJSON_STRING_(AnyOf, 'a', 'n', 'y', 'O', 'f') + RAPIDJSON_STRING_(OneOf, 'o', 'n', 'e', 'O', 'f') + RAPIDJSON_STRING_(Not, 'n', 'o', 't') + RAPIDJSON_STRING_(Properties, 'p', 'r', 'o', 'p', 'e', 'r', 't', 'i', 'e', 's') + RAPIDJSON_STRING_(Required, 'r', 'e', 'q', 'u', 'i', 'r', 'e', 'd') + RAPIDJSON_STRING_(Dependencies, 'd', 'e', 'p', 'e', 'n', 'd', 'e', 'n', 'c', 'i', 'e', 's') + RAPIDJSON_STRING_(PatternProperties, 'p', 'a', 't', 't', 'e', 'r', 'n', 'P', 'r', 'o', 'p', 'e', 'r', 't', 'i', 'e', 's') + RAPIDJSON_STRING_(AdditionalProperties, 'a', 'd', 'd', 'i', 't', 'i', 'o', 'n', 'a', 'l', 'P', 'r', 'o', 'p', 'e', 'r', 't', 'i', 'e', 's') + RAPIDJSON_STRING_(MinProperties, 'm', 'i', 'n', 'P', 'r', 'o', 'p', 'e', 'r', 't', 'i', 'e', 's') + RAPIDJSON_STRING_(MaxProperties, 'm', 'a', 'x', 'P', 'r', 'o', 'p', 'e', 'r', 't', 'i', 'e', 's') + RAPIDJSON_STRING_(Items, 'i', 't', 'e', 'm', 's') + RAPIDJSON_STRING_(MinItems, 'm', 'i', 'n', 'I', 't', 'e', 'm', 's') + RAPIDJSON_STRING_(MaxItems, 'm', 'a', 'x', 'I', 't', 'e', 'm', 's') + RAPIDJSON_STRING_(AdditionalItems, 'a', 'd', 'd', 'i', 't', 'i', 'o', 'n', 'a', 'l', 'I', 't', 'e', 'm', 's') + RAPIDJSON_STRING_(UniqueItems, 'u', 'n', 'i', 'q', 'u', 'e', 'I', 't', 'e', 'm', 's') + RAPIDJSON_STRING_(MinLength, 'm', 'i', 'n', 'L', 'e', 'n', 'g', 't', 'h') + RAPIDJSON_STRING_(MaxLength, 'm', 'a', 'x', 'L', 'e', 'n', 'g', 't', 'h') + RAPIDJSON_STRING_(Pattern, 'p', 'a', 't', 't', 'e', 'r', 'n') + RAPIDJSON_STRING_(Minimum, 'm', 'i', 'n', 'i', 'm', 'u', 'm') + RAPIDJSON_STRING_(Maximum, 'm', 'a', 'x', 'i', 'm', 'u', 'm') + RAPIDJSON_STRING_(ExclusiveMinimum, 'e', 'x', 'c', 'l', 'u', 's', 'i', 'v', 'e', 'M', 'i', 'n', 'i', 'm', 'u', 'm') + RAPIDJSON_STRING_(ExclusiveMaximum, 'e', 'x', 'c', 'l', 'u', 's', 'i', 'v', 'e', 'M', 'a', 'x', 'i', 'm', 'u', 'm') + RAPIDJSON_STRING_(MultipleOf, 'm', 'u', 'l', 't', 'i', 'p', 'l', 'e', 'O', 'f') + RAPIDJSON_STRING_(DefaultValue, 'd', 'e', 'f', 'a', 'u', 'l', 't') + RAPIDJSON_STRING_(Schema, '$', 's', 'c', 'h', 'e', 'm', 'a') + RAPIDJSON_STRING_(Ref, '$', 'r', 'e', 'f') + RAPIDJSON_STRING_(Id, 'i', 'd') + RAPIDJSON_STRING_(Swagger, 's', 'w', 'a', 'g', 'g', 'e', 'r') + RAPIDJSON_STRING_(OpenApi, 'o', 'p', 'e', 'n', 'a', 'p', 'i') + RAPIDJSON_STRING_(ReadOnly, 'r', 'e', 'a', 'd', 'O', 'n', 'l', 'y') + RAPIDJSON_STRING_(WriteOnly, 'w', 'r', 'i', 't', 'e', 'O', 'n', 'l', 'y') + RAPIDJSON_STRING_(Nullable, 'n', 'u', 'l', 'l', 'a', 'b', 'l', 'e') + +#undef RAPIDJSON_STRING_ + +private: + enum SchemaValueType { + kNullSchemaType, + kBooleanSchemaType, + kObjectSchemaType, + kArraySchemaType, + kStringSchemaType, + kNumberSchemaType, + kIntegerSchemaType, + kTotalSchemaType + }; + +#if RAPIDJSON_SCHEMA_USE_INTERNALREGEX + typedef internal::GenericRegex RegexType; +#elif RAPIDJSON_SCHEMA_USE_STDREGEX + typedef std::basic_regex RegexType; +#else + typedef char RegexType; +#endif + + struct SchemaArray { + SchemaArray() : schemas(), count() {} + ~SchemaArray() { AllocatorType::Free(schemas); } + const SchemaType** schemas; + SizeType begin; // begin index of context.validators + SizeType count; + }; + + template + void AddUniqueElement(V1& a, const V2& v) { + for (typename V1::ConstValueIterator itr = a.Begin(); itr != a.End(); ++itr) + if (*itr == v) + return; + V1 c(v, *allocator_); + a.PushBack(c, *allocator_); + } + + static const ValueType* GetMember(const ValueType& value, const ValueType& name) { + typename ValueType::ConstMemberIterator itr = value.FindMember(name); + return itr != value.MemberEnd() ? &(itr->value) : 0; + } + + static void AssignIfExist(bool& out, const ValueType& value, const ValueType& name) { + if (const ValueType* v = GetMember(value, name)) + if (v->IsBool()) + out = v->GetBool(); + } + + static void AssignIfExist(SizeType& out, const ValueType& value, const ValueType& name) { + if (const ValueType* v = GetMember(value, name)) + if (v->IsUint64() && v->GetUint64() <= SizeType(~0)) + out = static_cast(v->GetUint64()); + } + + void AssignIfExist(SchemaArray& out, SchemaDocumentType& schemaDocument, const PointerType& p, const ValueType& value, const ValueType& name, const ValueType& document) { + if (const ValueType* v = GetMember(value, name)) { + if (v->IsArray() && v->Size() > 0) { + PointerType q = p.Append(name, allocator_); + out.count = v->Size(); + out.schemas = static_cast(allocator_->Malloc(out.count * sizeof(const Schema*))); + memset(out.schemas, 0, sizeof(Schema*)* out.count); + for (SizeType i = 0; i < out.count; i++) + schemaDocument.CreateSchema(&out.schemas[i], q.Append(i, allocator_), (*v)[i], document, id_); + out.begin = validatorCount_; + validatorCount_ += out.count; + } + } + } + +#if RAPIDJSON_SCHEMA_USE_INTERNALREGEX + template + RegexType* CreatePattern(const ValueType& value, SchemaDocumentType* sd, const PointerType& p) { + if (value.IsString()) { + RegexType* r = new (allocator_->Malloc(sizeof(RegexType))) RegexType(value.GetString(), allocator_); + if (!r->IsValid()) { + sd->SchemaErrorValue(kSchemaErrorRegexInvalid, p, value.GetString(), value.GetStringLength()); + r->~RegexType(); + AllocatorType::Free(r); + r = 0; + } + return r; + } + return 0; + } + + static bool IsPatternMatch(const RegexType* pattern, const Ch *str, SizeType) { + GenericRegexSearch rs(*pattern); + return rs.Search(str); + } +#elif RAPIDJSON_SCHEMA_USE_STDREGEX + template + RegexType* CreatePattern(const ValueType& value, SchemaDocumentType* sd, const PointerType& p) { + if (value.IsString()) { + RegexType *r = static_cast(allocator_->Malloc(sizeof(RegexType))); + try { + return new (r) RegexType(value.GetString(), std::size_t(value.GetStringLength()), std::regex_constants::ECMAScript); + } + catch (const std::regex_error& e) { + sd->SchemaErrorValue(kSchemaErrorRegexInvalid, p, value.GetString(), value.GetStringLength()); + AllocatorType::Free(r); + } + } + return 0; + } + + static bool IsPatternMatch(const RegexType* pattern, const Ch *str, SizeType length) { + std::match_results r; + return std::regex_search(str, str + length, r, *pattern); + } +#else + template + RegexType* CreatePattern(const ValueType&) { + return 0; + } + + static bool IsPatternMatch(const RegexType*, const Ch *, SizeType) { return true; } +#endif // RAPIDJSON_SCHEMA_USE_STDREGEX + + void AddType(const ValueType& type) { + if (type == GetNullString() ) type_ |= 1 << kNullSchemaType; + else if (type == GetBooleanString()) type_ |= 1 << kBooleanSchemaType; + else if (type == GetObjectString() ) type_ |= 1 << kObjectSchemaType; + else if (type == GetArrayString() ) type_ |= 1 << kArraySchemaType; + else if (type == GetStringString() ) type_ |= 1 << kStringSchemaType; + else if (type == GetIntegerString()) type_ |= 1 << kIntegerSchemaType; + else if (type == GetNumberString() ) type_ |= (1 << kNumberSchemaType) | (1 << kIntegerSchemaType); + } + + // Creates parallel validators for allOf, anyOf, oneOf, not and schema dependencies, if required. + // Also creates a hasher for enums and array uniqueness, if required. + // Also a useful place to add type-independent error checks. + bool CreateParallelValidator(Context& context) const { + if (enum_ || context.arrayUniqueness) + context.hasher = context.factory.CreateHasher(); + + if (validatorCount_) { + RAPIDJSON_ASSERT(context.validators == 0); + context.validators = static_cast(context.factory.MallocState(sizeof(ISchemaValidator*) * validatorCount_)); + std::memset(context.validators, 0, sizeof(ISchemaValidator*) * validatorCount_); + context.validatorCount = validatorCount_; + + // Always return after first failure for these sub-validators + if (allOf_.schemas) + CreateSchemaValidators(context, allOf_, false); + + if (anyOf_.schemas) + CreateSchemaValidators(context, anyOf_, false); + + if (oneOf_.schemas) + CreateSchemaValidators(context, oneOf_, false); + + if (not_) + context.validators[notValidatorIndex_] = context.factory.CreateSchemaValidator(*not_, false); + + if (hasSchemaDependencies_) { + for (SizeType i = 0; i < propertyCount_; i++) + if (properties_[i].dependenciesSchema) + context.validators[properties_[i].dependenciesValidatorIndex] = context.factory.CreateSchemaValidator(*properties_[i].dependenciesSchema, false); + } + } + + // Add any other type-independent checks here + if (readOnly_ && (context.flags & kValidateWriteFlag)) { + context.error_handler.DisallowedWhenWriting(); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorReadOnly); + } + if (writeOnly_ && (context.flags & kValidateReadFlag)) { + context.error_handler.DisallowedWhenReading(); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorWriteOnly); + } + + return true; + } + + void CreateSchemaValidators(Context& context, const SchemaArray& schemas, const bool inheritContinueOnErrors) const { + for (SizeType i = 0; i < schemas.count; i++) + context.validators[schemas.begin + i] = context.factory.CreateSchemaValidator(*schemas.schemas[i], inheritContinueOnErrors); + } + + // O(n) + bool FindPropertyIndex(const ValueType& name, SizeType* outIndex) const { + SizeType len = name.GetStringLength(); + const Ch* str = name.GetString(); + for (SizeType index = 0; index < propertyCount_; index++) + if (properties_[index].name.GetStringLength() == len && + (std::memcmp(properties_[index].name.GetString(), str, sizeof(Ch) * len) == 0)) + { + *outIndex = index; + return true; + } + return false; + } + + bool CheckBool(Context& context, bool) const { + if (!(type_ & (1 << kBooleanSchemaType))) { + DisallowedType(context, GetBooleanString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + return true; + } + + bool CheckInt(Context& context, int64_t i) const { + if (!(type_ & ((1 << kIntegerSchemaType) | (1 << kNumberSchemaType)))) { + DisallowedType(context, GetIntegerString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + if (!minimum_.IsNull()) { + if (minimum_.IsInt64()) { + if (exclusiveMinimum_ ? i <= minimum_.GetInt64() : i < minimum_.GetInt64()) { + context.error_handler.BelowMinimum(i, minimum_, exclusiveMinimum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMinimum_ ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum); + } + } + else if (minimum_.IsUint64()) { + context.error_handler.BelowMinimum(i, minimum_, exclusiveMinimum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMinimum_ ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum); // i <= max(int64_t) < minimum.GetUint64() + } + else if (!CheckDoubleMinimum(context, static_cast(i))) + return false; + } + + if (!maximum_.IsNull()) { + if (maximum_.IsInt64()) { + if (exclusiveMaximum_ ? i >= maximum_.GetInt64() : i > maximum_.GetInt64()) { + context.error_handler.AboveMaximum(i, maximum_, exclusiveMaximum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMaximum_ ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum); + } + } + else if (maximum_.IsUint64()) { } + /* do nothing */ // i <= max(int64_t) < maximum_.GetUint64() + else if (!CheckDoubleMaximum(context, static_cast(i))) + return false; + } + + if (!multipleOf_.IsNull()) { + if (multipleOf_.IsUint64()) { + if (static_cast(i >= 0 ? i : -i) % multipleOf_.GetUint64() != 0) { + context.error_handler.NotMultipleOf(i, multipleOf_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMultipleOf); + } + } + else if (!CheckDoubleMultipleOf(context, static_cast(i))) + return false; + } + + return true; + } + + bool CheckUint(Context& context, uint64_t i) const { + if (!(type_ & ((1 << kIntegerSchemaType) | (1 << kNumberSchemaType)))) { + DisallowedType(context, GetIntegerString()); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorType); + } + + if (!minimum_.IsNull()) { + if (minimum_.IsUint64()) { + if (exclusiveMinimum_ ? i <= minimum_.GetUint64() : i < minimum_.GetUint64()) { + context.error_handler.BelowMinimum(i, minimum_, exclusiveMinimum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMinimum_ ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum); + } + } + else if (minimum_.IsInt64()) + /* do nothing */; // i >= 0 > minimum.Getint64() + else if (!CheckDoubleMinimum(context, static_cast(i))) + return false; + } + + if (!maximum_.IsNull()) { + if (maximum_.IsUint64()) { + if (exclusiveMaximum_ ? i >= maximum_.GetUint64() : i > maximum_.GetUint64()) { + context.error_handler.AboveMaximum(i, maximum_, exclusiveMaximum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMaximum_ ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum); + } + } + else if (maximum_.IsInt64()) { + context.error_handler.AboveMaximum(i, maximum_, exclusiveMaximum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMaximum_ ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum); // i >= 0 > maximum_ + } + else if (!CheckDoubleMaximum(context, static_cast(i))) + return false; + } + + if (!multipleOf_.IsNull()) { + if (multipleOf_.IsUint64()) { + if (i % multipleOf_.GetUint64() != 0) { + context.error_handler.NotMultipleOf(i, multipleOf_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMultipleOf); + } + } + else if (!CheckDoubleMultipleOf(context, static_cast(i))) + return false; + } + + return true; + } + + bool CheckDoubleMinimum(Context& context, double d) const { + if (exclusiveMinimum_ ? d <= minimum_.GetDouble() : d < minimum_.GetDouble()) { + context.error_handler.BelowMinimum(d, minimum_, exclusiveMinimum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMinimum_ ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum); + } + return true; + } + + bool CheckDoubleMaximum(Context& context, double d) const { + if (exclusiveMaximum_ ? d >= maximum_.GetDouble() : d > maximum_.GetDouble()) { + context.error_handler.AboveMaximum(d, maximum_, exclusiveMaximum_); + RAPIDJSON_INVALID_KEYWORD_RETURN(exclusiveMaximum_ ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum); + } + return true; + } + + bool CheckDoubleMultipleOf(Context& context, double d) const { + double a = std::abs(d), b = std::abs(multipleOf_.GetDouble()); + double q = a / b; + double qRounded = std::floor(q + 0.5); + double scaledEpsilon = (q + qRounded) * std::numeric_limits::epsilon(); + double difference = std::abs(qRounded - q); + bool isMultiple = difference <= scaledEpsilon || difference < (std::numeric_limits::min)(); + if (!isMultiple) { + context.error_handler.NotMultipleOf(d, multipleOf_); + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorMultipleOf); + } + return true; + } + + void DisallowedType(Context& context, const ValueType& actualType) const { + ErrorHandler& eh = context.error_handler; + eh.StartDisallowedType(); + + if (type_ & (1 << kNullSchemaType)) eh.AddExpectedType(GetNullString()); + if (type_ & (1 << kBooleanSchemaType)) eh.AddExpectedType(GetBooleanString()); + if (type_ & (1 << kObjectSchemaType)) eh.AddExpectedType(GetObjectString()); + if (type_ & (1 << kArraySchemaType)) eh.AddExpectedType(GetArrayString()); + if (type_ & (1 << kStringSchemaType)) eh.AddExpectedType(GetStringString()); + + if (type_ & (1 << kNumberSchemaType)) eh.AddExpectedType(GetNumberString()); + else if (type_ & (1 << kIntegerSchemaType)) eh.AddExpectedType(GetIntegerString()); + + eh.EndDisallowedType(actualType); + } + + struct Property { + Property() : schema(), dependenciesSchema(), dependenciesValidatorIndex(), dependencies(), required(false) {} + ~Property() { AllocatorType::Free(dependencies); } + SValue name; + const SchemaType* schema; + const SchemaType* dependenciesSchema; + SizeType dependenciesValidatorIndex; + bool* dependencies; + bool required; + }; + + struct PatternProperty { + PatternProperty() : schema(), pattern() {} + ~PatternProperty() { + if (pattern) { + pattern->~RegexType(); + AllocatorType::Free(pattern); + } + } + const SchemaType* schema; + RegexType* pattern; + }; + + AllocatorType* allocator_; + SValue uri_; + UriType id_; + Specification spec_; + PointerType pointer_; + const SchemaType* typeless_; + uint64_t* enum_; + SizeType enumCount_; + SchemaArray allOf_; + SchemaArray anyOf_; + SchemaArray oneOf_; + const SchemaType* not_; + unsigned type_; // bitmask of kSchemaType + SizeType validatorCount_; + SizeType notValidatorIndex_; + + Property* properties_; + const SchemaType* additionalPropertiesSchema_; + PatternProperty* patternProperties_; + SizeType patternPropertyCount_; + SizeType propertyCount_; + SizeType minProperties_; + SizeType maxProperties_; + bool additionalProperties_; + bool hasDependencies_; + bool hasRequired_; + bool hasSchemaDependencies_; + + const SchemaType* additionalItemsSchema_; + const SchemaType* itemsList_; + const SchemaType** itemsTuple_; + SizeType itemsTupleCount_; + SizeType minItems_; + SizeType maxItems_; + bool additionalItems_; + bool uniqueItems_; + + RegexType* pattern_; + SizeType minLength_; + SizeType maxLength_; + + SValue minimum_; + SValue maximum_; + SValue multipleOf_; + bool exclusiveMinimum_; + bool exclusiveMaximum_; + + SizeType defaultValueLength_; + + bool readOnly_; + bool writeOnly_; + bool nullable_; +}; + +template +struct TokenHelper { + RAPIDJSON_FORCEINLINE static void AppendIndexToken(Stack& documentStack, SizeType index) { + *documentStack.template Push() = '/'; + char buffer[21]; + size_t length = static_cast((sizeof(SizeType) == 4 ? u32toa(index, buffer) : u64toa(index, buffer)) - buffer); + for (size_t i = 0; i < length; i++) + *documentStack.template Push() = static_cast(buffer[i]); + } +}; + +// Partial specialized version for char to prevent buffer copying. +template +struct TokenHelper { + RAPIDJSON_FORCEINLINE static void AppendIndexToken(Stack& documentStack, SizeType index) { + RAPIDJSON_IF_CONSTEXPR (sizeof(SizeType) == 4) { + char *buffer = documentStack.template Push(1 + 10); // '/' + uint + *buffer++ = '/'; + const char* end = internal::u32toa(index, buffer); + documentStack.template Pop(static_cast(10 - (end - buffer))); + } + else { + char *buffer = documentStack.template Push(1 + 20); // '/' + uint64 + *buffer++ = '/'; + const char* end = internal::u64toa(index, buffer); + documentStack.template Pop(static_cast(20 - (end - buffer))); + } + } +}; + +} // namespace internal + +/////////////////////////////////////////////////////////////////////////////// +// IGenericRemoteSchemaDocumentProvider + +template +class IGenericRemoteSchemaDocumentProvider { +public: + typedef typename SchemaDocumentType::Ch Ch; + typedef typename SchemaDocumentType::ValueType ValueType; + typedef typename SchemaDocumentType::AllocatorType AllocatorType; + + virtual ~IGenericRemoteSchemaDocumentProvider() {} + virtual const SchemaDocumentType* GetRemoteDocument(const Ch* uri, SizeType length) = 0; + virtual const SchemaDocumentType* GetRemoteDocument(const GenericUri uri, Specification& spec) { + // Default implementation just calls through for compatibility + // Following line suppresses unused parameter warning + (void)spec; + // printf("GetRemoteDocument: %d %d\n", spec.draft, spec.oapi); + return GetRemoteDocument(uri.GetBaseString(), uri.GetBaseStringLength()); + } +}; + +/////////////////////////////////////////////////////////////////////////////// +// GenericSchemaDocument + +//! JSON schema document. +/*! + A JSON schema document is a compiled version of a JSON schema. + It is basically a tree of internal::Schema. + + \note This is an immutable class (i.e. its instance cannot be modified after construction). + \tparam ValueT Type of JSON value (e.g. \c Value ), which also determine the encoding. + \tparam Allocator Allocator type for allocating memory of this document. +*/ +template +class GenericSchemaDocument { +public: + typedef ValueT ValueType; + typedef IGenericRemoteSchemaDocumentProvider IRemoteSchemaDocumentProviderType; + typedef Allocator AllocatorType; + typedef typename ValueType::EncodingType EncodingType; + typedef typename EncodingType::Ch Ch; + typedef internal::Schema SchemaType; + typedef GenericPointer PointerType; + typedef GenericValue GValue; + typedef GenericUri UriType; + typedef GenericStringRef StringRefType; + friend class internal::Schema; + template + friend class GenericSchemaValidator; + + //! Constructor. + /*! + Compile a JSON document into schema document. + + \param document A JSON document as source. + \param uri The base URI of this schema document for purposes of violation reporting. + \param uriLength Length of \c name, in code points. + \param remoteProvider An optional remote schema document provider for resolving remote reference. Can be null. + \param allocator An optional allocator instance for allocating memory. Can be null. + \param pointer An optional JSON pointer to the start of the schema document + \param spec Optional schema draft or OpenAPI version. Used if no specification in document. Defaults to draft-04. + */ + explicit GenericSchemaDocument(const ValueType& document, const Ch* uri = 0, SizeType uriLength = 0, + IRemoteSchemaDocumentProviderType* remoteProvider = 0, Allocator* allocator = 0, + const PointerType& pointer = PointerType(), // PR #1393 + const Specification& spec = Specification(kDraft04)) : + remoteProvider_(remoteProvider), + allocator_(allocator), + ownAllocator_(), + root_(), + typeless_(), + schemaMap_(allocator, kInitialSchemaMapSize), + schemaRef_(allocator, kInitialSchemaRefSize), + spec_(spec), + error_(kObjectType), + currentError_() + { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaDocument::GenericSchemaDocument"); + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + + Ch noUri[1] = {0}; + uri_.SetString(uri ? uri : noUri, uriLength, *allocator_); + docId_ = UriType(uri_, allocator_); + + typeless_ = static_cast(allocator_->Malloc(sizeof(SchemaType))); + new (typeless_) SchemaType(this, PointerType(), ValueType(kObjectType).Move(), ValueType(kObjectType).Move(), allocator_, docId_); + + // Establish the schema draft or open api version. + // We only ever look for '$schema' or 'swagger' or 'openapi' at the root of the document. + SetSchemaSpecification(document); + + // Generate root schema, it will call CreateSchema() to create sub-schemas, + // And call HandleRefSchema() if there are $ref. + // PR #1393 use input pointer if supplied + root_ = typeless_; + if (pointer.GetTokenCount() == 0) { + CreateSchemaRecursive(&root_, pointer, document, document, docId_); + } + else if (const ValueType* v = pointer.Get(document)) { + CreateSchema(&root_, pointer, *v, document, docId_); + } + else { + GenericStringBuffer sb; + pointer.StringifyUriFragment(sb); + SchemaErrorValue(kSchemaErrorStartUnknown, PointerType(), sb.GetString(), static_cast(sb.GetSize() / sizeof(Ch))); + } + + RAPIDJSON_ASSERT(root_ != 0); + + schemaRef_.ShrinkToFit(); // Deallocate all memory for ref + } + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + //! Move constructor in C++11 + GenericSchemaDocument(GenericSchemaDocument&& rhs) RAPIDJSON_NOEXCEPT : + remoteProvider_(rhs.remoteProvider_), + allocator_(rhs.allocator_), + ownAllocator_(rhs.ownAllocator_), + root_(rhs.root_), + typeless_(rhs.typeless_), + schemaMap_(std::move(rhs.schemaMap_)), + schemaRef_(std::move(rhs.schemaRef_)), + uri_(std::move(rhs.uri_)), + docId_(std::move(rhs.docId_)), + spec_(rhs.spec_), + error_(std::move(rhs.error_)), + currentError_(std::move(rhs.currentError_)) + { + rhs.remoteProvider_ = 0; + rhs.allocator_ = 0; + rhs.ownAllocator_ = 0; + rhs.typeless_ = 0; + } +#endif + + //! Destructor + ~GenericSchemaDocument() { + while (!schemaMap_.Empty()) + schemaMap_.template Pop(1)->~SchemaEntry(); + + if (typeless_) { + typeless_->~SchemaType(); + Allocator::Free(typeless_); + } + + // these may contain some allocator data so clear before deleting ownAllocator_ + uri_.SetNull(); + error_.SetNull(); + currentError_.SetNull(); + + RAPIDJSON_DELETE(ownAllocator_); + } + + const GValue& GetURI() const { return uri_; } + + const Specification& GetSpecification() const { return spec_; } + bool IsSupportedSpecification() const { return spec_.IsSupported(); } + + //! Static method to get the specification of any schema document + // Returns kDraftNone if document is silent + static const Specification GetSpecification(const ValueType& document) { + SchemaDraft draft = GetSchemaDraft(document); + if (draft != kDraftNone) + return Specification(draft); + else { + OpenApiVersion oapi = GetOpenApiVersion(document); + if (oapi != kVersionNone) + return Specification(oapi); + } + return Specification(kDraftNone); + } + + //! Get the root schema. + const SchemaType& GetRoot() const { return *root_; } + + //! Gets the error object. + GValue& GetError() { return error_; } + const GValue& GetError() const { return error_; } + + static const StringRefType& GetSchemaErrorKeyword(SchemaErrorCode schemaErrorCode) { + switch (schemaErrorCode) { + case kSchemaErrorStartUnknown: return GetStartUnknownString(); + case kSchemaErrorRefPlainName: return GetRefPlainNameString(); + case kSchemaErrorRefInvalid: return GetRefInvalidString(); + case kSchemaErrorRefPointerInvalid: return GetRefPointerInvalidString(); + case kSchemaErrorRefUnknown: return GetRefUnknownString(); + case kSchemaErrorRefCyclical: return GetRefCyclicalString(); + case kSchemaErrorRefNoRemoteProvider: return GetRefNoRemoteProviderString(); + case kSchemaErrorRefNoRemoteSchema: return GetRefNoRemoteSchemaString(); + case kSchemaErrorRegexInvalid: return GetRegexInvalidString(); + case kSchemaErrorSpecUnknown: return GetSpecUnknownString(); + case kSchemaErrorSpecUnsupported: return GetSpecUnsupportedString(); + case kSchemaErrorSpecIllegal: return GetSpecIllegalString(); + case kSchemaErrorReadOnlyAndWriteOnly: return GetReadOnlyAndWriteOnlyString(); + default: return GetNullString(); + } + } + + //! Default error method + void SchemaError(const SchemaErrorCode code, const PointerType& location) { + currentError_ = GValue(kObjectType); + AddCurrentError(code, location); + } + + //! Method for error with single string value insert + void SchemaErrorValue(const SchemaErrorCode code, const PointerType& location, const Ch* value, SizeType length) { + currentError_ = GValue(kObjectType); + currentError_.AddMember(GetValueString(), GValue(value, length, *allocator_).Move(), *allocator_); + AddCurrentError(code, location); + } + + //! Method for error with invalid pointer + void SchemaErrorPointer(const SchemaErrorCode code, const PointerType& location, const Ch* value, SizeType length, const PointerType& pointer) { + currentError_ = GValue(kObjectType); + currentError_.AddMember(GetValueString(), GValue(value, length, *allocator_).Move(), *allocator_); + currentError_.AddMember(GetOffsetString(), static_cast(pointer.GetParseErrorOffset() / sizeof(Ch)), *allocator_); + AddCurrentError(code, location); + } + + private: + //! Prohibit copying + GenericSchemaDocument(const GenericSchemaDocument&); + //! Prohibit assignment + GenericSchemaDocument& operator=(const GenericSchemaDocument&); + + typedef const PointerType* SchemaRefPtr; // PR #1393 + + struct SchemaEntry { + SchemaEntry(const PointerType& p, SchemaType* s, bool o, Allocator* allocator) : pointer(p, allocator), schema(s), owned(o) {} + ~SchemaEntry() { + if (owned) { + schema->~SchemaType(); + Allocator::Free(schema); + } + } + PointerType pointer; + SchemaType* schema; + bool owned; + }; + + void AddErrorInstanceLocation(GValue& result, const PointerType& location) { + GenericStringBuffer sb; + location.StringifyUriFragment(sb); + GValue instanceRef(sb.GetString(), static_cast(sb.GetSize() / sizeof(Ch)), *allocator_); + result.AddMember(GetInstanceRefString(), instanceRef, *allocator_); + } + + void AddError(GValue& keyword, GValue& error) { + typename GValue::MemberIterator member = error_.FindMember(keyword); + if (member == error_.MemberEnd()) + error_.AddMember(keyword, error, *allocator_); + else { + if (member->value.IsObject()) { + GValue errors(kArrayType); + errors.PushBack(member->value, *allocator_); + member->value = errors; + } + member->value.PushBack(error, *allocator_); + } + } + + void AddCurrentError(const SchemaErrorCode code, const PointerType& location) { + RAPIDJSON_SCHEMA_PRINT(InvalidKeyword, GetSchemaErrorKeyword(code)); + currentError_.AddMember(GetErrorCodeString(), code, *allocator_); + AddErrorInstanceLocation(currentError_, location); + AddError(GValue(GetSchemaErrorKeyword(code)).Move(), currentError_); + } + +#define RAPIDJSON_STRING_(name, ...) \ + static const StringRefType& Get##name##String() {\ + static const Ch s[] = { __VA_ARGS__, '\0' };\ + static const StringRefType v(s, static_cast(sizeof(s) / sizeof(Ch) - 1)); \ + return v;\ + } + + RAPIDJSON_STRING_(InstanceRef, 'i', 'n', 's', 't', 'a', 'n', 'c', 'e', 'R', 'e', 'f') + RAPIDJSON_STRING_(ErrorCode, 'e', 'r', 'r', 'o', 'r', 'C', 'o', 'd', 'e') + RAPIDJSON_STRING_(Value, 'v', 'a', 'l', 'u', 'e') + RAPIDJSON_STRING_(Offset, 'o', 'f', 'f', 's', 'e', 't') + + RAPIDJSON_STRING_(Null, 'n', 'u', 'l', 'l') + RAPIDJSON_STRING_(SpecUnknown, 'S', 'p', 'e', 'c', 'U', 'n', 'k', 'n', 'o', 'w', 'n') + RAPIDJSON_STRING_(SpecUnsupported, 'S', 'p', 'e', 'c', 'U', 'n', 's', 'u', 'p', 'p', 'o', 'r', 't', 'e', 'd') + RAPIDJSON_STRING_(SpecIllegal, 'S', 'p', 'e', 'c', 'I', 'l', 'l', 'e', 'g', 'a', 'l') + RAPIDJSON_STRING_(StartUnknown, 'S', 't', 'a', 'r', 't', 'U', 'n', 'k', 'n', 'o', 'w', 'n') + RAPIDJSON_STRING_(RefPlainName, 'R', 'e', 'f', 'P', 'l', 'a', 'i', 'n', 'N', 'a', 'm', 'e') + RAPIDJSON_STRING_(RefInvalid, 'R', 'e', 'f', 'I', 'n', 'v', 'a', 'l', 'i', 'd') + RAPIDJSON_STRING_(RefPointerInvalid, 'R', 'e', 'f', 'P', 'o', 'i', 'n', 't', 'e', 'r', 'I', 'n', 'v', 'a', 'l', 'i', 'd') + RAPIDJSON_STRING_(RefUnknown, 'R', 'e', 'f', 'U', 'n', 'k', 'n', 'o', 'w', 'n') + RAPIDJSON_STRING_(RefCyclical, 'R', 'e', 'f', 'C', 'y', 'c', 'l', 'i', 'c', 'a', 'l') + RAPIDJSON_STRING_(RefNoRemoteProvider, 'R', 'e', 'f', 'N', 'o', 'R', 'e', 'm', 'o', 't', 'e', 'P', 'r', 'o', 'v', 'i', 'd', 'e', 'r') + RAPIDJSON_STRING_(RefNoRemoteSchema, 'R', 'e', 'f', 'N', 'o', 'R', 'e', 'm', 'o', 't', 'e', 'S', 'c', 'h', 'e', 'm', 'a') + RAPIDJSON_STRING_(ReadOnlyAndWriteOnly, 'R', 'e', 'a', 'd', 'O', 'n', 'l', 'y', 'A', 'n', 'd', 'W', 'r', 'i', 't', 'e', 'O', 'n', 'l', 'y') + RAPIDJSON_STRING_(RegexInvalid, 'R', 'e', 'g', 'e', 'x', 'I', 'n', 'v', 'a', 'l', 'i', 'd') + +#undef RAPIDJSON_STRING_ + + // Static method to get schema draft of any schema document + static SchemaDraft GetSchemaDraft(const ValueType& document) { + static const Ch kDraft03String[] = { 'h', 't', 't', 'p', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '-', '0', '3', '/', 's', 'c', 'h', 'e', 'm', 'a', '#', '\0' }; + static const Ch kDraft04String[] = { 'h', 't', 't', 'p', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '-', '0', '4', '/', 's', 'c', 'h', 'e', 'm', 'a', '#', '\0' }; + static const Ch kDraft05String[] = { 'h', 't', 't', 'p', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '-', '0', '5', '/', 's', 'c', 'h', 'e', 'm', 'a', '#', '\0' }; + static const Ch kDraft06String[] = { 'h', 't', 't', 'p', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '-', '0', '6', '/', 's', 'c', 'h', 'e', 'm', 'a', '#', '\0' }; + static const Ch kDraft07String[] = { 'h', 't', 't', 'p', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '-', '0', '7', '/', 's', 'c', 'h', 'e', 'm', 'a', '#', '\0' }; + static const Ch kDraft2019_09String[] = { 'h', 't', 't', 'p', 's', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '/', '2', '0', '1', '9', '-', '0', '9', '/', 's', 'c', 'h', 'e', 'm', 'a', '\0' }; + static const Ch kDraft2020_12String[] = { 'h', 't', 't', 'p', 's', ':', '/', '/', 'j', 's', 'o', 'n', '-', 's', 'c', 'h', 'e', 'm', 'a', '.', 'o', 'r', 'g', '/', 'd', 'r', 'a', 'f', 't', '/', '2', '0', '2', '0', '-', '1', '2', '/', 's', 'c', 'h', 'e', 'm', 'a', '\0' }; + + if (!document.IsObject()) { + return kDraftNone; + } + + // Get the schema draft from the $schema keyword at the supplied location + typename ValueType::ConstMemberIterator itr = document.FindMember(SchemaType::GetSchemaString()); + if (itr != document.MemberEnd()) { + if (!itr->value.IsString()) return kDraftUnknown; + const UriType draftUri(itr->value); + // Check base uri for match + if (draftUri.Match(UriType(kDraft04String), false)) return kDraft04; + if (draftUri.Match(UriType(kDraft05String), false)) return kDraft05; + if (draftUri.Match(UriType(kDraft06String), false)) return kDraft06; + if (draftUri.Match(UriType(kDraft07String), false)) return kDraft07; + if (draftUri.Match(UriType(kDraft03String), false)) return kDraft03; + if (draftUri.Match(UriType(kDraft2019_09String), false)) return kDraft2019_09; + if (draftUri.Match(UriType(kDraft2020_12String), false)) return kDraft2020_12; + return kDraftUnknown; + } + // $schema not found + return kDraftNone; + } + + + // Get open api version of any schema document + static OpenApiVersion GetOpenApiVersion(const ValueType& document) { + static const Ch kVersion20String[] = { '2', '.', '0', '\0' }; + static const Ch kVersion30String[] = { '3', '.', '0', '.', '\0' }; // ignore patch level + static const Ch kVersion31String[] = { '3', '.', '1', '.', '\0' }; // ignore patch level + static SizeType len = internal::StrLen(kVersion30String); + + if (!document.IsObject()) { + return kVersionNone; + } + + // Get the open api version from the swagger / openapi keyword at the supplied location + typename ValueType::ConstMemberIterator itr = document.FindMember(SchemaType::GetSwaggerString()); + if (itr == document.MemberEnd()) itr = document.FindMember(SchemaType::GetOpenApiString()); + if (itr != document.MemberEnd()) { + if (!itr->value.IsString()) return kVersionUnknown; + const ValueType kVersion20Value(kVersion20String); + if (kVersion20Value == itr->value) return kVersion20; // must match 2.0 exactly + const ValueType kVersion30Value(kVersion30String); + if (itr->value.GetStringLength() > len && kVersion30Value == ValueType(itr->value.GetString(), len)) return kVersion30; // must match 3.0.x + const ValueType kVersion31Value(kVersion31String); + if (itr->value.GetStringLength() > len && kVersion31Value == ValueType(itr->value.GetString(), len)) return kVersion31; // must match 3.1.x + return kVersionUnknown; + } + // swagger or openapi not found + return kVersionNone; + } + + // Get the draft of the schema or the open api version (which implies the draft). + // Report an error if schema draft or open api version not supported or not recognized, or both in document, and carry on. + void SetSchemaSpecification(const ValueType& document) { + // Look for '$schema', 'swagger' or 'openapi' keyword at document root + SchemaDraft docDraft = GetSchemaDraft(document); + OpenApiVersion docOapi = GetOpenApiVersion(document); + // Error if both in document + if (docDraft != kDraftNone && docOapi != kVersionNone) + SchemaError(kSchemaErrorSpecIllegal, PointerType()); + // Use document draft or open api version if present or use spec from constructor + if (docDraft != kDraftNone) + spec_ = Specification(docDraft); + else if (docOapi != kVersionNone) + spec_ = Specification(docOapi); + // Error if draft or version unknown + if (spec_.draft == kDraftUnknown || spec_.oapi == kVersionUnknown) + SchemaError(kSchemaErrorSpecUnknown, PointerType()); + else if (!spec_.IsSupported()) + SchemaError(kSchemaErrorSpecUnsupported, PointerType()); + } + + // Changed by PR #1393 + void CreateSchemaRecursive(const SchemaType** schema, const PointerType& pointer, const ValueType& v, const ValueType& document, const UriType& id) { + if (v.GetType() == kObjectType) { + UriType newid = UriType(CreateSchema(schema, pointer, v, document, id), allocator_); + + for (typename ValueType::ConstMemberIterator itr = v.MemberBegin(); itr != v.MemberEnd(); ++itr) + CreateSchemaRecursive(0, pointer.Append(itr->name, allocator_), itr->value, document, newid); + } + else if (v.GetType() == kArrayType) + for (SizeType i = 0; i < v.Size(); i++) + CreateSchemaRecursive(0, pointer.Append(i, allocator_), v[i], document, id); + } + + // Changed by PR #1393 + const UriType& CreateSchema(const SchemaType** schema, const PointerType& pointer, const ValueType& v, const ValueType& document, const UriType& id) { + RAPIDJSON_ASSERT(pointer.IsValid()); + GenericStringBuffer sb; + pointer.StringifyUriFragment(sb); + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaDocument::CreateSchema", sb.GetString(), id.GetString()); + if (v.IsObject()) { + if (const SchemaType* sc = GetSchema(pointer)) { + if (schema) + *schema = sc; + AddSchemaRefs(const_cast(sc)); + } + else if (!HandleRefSchema(pointer, schema, v, document, id)) { + // The new schema constructor adds itself and its $ref(s) to schemaMap_ + SchemaType* s = new (allocator_->Malloc(sizeof(SchemaType))) SchemaType(this, pointer, v, document, allocator_, id); + if (schema) + *schema = s; + return s->GetId(); + } + } + else { + if (schema) + *schema = typeless_; + AddSchemaRefs(typeless_); + } + return id; + } + + // Changed by PR #1393 + // TODO should this return a UriType& ? + bool HandleRefSchema(const PointerType& source, const SchemaType** schema, const ValueType& v, const ValueType& document, const UriType& id) { + typename ValueType::ConstMemberIterator itr = v.FindMember(SchemaType::GetRefString()); + if (itr == v.MemberEnd()) + return false; + + GenericStringBuffer sb; + source.StringifyUriFragment(sb); + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaDocument::HandleRefSchema", sb.GetString(), id.GetString()); + // Resolve the source pointer to the $ref'ed schema (finally) + new (schemaRef_.template Push()) SchemaRefPtr(&source); + + if (itr->value.IsString()) { + SizeType len = itr->value.GetStringLength(); + if (len == 0) + SchemaError(kSchemaErrorRefInvalid, source); + else { + // First resolve $ref against the in-scope id + UriType scopeId = UriType(id, allocator_); + UriType ref = UriType(itr->value, allocator_).Resolve(scopeId, allocator_); + RAPIDJSON_SCHEMA_PRINT(SchemaIds, id.GetString(), itr->value.GetString(), ref.GetString()); + // See if the resolved $ref minus the fragment matches a resolved id in this document + // Search from the root. Returns the subschema in the document and its absolute JSON pointer. + PointerType basePointer = PointerType(); + const ValueType *base = FindId(document, ref, basePointer, docId_, false); + if (!base) { + // Remote reference - call the remote document provider + if (!remoteProvider_) + SchemaError(kSchemaErrorRefNoRemoteProvider, source); + else { + if (const GenericSchemaDocument* remoteDocument = remoteProvider_->GetRemoteDocument(ref, spec_)) { + const Ch* s = ref.GetFragString(); + len = ref.GetFragStringLength(); + if (len <= 1 || s[1] == '/') { + // JSON pointer fragment, absolute in the remote schema + const PointerType pointer(s, len, allocator_); + if (!pointer.IsValid()) + SchemaErrorPointer(kSchemaErrorRefPointerInvalid, source, s, len, pointer); + else { + // Get the subschema + if (const SchemaType *sc = remoteDocument->GetSchema(pointer)) { + if (schema) + *schema = sc; + AddSchemaRefs(const_cast(sc)); + return true; + } else + SchemaErrorValue(kSchemaErrorRefUnknown, source, ref.GetString(), ref.GetStringLength()); + } + } else + // Plain name fragment, not allowed in remote schema + SchemaErrorValue(kSchemaErrorRefPlainName, source, s, len); + } else + SchemaErrorValue(kSchemaErrorRefNoRemoteSchema, source, ref.GetString(), ref.GetStringLength()); + } + } + else { // Local reference + const Ch* s = ref.GetFragString(); + len = ref.GetFragStringLength(); + if (len <= 1 || s[1] == '/') { + // JSON pointer fragment, relative to the resolved URI + const PointerType relPointer(s, len, allocator_); + if (!relPointer.IsValid()) + SchemaErrorPointer(kSchemaErrorRefPointerInvalid, source, s, len, relPointer); + else { + // Get the subschema + if (const ValueType *pv = relPointer.Get(*base)) { + // Now get the absolute JSON pointer by adding relative to base + PointerType pointer(basePointer, allocator_); + for (SizeType i = 0; i < relPointer.GetTokenCount(); i++) + pointer = pointer.Append(relPointer.GetTokens()[i], allocator_); + if (IsCyclicRef(pointer)) + SchemaErrorValue(kSchemaErrorRefCyclical, source, ref.GetString(), ref.GetStringLength()); + else { + // Call CreateSchema recursively, but first compute the in-scope id for the $ref target as we have jumped there + // TODO: cache pointer <-> id mapping + size_t unresolvedTokenIndex; + scopeId = pointer.GetUri(document, docId_, &unresolvedTokenIndex, allocator_); + CreateSchema(schema, pointer, *pv, document, scopeId); + return true; + } + } else + SchemaErrorValue(kSchemaErrorRefUnknown, source, ref.GetString(), ref.GetStringLength()); + } + } else { + // Plain name fragment, relative to the resolved URI + // Not supported in open api 2.0 and 3.0 + PointerType pointer(allocator_); + if (spec_.oapi == kVersion20 || spec_.oapi == kVersion30) + SchemaErrorValue(kSchemaErrorRefPlainName, source, s, len); + // See if the fragment matches an id in this document. + // Search from the base we just established. Returns the subschema in the document and its absolute JSON pointer. + else if (const ValueType *pv = FindId(*base, ref, pointer, UriType(ref.GetBaseString(), ref.GetBaseStringLength(), allocator_), true, basePointer)) { + if (IsCyclicRef(pointer)) + SchemaErrorValue(kSchemaErrorRefCyclical, source, ref.GetString(), ref.GetStringLength()); + else { + // Call CreateSchema recursively, but first compute the in-scope id for the $ref target as we have jumped there + // TODO: cache pointer <-> id mapping + size_t unresolvedTokenIndex; + scopeId = pointer.GetUri(document, docId_, &unresolvedTokenIndex, allocator_); + CreateSchema(schema, pointer, *pv, document, scopeId); + return true; + } + } else + SchemaErrorValue(kSchemaErrorRefUnknown, source, ref.GetString(), ref.GetStringLength()); + } + } + } + } + + // Invalid/Unknown $ref + if (schema) + *schema = typeless_; + AddSchemaRefs(typeless_); + return true; + } + + //! Find the first subschema with a resolved 'id' that matches the specified URI. + // If full specified use all URI else ignore fragment. + // If found, return a pointer to the subschema and its JSON pointer. + // TODO cache pointer <-> id mapping + ValueType* FindId(const ValueType& doc, const UriType& finduri, PointerType& resptr, const UriType& baseuri, bool full, const PointerType& here = PointerType()) const { + SizeType i = 0; + ValueType* resval = 0; + UriType tempuri = UriType(finduri, allocator_); + UriType localuri = UriType(baseuri, allocator_); + if (doc.GetType() == kObjectType) { + // Establish the base URI of this object + typename ValueType::ConstMemberIterator m = doc.FindMember(SchemaType::GetIdString()); + if (m != doc.MemberEnd() && m->value.GetType() == kStringType) { + localuri = UriType(m->value, allocator_).Resolve(baseuri, allocator_); + } + // See if it matches + if (localuri.Match(finduri, full)) { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaDocument::FindId (match)", full ? localuri.GetString() : localuri.GetBaseString()); + resval = const_cast(&doc); + resptr = here; + return resval; + } + // No match, continue looking + for (m = doc.MemberBegin(); m != doc.MemberEnd(); ++m) { + if (m->value.GetType() == kObjectType || m->value.GetType() == kArrayType) { + resval = FindId(m->value, finduri, resptr, localuri, full, here.Append(m->name.GetString(), m->name.GetStringLength(), allocator_)); + } + if (resval) break; + } + } else if (doc.GetType() == kArrayType) { + // Continue looking + for (typename ValueType::ConstValueIterator v = doc.Begin(); v != doc.End(); ++v) { + if (v->GetType() == kObjectType || v->GetType() == kArrayType) { + resval = FindId(*v, finduri, resptr, localuri, full, here.Append(i, allocator_)); + } + if (resval) break; + i++; + } + } + return resval; + } + + // Added by PR #1393 + void AddSchemaRefs(SchemaType* schema) { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaDocument::AddSchemaRefs"); + while (!schemaRef_.Empty()) { + SchemaRefPtr *ref = schemaRef_.template Pop(1); + SchemaEntry *entry = schemaMap_.template Push(); + new (entry) SchemaEntry(**ref, schema, false, allocator_); + } + } + + // Added by PR #1393 + bool IsCyclicRef(const PointerType& pointer) const { + for (const SchemaRefPtr* ref = schemaRef_.template Bottom(); ref != schemaRef_.template End(); ++ref) + if (pointer == **ref) + return true; + return false; + } + + const SchemaType* GetSchema(const PointerType& pointer) const { + for (const SchemaEntry* target = schemaMap_.template Bottom(); target != schemaMap_.template End(); ++target) + if (pointer == target->pointer) + return target->schema; + return 0; + } + + PointerType GetPointer(const SchemaType* schema) const { + for (const SchemaEntry* target = schemaMap_.template Bottom(); target != schemaMap_.template End(); ++target) + if (schema == target->schema) + return target->pointer; + return PointerType(); + } + + const SchemaType* GetTypeless() const { return typeless_; } + + static const size_t kInitialSchemaMapSize = 64; + static const size_t kInitialSchemaRefSize = 64; + + IRemoteSchemaDocumentProviderType* remoteProvider_; + Allocator *allocator_; + Allocator *ownAllocator_; + const SchemaType* root_; //!< Root schema. + SchemaType* typeless_; + internal::Stack schemaMap_; // Stores created Pointer -> Schemas + internal::Stack schemaRef_; // Stores Pointer(s) from $ref(s) until resolved + GValue uri_; // Schema document URI + UriType docId_; + Specification spec_; + GValue error_; + GValue currentError_; +}; + +//! GenericSchemaDocument using Value type. +typedef GenericSchemaDocument SchemaDocument; +//! IGenericRemoteSchemaDocumentProvider using SchemaDocument. +typedef IGenericRemoteSchemaDocumentProvider IRemoteSchemaDocumentProvider; + +/////////////////////////////////////////////////////////////////////////////// +// GenericSchemaValidator + +//! JSON Schema Validator. +/*! + A SAX style JSON schema validator. + It uses a \c GenericSchemaDocument to validate SAX events. + It delegates the incoming SAX events to an output handler. + The default output handler does nothing. + It can be reused multiple times by calling \c Reset(). + + \tparam SchemaDocumentType Type of schema document. + \tparam OutputHandler Type of output handler. Default handler does nothing. + \tparam StateAllocator Allocator for storing the internal validation states. +*/ +template < + typename SchemaDocumentType, + typename OutputHandler = BaseReaderHandler, + typename StateAllocator = CrtAllocator> +class GenericSchemaValidator : + public internal::ISchemaStateFactory, + public internal::ISchemaValidator, + public internal::IValidationErrorHandler { +public: + typedef typename SchemaDocumentType::SchemaType SchemaType; + typedef typename SchemaDocumentType::PointerType PointerType; + typedef typename SchemaType::EncodingType EncodingType; + typedef typename SchemaType::SValue SValue; + typedef typename EncodingType::Ch Ch; + typedef GenericStringRef StringRefType; + typedef GenericValue ValueType; + + //! Constructor without output handler. + /*! + \param schemaDocument The schema document to conform to. + \param allocator Optional allocator for storing internal validation states. + \param schemaStackCapacity Optional initial capacity of schema path stack. + \param documentStackCapacity Optional initial capacity of document path stack. + */ + GenericSchemaValidator( + const SchemaDocumentType& schemaDocument, + StateAllocator* allocator = 0, + size_t schemaStackCapacity = kDefaultSchemaStackCapacity, + size_t documentStackCapacity = kDefaultDocumentStackCapacity) + : + schemaDocument_(&schemaDocument), + root_(schemaDocument.GetRoot()), + stateAllocator_(allocator), + ownStateAllocator_(0), + schemaStack_(allocator, schemaStackCapacity), + documentStack_(allocator, documentStackCapacity), + outputHandler_(0), + error_(kObjectType), + currentError_(), + missingDependents_(), + valid_(true), + flags_(kValidateDefaultFlags), + depth_(0) + { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::GenericSchemaValidator"); + } + + //! Constructor with output handler. + /*! + \param schemaDocument The schema document to conform to. + \param allocator Optional allocator for storing internal validation states. + \param schemaStackCapacity Optional initial capacity of schema path stack. + \param documentStackCapacity Optional initial capacity of document path stack. + */ + GenericSchemaValidator( + const SchemaDocumentType& schemaDocument, + OutputHandler& outputHandler, + StateAllocator* allocator = 0, + size_t schemaStackCapacity = kDefaultSchemaStackCapacity, + size_t documentStackCapacity = kDefaultDocumentStackCapacity) + : + schemaDocument_(&schemaDocument), + root_(schemaDocument.GetRoot()), + stateAllocator_(allocator), + ownStateAllocator_(0), + schemaStack_(allocator, schemaStackCapacity), + documentStack_(allocator, documentStackCapacity), + outputHandler_(&outputHandler), + error_(kObjectType), + currentError_(), + missingDependents_(), + valid_(true), + flags_(kValidateDefaultFlags), + depth_(0) + { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::GenericSchemaValidator (output handler)"); + } + + //! Destructor. + ~GenericSchemaValidator() { + Reset(); + RAPIDJSON_DELETE(ownStateAllocator_); + } + + //! Reset the internal states. + void Reset() { + while (!schemaStack_.Empty()) + PopSchema(); + documentStack_.Clear(); + ResetError(); + } + + //! Reset the error state. + void ResetError() { + error_.SetObject(); + currentError_.SetNull(); + missingDependents_.SetNull(); + valid_ = true; + } + + //! Implementation of ISchemaValidator + void SetValidateFlags(unsigned flags) { + flags_ = flags; + } + virtual unsigned GetValidateFlags() const { + return flags_; + } + + virtual bool IsValid() const { + if (!valid_) return false; + if (GetContinueOnErrors() && !error_.ObjectEmpty()) return false; + return true; + } + //! End of Implementation of ISchemaValidator + + //! Gets the error object. + ValueType& GetError() { return error_; } + const ValueType& GetError() const { return error_; } + + //! Gets the JSON pointer pointed to the invalid schema. + // If reporting all errors, the stack will be empty. + PointerType GetInvalidSchemaPointer() const { + return schemaStack_.Empty() ? PointerType() : CurrentSchema().GetPointer(); + } + + //! Gets the keyword of invalid schema. + // If reporting all errors, the stack will be empty, so return "errors". + const Ch* GetInvalidSchemaKeyword() const { + if (!schemaStack_.Empty()) return CurrentContext().invalidKeyword; + if (GetContinueOnErrors() && !error_.ObjectEmpty()) return static_cast(GetErrorsString()); + return 0; + } + + //! Gets the error code of invalid schema. + // If reporting all errors, the stack will be empty, so return kValidateErrors. + ValidateErrorCode GetInvalidSchemaCode() const { + if (!schemaStack_.Empty()) return CurrentContext().invalidCode; + if (GetContinueOnErrors() && !error_.ObjectEmpty()) return kValidateErrors; + return kValidateErrorNone; + } + + //! Gets the JSON pointer pointed to the invalid value. + // If reporting all errors, the stack will be empty. + PointerType GetInvalidDocumentPointer() const { + if (documentStack_.Empty()) { + return PointerType(); + } + else { + return PointerType(documentStack_.template Bottom(), documentStack_.GetSize() / sizeof(Ch)); + } + } + + void NotMultipleOf(int64_t actual, const SValue& expected) { + AddNumberError(kValidateErrorMultipleOf, ValueType(actual).Move(), expected); + } + void NotMultipleOf(uint64_t actual, const SValue& expected) { + AddNumberError(kValidateErrorMultipleOf, ValueType(actual).Move(), expected); + } + void NotMultipleOf(double actual, const SValue& expected) { + AddNumberError(kValidateErrorMultipleOf, ValueType(actual).Move(), expected); + } + void AboveMaximum(int64_t actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMaximumString : 0); + } + void AboveMaximum(uint64_t actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMaximumString : 0); + } + void AboveMaximum(double actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMaximum : kValidateErrorMaximum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMaximumString : 0); + } + void BelowMinimum(int64_t actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMinimumString : 0); + } + void BelowMinimum(uint64_t actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMinimumString : 0); + } + void BelowMinimum(double actual, const SValue& expected, bool exclusive) { + AddNumberError(exclusive ? kValidateErrorExclusiveMinimum : kValidateErrorMinimum, ValueType(actual).Move(), expected, + exclusive ? &SchemaType::GetExclusiveMinimumString : 0); + } + + void TooLong(const Ch* str, SizeType length, SizeType expected) { + AddNumberError(kValidateErrorMaxLength, + ValueType(str, length, GetStateAllocator()).Move(), SValue(expected).Move()); + } + void TooShort(const Ch* str, SizeType length, SizeType expected) { + AddNumberError(kValidateErrorMinLength, + ValueType(str, length, GetStateAllocator()).Move(), SValue(expected).Move()); + } + void DoesNotMatch(const Ch* str, SizeType length) { + currentError_.SetObject(); + currentError_.AddMember(GetActualString(), ValueType(str, length, GetStateAllocator()).Move(), GetStateAllocator()); + AddCurrentError(kValidateErrorPattern); + } + + void DisallowedItem(SizeType index) { + currentError_.SetObject(); + currentError_.AddMember(GetDisallowedString(), ValueType(index).Move(), GetStateAllocator()); + AddCurrentError(kValidateErrorAdditionalItems, true); + } + void TooFewItems(SizeType actualCount, SizeType expectedCount) { + AddNumberError(kValidateErrorMinItems, + ValueType(actualCount).Move(), SValue(expectedCount).Move()); + } + void TooManyItems(SizeType actualCount, SizeType expectedCount) { + AddNumberError(kValidateErrorMaxItems, + ValueType(actualCount).Move(), SValue(expectedCount).Move()); + } + void DuplicateItems(SizeType index1, SizeType index2) { + ValueType duplicates(kArrayType); + duplicates.PushBack(index1, GetStateAllocator()); + duplicates.PushBack(index2, GetStateAllocator()); + currentError_.SetObject(); + currentError_.AddMember(GetDuplicatesString(), duplicates, GetStateAllocator()); + AddCurrentError(kValidateErrorUniqueItems, true); + } + + void TooManyProperties(SizeType actualCount, SizeType expectedCount) { + AddNumberError(kValidateErrorMaxProperties, + ValueType(actualCount).Move(), SValue(expectedCount).Move()); + } + void TooFewProperties(SizeType actualCount, SizeType expectedCount) { + AddNumberError(kValidateErrorMinProperties, + ValueType(actualCount).Move(), SValue(expectedCount).Move()); + } + void StartMissingProperties() { + currentError_.SetArray(); + } + void AddMissingProperty(const SValue& name) { + currentError_.PushBack(ValueType(name, GetStateAllocator()).Move(), GetStateAllocator()); + } + bool EndMissingProperties() { + if (currentError_.Empty()) + return false; + ValueType error(kObjectType); + error.AddMember(GetMissingString(), currentError_, GetStateAllocator()); + currentError_ = error; + AddCurrentError(kValidateErrorRequired); + return true; + } + void PropertyViolations(ISchemaValidator** subvalidators, SizeType count) { + for (SizeType i = 0; i < count; ++i) + MergeError(static_cast(subvalidators[i])->GetError()); + } + void DisallowedProperty(const Ch* name, SizeType length) { + currentError_.SetObject(); + currentError_.AddMember(GetDisallowedString(), ValueType(name, length, GetStateAllocator()).Move(), GetStateAllocator()); + AddCurrentError(kValidateErrorAdditionalProperties, true); + } + + void StartDependencyErrors() { + currentError_.SetObject(); + } + void StartMissingDependentProperties() { + missingDependents_.SetArray(); + } + void AddMissingDependentProperty(const SValue& targetName) { + missingDependents_.PushBack(ValueType(targetName, GetStateAllocator()).Move(), GetStateAllocator()); + } + void EndMissingDependentProperties(const SValue& sourceName) { + if (!missingDependents_.Empty()) { + // Create equivalent 'required' error + ValueType error(kObjectType); + ValidateErrorCode code = kValidateErrorRequired; + error.AddMember(GetMissingString(), missingDependents_.Move(), GetStateAllocator()); + AddErrorCode(error, code); + AddErrorInstanceLocation(error, false); + // When appending to a pointer ensure its allocator is used + PointerType schemaRef = GetInvalidSchemaPointer().Append(SchemaType::GetValidateErrorKeyword(kValidateErrorDependencies), &GetInvalidSchemaPointer().GetAllocator()); + AddErrorSchemaLocation(error, schemaRef.Append(sourceName.GetString(), sourceName.GetStringLength(), &GetInvalidSchemaPointer().GetAllocator())); + ValueType wrapper(kObjectType); + wrapper.AddMember(ValueType(SchemaType::GetValidateErrorKeyword(code), GetStateAllocator()).Move(), error, GetStateAllocator()); + currentError_.AddMember(ValueType(sourceName, GetStateAllocator()).Move(), wrapper, GetStateAllocator()); + } + } + void AddDependencySchemaError(const SValue& sourceName, ISchemaValidator* subvalidator) { + currentError_.AddMember(ValueType(sourceName, GetStateAllocator()).Move(), + static_cast(subvalidator)->GetError(), GetStateAllocator()); + } + bool EndDependencyErrors() { + if (currentError_.ObjectEmpty()) + return false; + ValueType error(kObjectType); + error.AddMember(GetErrorsString(), currentError_, GetStateAllocator()); + currentError_ = error; + AddCurrentError(kValidateErrorDependencies); + return true; + } + + void DisallowedValue(const ValidateErrorCode code = kValidateErrorEnum) { + currentError_.SetObject(); + AddCurrentError(code); + } + void StartDisallowedType() { + currentError_.SetArray(); + } + void AddExpectedType(const typename SchemaType::ValueType& expectedType) { + currentError_.PushBack(ValueType(expectedType, GetStateAllocator()).Move(), GetStateAllocator()); + } + void EndDisallowedType(const typename SchemaType::ValueType& actualType) { + ValueType error(kObjectType); + error.AddMember(GetExpectedString(), currentError_, GetStateAllocator()); + error.AddMember(GetActualString(), ValueType(actualType, GetStateAllocator()).Move(), GetStateAllocator()); + currentError_ = error; + AddCurrentError(kValidateErrorType); + } + void NotAllOf(ISchemaValidator** subvalidators, SizeType count) { + // Treat allOf like oneOf and anyOf to match https://rapidjson.org/md_doc_schema.html#allOf-anyOf-oneOf + AddErrorArray(kValidateErrorAllOf, subvalidators, count); + //for (SizeType i = 0; i < count; ++i) { + // MergeError(static_cast(subvalidators[i])->GetError()); + //} + } + void NoneOf(ISchemaValidator** subvalidators, SizeType count) { + AddErrorArray(kValidateErrorAnyOf, subvalidators, count); + } + void NotOneOf(ISchemaValidator** subvalidators, SizeType count) { + AddErrorArray(kValidateErrorOneOf, subvalidators, count); + } + void MultipleOneOf(SizeType index1, SizeType index2) { + ValueType matches(kArrayType); + matches.PushBack(index1, GetStateAllocator()); + matches.PushBack(index2, GetStateAllocator()); + currentError_.SetObject(); + currentError_.AddMember(GetMatchesString(), matches, GetStateAllocator()); + AddCurrentError(kValidateErrorOneOfMatch); + } + void Disallowed() { + currentError_.SetObject(); + AddCurrentError(kValidateErrorNot); + } + void DisallowedWhenWriting() { + currentError_.SetObject(); + AddCurrentError(kValidateErrorReadOnly); + } + void DisallowedWhenReading() { + currentError_.SetObject(); + AddCurrentError(kValidateErrorWriteOnly); + } + +#define RAPIDJSON_STRING_(name, ...) \ + static const StringRefType& Get##name##String() {\ + static const Ch s[] = { __VA_ARGS__, '\0' };\ + static const StringRefType v(s, static_cast(sizeof(s) / sizeof(Ch) - 1)); \ + return v;\ + } + + RAPIDJSON_STRING_(InstanceRef, 'i', 'n', 's', 't', 'a', 'n', 'c', 'e', 'R', 'e', 'f') + RAPIDJSON_STRING_(SchemaRef, 's', 'c', 'h', 'e', 'm', 'a', 'R', 'e', 'f') + RAPIDJSON_STRING_(Expected, 'e', 'x', 'p', 'e', 'c', 't', 'e', 'd') + RAPIDJSON_STRING_(Actual, 'a', 'c', 't', 'u', 'a', 'l') + RAPIDJSON_STRING_(Disallowed, 'd', 'i', 's', 'a', 'l', 'l', 'o', 'w', 'e', 'd') + RAPIDJSON_STRING_(Missing, 'm', 'i', 's', 's', 'i', 'n', 'g') + RAPIDJSON_STRING_(Errors, 'e', 'r', 'r', 'o', 'r', 's') + RAPIDJSON_STRING_(ErrorCode, 'e', 'r', 'r', 'o', 'r', 'C', 'o', 'd', 'e') + RAPIDJSON_STRING_(ErrorMessage, 'e', 'r', 'r', 'o', 'r', 'M', 'e', 's', 's', 'a', 'g', 'e') + RAPIDJSON_STRING_(Duplicates, 'd', 'u', 'p', 'l', 'i', 'c', 'a', 't', 'e', 's') + RAPIDJSON_STRING_(Matches, 'm', 'a', 't', 'c', 'h', 'e', 's') + +#undef RAPIDJSON_STRING_ + +#define RAPIDJSON_SCHEMA_HANDLE_BEGIN_(method, arg1)\ + if (!valid_) return false; \ + if ((!BeginValue() && !GetContinueOnErrors()) || (!CurrentSchema().method arg1 && !GetContinueOnErrors())) {\ + *documentStack_.template Push() = '\0';\ + documentStack_.template Pop(1);\ + RAPIDJSON_SCHEMA_PRINT(InvalidDocument, documentStack_.template Bottom());\ + valid_ = false;\ + return valid_;\ + } + +#define RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(method, arg2)\ + for (Context* context = schemaStack_.template Bottom(); context != schemaStack_.template End(); context++) {\ + if (context->hasher)\ + static_cast(context->hasher)->method arg2;\ + if (context->validators)\ + for (SizeType i_ = 0; i_ < context->validatorCount; i_++)\ + static_cast(context->validators[i_])->method arg2;\ + if (context->patternPropertiesValidators)\ + for (SizeType i_ = 0; i_ < context->patternPropertiesValidatorCount; i_++)\ + static_cast(context->patternPropertiesValidators[i_])->method arg2;\ + } + +#define RAPIDJSON_SCHEMA_HANDLE_END_(method, arg2)\ + valid_ = (EndValue() || GetContinueOnErrors()) && (!outputHandler_ || outputHandler_->method arg2);\ + return valid_; + +#define RAPIDJSON_SCHEMA_HANDLE_VALUE_(method, arg1, arg2) \ + RAPIDJSON_SCHEMA_HANDLE_BEGIN_ (method, arg1);\ + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(method, arg2);\ + RAPIDJSON_SCHEMA_HANDLE_END_ (method, arg2) + + bool Null() { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Null, (CurrentContext()), ( )); } + bool Bool(bool b) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Bool, (CurrentContext(), b), (b)); } + bool Int(int i) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Int, (CurrentContext(), i), (i)); } + bool Uint(unsigned u) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Uint, (CurrentContext(), u), (u)); } + bool Int64(int64_t i) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Int64, (CurrentContext(), i), (i)); } + bool Uint64(uint64_t u) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Uint64, (CurrentContext(), u), (u)); } + bool Double(double d) { RAPIDJSON_SCHEMA_HANDLE_VALUE_(Double, (CurrentContext(), d), (d)); } + bool RawNumber(const Ch* str, SizeType length, bool copy) + { RAPIDJSON_SCHEMA_HANDLE_VALUE_(String, (CurrentContext(), str, length, copy), (str, length, copy)); } + bool String(const Ch* str, SizeType length, bool copy) + { RAPIDJSON_SCHEMA_HANDLE_VALUE_(String, (CurrentContext(), str, length, copy), (str, length, copy)); } + + bool StartObject() { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::StartObject"); + RAPIDJSON_SCHEMA_HANDLE_BEGIN_(StartObject, (CurrentContext())); + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(StartObject, ()); + valid_ = !outputHandler_ || outputHandler_->StartObject(); + return valid_; + } + + bool Key(const Ch* str, SizeType len, bool copy) { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::Key", str); + if (!valid_) return false; + AppendToken(str, len); + if (!CurrentSchema().Key(CurrentContext(), str, len, copy) && !GetContinueOnErrors()) { + valid_ = false; + return valid_; + } + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(Key, (str, len, copy)); + valid_ = !outputHandler_ || outputHandler_->Key(str, len, copy); + return valid_; + } + + bool EndObject(SizeType memberCount) { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::EndObject"); + if (!valid_) return false; + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(EndObject, (memberCount)); + if (!CurrentSchema().EndObject(CurrentContext(), memberCount) && !GetContinueOnErrors()) { + valid_ = false; + return valid_; + } + RAPIDJSON_SCHEMA_HANDLE_END_(EndObject, (memberCount)); + } + + bool StartArray() { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::StartArray"); + RAPIDJSON_SCHEMA_HANDLE_BEGIN_(StartArray, (CurrentContext())); + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(StartArray, ()); + valid_ = !outputHandler_ || outputHandler_->StartArray(); + return valid_; + } + + bool EndArray(SizeType elementCount) { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::EndArray"); + if (!valid_) return false; + RAPIDJSON_SCHEMA_HANDLE_PARALLEL_(EndArray, (elementCount)); + if (!CurrentSchema().EndArray(CurrentContext(), elementCount) && !GetContinueOnErrors()) { + valid_ = false; + return valid_; + } + RAPIDJSON_SCHEMA_HANDLE_END_(EndArray, (elementCount)); + } + +#undef RAPIDJSON_SCHEMA_HANDLE_BEGIN_ +#undef RAPIDJSON_SCHEMA_HANDLE_PARALLEL_ +#undef RAPIDJSON_SCHEMA_HANDLE_VALUE_ + + // Implementation of ISchemaStateFactory + virtual ISchemaValidator* CreateSchemaValidator(const SchemaType& root, const bool inheritContinueOnErrors) { + *documentStack_.template Push() = '\0'; + documentStack_.template Pop(1); + ISchemaValidator* sv = new (GetStateAllocator().Malloc(sizeof(GenericSchemaValidator))) GenericSchemaValidator(*schemaDocument_, root, documentStack_.template Bottom(), documentStack_.GetSize(), + depth_ + 1, + &GetStateAllocator()); + sv->SetValidateFlags(inheritContinueOnErrors ? GetValidateFlags() : GetValidateFlags() & ~static_cast(kValidateContinueOnErrorFlag)); + return sv; + } + + virtual void DestroySchemaValidator(ISchemaValidator* validator) { + GenericSchemaValidator* v = static_cast(validator); + v->~GenericSchemaValidator(); + StateAllocator::Free(v); + } + + virtual void* CreateHasher() { + return new (GetStateAllocator().Malloc(sizeof(HasherType))) HasherType(&GetStateAllocator()); + } + + virtual uint64_t GetHashCode(void* hasher) { + return static_cast(hasher)->GetHashCode(); + } + + virtual void DestroryHasher(void* hasher) { + HasherType* h = static_cast(hasher); + h->~HasherType(); + StateAllocator::Free(h); + } + + virtual void* MallocState(size_t size) { + return GetStateAllocator().Malloc(size); + } + + virtual void FreeState(void* p) { + StateAllocator::Free(p); + } + // End of implementation of ISchemaStateFactory + +private: + typedef typename SchemaType::Context Context; + typedef GenericValue, StateAllocator> HashCodeArray; + typedef internal::Hasher HasherType; + + GenericSchemaValidator( + const SchemaDocumentType& schemaDocument, + const SchemaType& root, + const char* basePath, size_t basePathSize, + unsigned depth, + StateAllocator* allocator = 0, + size_t schemaStackCapacity = kDefaultSchemaStackCapacity, + size_t documentStackCapacity = kDefaultDocumentStackCapacity) + : + schemaDocument_(&schemaDocument), + root_(root), + stateAllocator_(allocator), + ownStateAllocator_(0), + schemaStack_(allocator, schemaStackCapacity), + documentStack_(allocator, documentStackCapacity), + outputHandler_(0), + error_(kObjectType), + currentError_(), + missingDependents_(), + valid_(true), + flags_(kValidateDefaultFlags), + depth_(depth) + { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::GenericSchemaValidator (internal)", basePath && basePathSize ? basePath : ""); + if (basePath && basePathSize) + memcpy(documentStack_.template Push(basePathSize), basePath, basePathSize); + } + + StateAllocator& GetStateAllocator() { + if (!stateAllocator_) + stateAllocator_ = ownStateAllocator_ = RAPIDJSON_NEW(StateAllocator)(); + return *stateAllocator_; + } + + bool GetContinueOnErrors() const { + return flags_ & kValidateContinueOnErrorFlag; + } + + bool BeginValue() { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::BeginValue"); + if (schemaStack_.Empty()) + PushSchema(root_); + else { + if (CurrentContext().inArray) + internal::TokenHelper, Ch>::AppendIndexToken(documentStack_, CurrentContext().arrayElementIndex); + + if (!CurrentSchema().BeginValue(CurrentContext()) && !GetContinueOnErrors()) + return false; + + SizeType count = CurrentContext().patternPropertiesSchemaCount; + const SchemaType** sa = CurrentContext().patternPropertiesSchemas; + typename Context::PatternValidatorType patternValidatorType = CurrentContext().valuePatternValidatorType; + bool valueUniqueness = CurrentContext().valueUniqueness; + RAPIDJSON_ASSERT(CurrentContext().valueSchema); + PushSchema(*CurrentContext().valueSchema); + + if (count > 0) { + CurrentContext().objectPatternValidatorType = patternValidatorType; + ISchemaValidator**& va = CurrentContext().patternPropertiesValidators; + SizeType& validatorCount = CurrentContext().patternPropertiesValidatorCount; + va = static_cast(MallocState(sizeof(ISchemaValidator*) * count)); + std::memset(va, 0, sizeof(ISchemaValidator*) * count); + for (SizeType i = 0; i < count; i++) + va[validatorCount++] = CreateSchemaValidator(*sa[i], true); // Inherit continueOnError + } + + CurrentContext().arrayUniqueness = valueUniqueness; + } + return true; + } + + bool EndValue() { + RAPIDJSON_SCHEMA_PRINT(Method, "GenericSchemaValidator::EndValue"); + if (!CurrentSchema().EndValue(CurrentContext()) && !GetContinueOnErrors()) + return false; + + GenericStringBuffer sb; + schemaDocument_->GetPointer(&CurrentSchema()).StringifyUriFragment(sb); + *documentStack_.template Push() = '\0'; + documentStack_.template Pop(1); + RAPIDJSON_SCHEMA_PRINT(ValidatorPointers, sb.GetString(), documentStack_.template Bottom(), depth_); + void* hasher = CurrentContext().hasher; + uint64_t h = hasher && CurrentContext().arrayUniqueness ? static_cast(hasher)->GetHashCode() : 0; + + PopSchema(); + + if (!schemaStack_.Empty()) { + Context& context = CurrentContext(); + // Only check uniqueness if there is a hasher + if (hasher && context.valueUniqueness) { + HashCodeArray* a = static_cast(context.arrayElementHashCodes); + if (!a) + CurrentContext().arrayElementHashCodes = a = new (GetStateAllocator().Malloc(sizeof(HashCodeArray))) HashCodeArray(kArrayType); + for (typename HashCodeArray::ConstValueIterator itr = a->Begin(); itr != a->End(); ++itr) + if (itr->GetUint64() == h) { + DuplicateItems(static_cast(itr - a->Begin()), a->Size()); + // Cleanup before returning if continuing + if (GetContinueOnErrors()) { + a->PushBack(h, GetStateAllocator()); + while (!documentStack_.Empty() && *documentStack_.template Pop(1) != '/'); + } + RAPIDJSON_INVALID_KEYWORD_RETURN(kValidateErrorUniqueItems); + } + a->PushBack(h, GetStateAllocator()); + } + } + + // Remove the last token of document pointer + while (!documentStack_.Empty() && *documentStack_.template Pop(1) != '/') + ; + + return true; + } + + void AppendToken(const Ch* str, SizeType len) { + documentStack_.template Reserve(1 + len * 2); // worst case all characters are escaped as two characters + *documentStack_.template PushUnsafe() = '/'; + for (SizeType i = 0; i < len; i++) { + if (str[i] == '~') { + *documentStack_.template PushUnsafe() = '~'; + *documentStack_.template PushUnsafe() = '0'; + } + else if (str[i] == '/') { + *documentStack_.template PushUnsafe() = '~'; + *documentStack_.template PushUnsafe() = '1'; + } + else + *documentStack_.template PushUnsafe() = str[i]; + } + } + + RAPIDJSON_FORCEINLINE void PushSchema(const SchemaType& schema) { new (schemaStack_.template Push()) Context(*this, *this, &schema, flags_); } + + RAPIDJSON_FORCEINLINE void PopSchema() { + Context* c = schemaStack_.template Pop(1); + if (HashCodeArray* a = static_cast(c->arrayElementHashCodes)) { + a->~HashCodeArray(); + StateAllocator::Free(a); + } + c->~Context(); + } + + void AddErrorInstanceLocation(ValueType& result, bool parent) { + GenericStringBuffer sb; + PointerType instancePointer = GetInvalidDocumentPointer(); + ((parent && instancePointer.GetTokenCount() > 0) + ? PointerType(instancePointer.GetTokens(), instancePointer.GetTokenCount() - 1) + : instancePointer).StringifyUriFragment(sb); + ValueType instanceRef(sb.GetString(), static_cast(sb.GetSize() / sizeof(Ch)), + GetStateAllocator()); + result.AddMember(GetInstanceRefString(), instanceRef, GetStateAllocator()); + } + + void AddErrorSchemaLocation(ValueType& result, PointerType schema = PointerType()) { + GenericStringBuffer sb; + SizeType len = CurrentSchema().GetURI().GetStringLength(); + if (len) memcpy(sb.Push(len), CurrentSchema().GetURI().GetString(), len * sizeof(Ch)); + if (schema.GetTokenCount()) schema.StringifyUriFragment(sb); + else GetInvalidSchemaPointer().StringifyUriFragment(sb); + ValueType schemaRef(sb.GetString(), static_cast(sb.GetSize() / sizeof(Ch)), + GetStateAllocator()); + result.AddMember(GetSchemaRefString(), schemaRef, GetStateAllocator()); + } + + void AddErrorCode(ValueType& result, const ValidateErrorCode code) { + result.AddMember(GetErrorCodeString(), code, GetStateAllocator()); + } + + void AddError(ValueType& keyword, ValueType& error) { + typename ValueType::MemberIterator member = error_.FindMember(keyword); + if (member == error_.MemberEnd()) + error_.AddMember(keyword, error, GetStateAllocator()); + else { + if (member->value.IsObject()) { + ValueType errors(kArrayType); + errors.PushBack(member->value, GetStateAllocator()); + member->value = errors; + } + member->value.PushBack(error, GetStateAllocator()); + } + } + + void AddCurrentError(const ValidateErrorCode code, bool parent = false) { + AddErrorCode(currentError_, code); + AddErrorInstanceLocation(currentError_, parent); + AddErrorSchemaLocation(currentError_); + AddError(ValueType(SchemaType::GetValidateErrorKeyword(code), GetStateAllocator(), false).Move(), currentError_); + } + + void MergeError(ValueType& other) { + for (typename ValueType::MemberIterator it = other.MemberBegin(), end = other.MemberEnd(); it != end; ++it) { + AddError(it->name, it->value); + } + } + + void AddNumberError(const ValidateErrorCode code, ValueType& actual, const SValue& expected, + const typename SchemaType::ValueType& (*exclusive)() = 0) { + currentError_.SetObject(); + currentError_.AddMember(GetActualString(), actual, GetStateAllocator()); + currentError_.AddMember(GetExpectedString(), ValueType(expected, GetStateAllocator()).Move(), GetStateAllocator()); + if (exclusive) + currentError_.AddMember(ValueType(exclusive(), GetStateAllocator()).Move(), true, GetStateAllocator()); + AddCurrentError(code); + } + + void AddErrorArray(const ValidateErrorCode code, + ISchemaValidator** subvalidators, SizeType count) { + ValueType errors(kArrayType); + for (SizeType i = 0; i < count; ++i) + errors.PushBack(static_cast(subvalidators[i])->GetError(), GetStateAllocator()); + currentError_.SetObject(); + currentError_.AddMember(GetErrorsString(), errors, GetStateAllocator()); + AddCurrentError(code); + } + + const SchemaType& CurrentSchema() const { return *schemaStack_.template Top()->schema; } + Context& CurrentContext() { return *schemaStack_.template Top(); } + const Context& CurrentContext() const { return *schemaStack_.template Top(); } + + static const size_t kDefaultSchemaStackCapacity = 1024; + static const size_t kDefaultDocumentStackCapacity = 256; + const SchemaDocumentType* schemaDocument_; + const SchemaType& root_; + StateAllocator* stateAllocator_; + StateAllocator* ownStateAllocator_; + internal::Stack schemaStack_; //!< stack to store the current path of schema (BaseSchemaType *) + internal::Stack documentStack_; //!< stack to store the current path of validating document (Ch) + OutputHandler* outputHandler_; + ValueType error_; + ValueType currentError_; + ValueType missingDependents_; + bool valid_; + unsigned flags_; + unsigned depth_; +}; + +typedef GenericSchemaValidator SchemaValidator; + +/////////////////////////////////////////////////////////////////////////////// +// SchemaValidatingReader + +//! A helper class for parsing with validation. +/*! + This helper class is a functor, designed as a parameter of \ref GenericDocument::Populate(). + + \tparam parseFlags Combination of \ref ParseFlag. + \tparam InputStream Type of input stream, implementing Stream concept. + \tparam SourceEncoding Encoding of the input stream. + \tparam SchemaDocumentType Type of schema document. + \tparam StackAllocator Allocator type for stack. +*/ +template < + unsigned parseFlags, + typename InputStream, + typename SourceEncoding, + typename SchemaDocumentType = SchemaDocument, + typename StackAllocator = CrtAllocator> +class SchemaValidatingReader { +public: + typedef typename SchemaDocumentType::PointerType PointerType; + typedef typename InputStream::Ch Ch; + typedef GenericValue ValueType; + + //! Constructor + /*! + \param is Input stream. + \param sd Schema document. + */ + SchemaValidatingReader(InputStream& is, const SchemaDocumentType& sd) : is_(is), sd_(sd), invalidSchemaKeyword_(), invalidSchemaCode_(kValidateErrorNone), error_(kObjectType), isValid_(true) {} + + template + bool operator()(Handler& handler) { + GenericReader reader; + GenericSchemaValidator validator(sd_, handler); + parseResult_ = reader.template Parse(is_, validator); + + isValid_ = validator.IsValid(); + if (isValid_) { + invalidSchemaPointer_ = PointerType(); + invalidSchemaKeyword_ = 0; + invalidDocumentPointer_ = PointerType(); + error_.SetObject(); + } + else { + invalidSchemaPointer_ = validator.GetInvalidSchemaPointer(); + invalidSchemaKeyword_ = validator.GetInvalidSchemaKeyword(); + invalidSchemaCode_ = validator.GetInvalidSchemaCode(); + invalidDocumentPointer_ = validator.GetInvalidDocumentPointer(); + error_.CopyFrom(validator.GetError(), allocator_); + } + + return parseResult_; + } + + const ParseResult& GetParseResult() const { return parseResult_; } + bool IsValid() const { return isValid_; } + const PointerType& GetInvalidSchemaPointer() const { return invalidSchemaPointer_; } + const Ch* GetInvalidSchemaKeyword() const { return invalidSchemaKeyword_; } + const PointerType& GetInvalidDocumentPointer() const { return invalidDocumentPointer_; } + const ValueType& GetError() const { return error_; } + ValidateErrorCode GetInvalidSchemaCode() const { return invalidSchemaCode_; } + +private: + InputStream& is_; + const SchemaDocumentType& sd_; + + ParseResult parseResult_; + PointerType invalidSchemaPointer_; + const Ch* invalidSchemaKeyword_; + PointerType invalidDocumentPointer_; + ValidateErrorCode invalidSchemaCode_; + StackAllocator allocator_; + ValueType error_; + bool isValid_; +}; + +RAPIDJSON_NAMESPACE_END +RAPIDJSON_DIAG_POP + +#endif // RAPIDJSON_SCHEMA_H_ diff --git a/include/rapidjson/stream.h b/include/rapidjson/stream.h new file mode 100644 index 0000000..1fd7091 --- /dev/null +++ b/include/rapidjson/stream.h @@ -0,0 +1,223 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#include "rapidjson.h" + +#ifndef RAPIDJSON_STREAM_H_ +#define RAPIDJSON_STREAM_H_ + +#include "encodings.h" + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// Stream + +/*! \class rapidjson::Stream + \brief Concept for reading and writing characters. + + For read-only stream, no need to implement PutBegin(), Put(), Flush() and PutEnd(). + + For write-only stream, only need to implement Put() and Flush(). + +\code +concept Stream { + typename Ch; //!< Character type of the stream. + + //! Read the current character from stream without moving the read cursor. + Ch Peek() const; + + //! Read the current character from stream and moving the read cursor to next character. + Ch Take(); + + //! Get the current read cursor. + //! \return Number of characters read from start. + size_t Tell(); + + //! Begin writing operation at the current read pointer. + //! \return The begin writer pointer. + Ch* PutBegin(); + + //! Write a character. + void Put(Ch c); + + //! Flush the buffer. + void Flush(); + + //! End the writing operation. + //! \param begin The begin write pointer returned by PutBegin(). + //! \return Number of characters written. + size_t PutEnd(Ch* begin); +} +\endcode +*/ + +//! Provides additional information for stream. +/*! + By using traits pattern, this type provides a default configuration for stream. + For custom stream, this type can be specialized for other configuration. + See TEST(Reader, CustomStringStream) in readertest.cpp for example. +*/ +template +struct StreamTraits { + //! Whether to make local copy of stream for optimization during parsing. + /*! + By default, for safety, streams do not use local copy optimization. + Stream that can be copied fast should specialize this, like StreamTraits. + */ + enum { copyOptimization = 0 }; +}; + +//! Reserve n characters for writing to a stream. +template +inline void PutReserve(Stream& stream, size_t count) { + (void)stream; + (void)count; +} + +//! Write character to a stream, presuming buffer is reserved. +template +inline void PutUnsafe(Stream& stream, typename Stream::Ch c) { + stream.Put(c); +} + +//! Put N copies of a character to a stream. +template +inline void PutN(Stream& stream, Ch c, size_t n) { + PutReserve(stream, n); + for (size_t i = 0; i < n; i++) + PutUnsafe(stream, c); +} + +/////////////////////////////////////////////////////////////////////////////// +// GenericStreamWrapper + +//! A Stream Wrapper +/*! \tThis string stream is a wrapper for any stream by just forwarding any + \treceived message to the origin stream. + \note implements Stream concept +*/ + +#if defined(_MSC_VER) && _MSC_VER <= 1800 +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4702) // unreachable code +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +template > +class GenericStreamWrapper { +public: + typedef typename Encoding::Ch Ch; + GenericStreamWrapper(InputStream& is): is_(is) {} + + Ch Peek() const { return is_.Peek(); } + Ch Take() { return is_.Take(); } + size_t Tell() { return is_.Tell(); } + Ch* PutBegin() { return is_.PutBegin(); } + void Put(Ch ch) { is_.Put(ch); } + void Flush() { is_.Flush(); } + size_t PutEnd(Ch* ch) { return is_.PutEnd(ch); } + + // wrapper for MemoryStream + const Ch* Peek4() const { return is_.Peek4(); } + + // wrapper for AutoUTFInputStream + UTFType GetType() const { return is_.GetType(); } + bool HasBOM() const { return is_.HasBOM(); } + +protected: + InputStream& is_; +}; + +#if defined(_MSC_VER) && _MSC_VER <= 1800 +RAPIDJSON_DIAG_POP +#endif + +/////////////////////////////////////////////////////////////////////////////// +// StringStream + +//! Read-only string stream. +/*! \note implements Stream concept +*/ +template +struct GenericStringStream { + typedef typename Encoding::Ch Ch; + + GenericStringStream(const Ch *src) : src_(src), head_(src) {} + + Ch Peek() const { return *src_; } + Ch Take() { return *src_++; } + size_t Tell() const { return static_cast(src_ - head_); } + + Ch* PutBegin() { RAPIDJSON_ASSERT(false); return 0; } + void Put(Ch) { RAPIDJSON_ASSERT(false); } + void Flush() { RAPIDJSON_ASSERT(false); } + size_t PutEnd(Ch*) { RAPIDJSON_ASSERT(false); return 0; } + + const Ch* src_; //!< Current read position. + const Ch* head_; //!< Original head of the string. +}; + +template +struct StreamTraits > { + enum { copyOptimization = 1 }; +}; + +//! String stream with UTF8 encoding. +typedef GenericStringStream > StringStream; + +/////////////////////////////////////////////////////////////////////////////// +// InsituStringStream + +//! A read-write string stream. +/*! This string stream is particularly designed for in-situ parsing. + \note implements Stream concept +*/ +template +struct GenericInsituStringStream { + typedef typename Encoding::Ch Ch; + + GenericInsituStringStream(Ch *src) : src_(src), dst_(0), head_(src) {} + + // Read + Ch Peek() { return *src_; } + Ch Take() { return *src_++; } + size_t Tell() { return static_cast(src_ - head_); } + + // Write + void Put(Ch c) { RAPIDJSON_ASSERT(dst_ != 0); *dst_++ = c; } + + Ch* PutBegin() { return dst_ = src_; } + size_t PutEnd(Ch* begin) { return static_cast(dst_ - begin); } + void Flush() {} + + Ch* Push(size_t count) { Ch* begin = dst_; dst_ += count; return begin; } + void Pop(size_t count) { dst_ -= count; } + + Ch* src_; + Ch* dst_; + Ch* head_; +}; + +template +struct StreamTraits > { + enum { copyOptimization = 1 }; +}; + +//! Insitu string stream with UTF8 encoding. +typedef GenericInsituStringStream > InsituStringStream; + +RAPIDJSON_NAMESPACE_END + +#endif // RAPIDJSON_STREAM_H_ diff --git a/include/rapidjson/stringbuffer.h b/include/rapidjson/stringbuffer.h new file mode 100644 index 0000000..82ad3ca --- /dev/null +++ b/include/rapidjson/stringbuffer.h @@ -0,0 +1,121 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_STRINGBUFFER_H_ +#define RAPIDJSON_STRINGBUFFER_H_ + +#include "stream.h" +#include "internal/stack.h" + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS +#include // std::move +#endif + +#include "internal/stack.h" + +#if defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(c++98-compat) +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +//! Represents an in-memory output stream. +/*! + \tparam Encoding Encoding of the stream. + \tparam Allocator type for allocating memory buffer. + \note implements Stream concept +*/ +template +class GenericStringBuffer { +public: + typedef typename Encoding::Ch Ch; + + GenericStringBuffer(Allocator* allocator = 0, size_t capacity = kDefaultCapacity) : stack_(allocator, capacity) {} + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + GenericStringBuffer(GenericStringBuffer&& rhs) : stack_(std::move(rhs.stack_)) {} + GenericStringBuffer& operator=(GenericStringBuffer&& rhs) { + if (&rhs != this) + stack_ = std::move(rhs.stack_); + return *this; + } +#endif + + void Put(Ch c) { *stack_.template Push() = c; } + void PutUnsafe(Ch c) { *stack_.template PushUnsafe() = c; } + void Flush() {} + + void Clear() { stack_.Clear(); } + void ShrinkToFit() { + // Push and pop a null terminator. This is safe. + *stack_.template Push() = '\0'; + stack_.ShrinkToFit(); + stack_.template Pop(1); + } + + void Reserve(size_t count) { stack_.template Reserve(count); } + Ch* Push(size_t count) { return stack_.template Push(count); } + Ch* PushUnsafe(size_t count) { return stack_.template PushUnsafe(count); } + void Pop(size_t count) { stack_.template Pop(count); } + + const Ch* GetString() const { + // Push and pop a null terminator. This is safe. + *stack_.template Push() = '\0'; + stack_.template Pop(1); + + return stack_.template Bottom(); + } + + //! Get the size of string in bytes in the string buffer. + size_t GetSize() const { return stack_.GetSize(); } + + //! Get the length of string in Ch in the string buffer. + size_t GetLength() const { return stack_.GetSize() / sizeof(Ch); } + + static const size_t kDefaultCapacity = 256; + mutable internal::Stack stack_; + +private: + // Prohibit copy constructor & assignment operator. + GenericStringBuffer(const GenericStringBuffer&); + GenericStringBuffer& operator=(const GenericStringBuffer&); +}; + +//! String buffer with UTF8 encoding +typedef GenericStringBuffer > StringBuffer; + +template +inline void PutReserve(GenericStringBuffer& stream, size_t count) { + stream.Reserve(count); +} + +template +inline void PutUnsafe(GenericStringBuffer& stream, typename Encoding::Ch c) { + stream.PutUnsafe(c); +} + +//! Implement specialized version of PutN() with memset() for better performance. +template<> +inline void PutN(GenericStringBuffer >& stream, char c, size_t n) { + std::memset(stream.stack_.Push(n), c, n * sizeof(c)); +} + +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_STRINGBUFFER_H_ diff --git a/include/rapidjson/uri.h b/include/rapidjson/uri.h new file mode 100644 index 0000000..f93e508 --- /dev/null +++ b/include/rapidjson/uri.h @@ -0,0 +1,481 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// (C) Copyright IBM Corporation 2021 +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_URI_H_ +#define RAPIDJSON_URI_H_ + +#include "internal/strfunc.h" + +#if defined(__clang__) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(c++98-compat) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_OFF(4512) // assignment operator could not be generated +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// GenericUri + +template +class GenericUri { +public: + typedef typename ValueType::Ch Ch; +#if RAPIDJSON_HAS_STDSTRING + typedef std::basic_string String; +#endif + + //! Constructors + GenericUri(Allocator* allocator = 0) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + } + + GenericUri(const Ch* uri, SizeType len, Allocator* allocator = 0) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + Parse(uri, len); + } + + GenericUri(const Ch* uri, Allocator* allocator = 0) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + Parse(uri, internal::StrLen(uri)); + } + + // Use with specializations of GenericValue + template GenericUri(const T& uri, Allocator* allocator = 0) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + const Ch* u = uri.template Get(); // TypeHelper from document.h + Parse(u, internal::StrLen(u)); + } + +#if RAPIDJSON_HAS_STDSTRING + GenericUri(const String& uri, Allocator* allocator = 0) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + Parse(uri.c_str(), internal::StrLen(uri.c_str())); + } +#endif + + //! Copy constructor + GenericUri(const GenericUri& rhs) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(), ownAllocator_() { + *this = rhs; + } + + //! Copy constructor + GenericUri(const GenericUri& rhs, Allocator* allocator) : uri_(), base_(), scheme_(), auth_(), path_(), query_(), frag_(), allocator_(allocator), ownAllocator_() { + *this = rhs; + } + + //! Destructor. + ~GenericUri() { + Free(); + RAPIDJSON_DELETE(ownAllocator_); + } + + //! Assignment operator + GenericUri& operator=(const GenericUri& rhs) { + if (this != &rhs) { + // Do not delete ownAllocator + Free(); + Allocate(rhs.GetStringLength()); + auth_ = CopyPart(scheme_, rhs.scheme_, rhs.GetSchemeStringLength()); + path_ = CopyPart(auth_, rhs.auth_, rhs.GetAuthStringLength()); + query_ = CopyPart(path_, rhs.path_, rhs.GetPathStringLength()); + frag_ = CopyPart(query_, rhs.query_, rhs.GetQueryStringLength()); + base_ = CopyPart(frag_, rhs.frag_, rhs.GetFragStringLength()); + uri_ = CopyPart(base_, rhs.base_, rhs.GetBaseStringLength()); + CopyPart(uri_, rhs.uri_, rhs.GetStringLength()); + } + return *this; + } + + //! Getters + // Use with specializations of GenericValue + template void Get(T& uri, Allocator& allocator) { + uri.template Set(this->GetString(), allocator); // TypeHelper from document.h + } + + const Ch* GetString() const { return uri_; } + SizeType GetStringLength() const { return uri_ == 0 ? 0 : internal::StrLen(uri_); } + const Ch* GetBaseString() const { return base_; } + SizeType GetBaseStringLength() const { return base_ == 0 ? 0 : internal::StrLen(base_); } + const Ch* GetSchemeString() const { return scheme_; } + SizeType GetSchemeStringLength() const { return scheme_ == 0 ? 0 : internal::StrLen(scheme_); } + const Ch* GetAuthString() const { return auth_; } + SizeType GetAuthStringLength() const { return auth_ == 0 ? 0 : internal::StrLen(auth_); } + const Ch* GetPathString() const { return path_; } + SizeType GetPathStringLength() const { return path_ == 0 ? 0 : internal::StrLen(path_); } + const Ch* GetQueryString() const { return query_; } + SizeType GetQueryStringLength() const { return query_ == 0 ? 0 : internal::StrLen(query_); } + const Ch* GetFragString() const { return frag_; } + SizeType GetFragStringLength() const { return frag_ == 0 ? 0 : internal::StrLen(frag_); } + +#if RAPIDJSON_HAS_STDSTRING + static String Get(const GenericUri& uri) { return String(uri.GetString(), uri.GetStringLength()); } + static String GetBase(const GenericUri& uri) { return String(uri.GetBaseString(), uri.GetBaseStringLength()); } + static String GetScheme(const GenericUri& uri) { return String(uri.GetSchemeString(), uri.GetSchemeStringLength()); } + static String GetAuth(const GenericUri& uri) { return String(uri.GetAuthString(), uri.GetAuthStringLength()); } + static String GetPath(const GenericUri& uri) { return String(uri.GetPathString(), uri.GetPathStringLength()); } + static String GetQuery(const GenericUri& uri) { return String(uri.GetQueryString(), uri.GetQueryStringLength()); } + static String GetFrag(const GenericUri& uri) { return String(uri.GetFragString(), uri.GetFragStringLength()); } +#endif + + //! Equality operators + bool operator==(const GenericUri& rhs) const { + return Match(rhs, true); + } + + bool operator!=(const GenericUri& rhs) const { + return !Match(rhs, true); + } + + bool Match(const GenericUri& uri, bool full = true) const { + Ch* s1; + Ch* s2; + if (full) { + s1 = uri_; + s2 = uri.uri_; + } else { + s1 = base_; + s2 = uri.base_; + } + if (s1 == s2) return true; + if (s1 == 0 || s2 == 0) return false; + return internal::StrCmp(s1, s2) == 0; + } + + //! Resolve this URI against another (base) URI in accordance with URI resolution rules. + // See https://tools.ietf.org/html/rfc3986 + // Use for resolving an id or $ref with an in-scope id. + // Returns a new GenericUri for the resolved URI. + GenericUri Resolve(const GenericUri& baseuri, Allocator* allocator = 0) { + GenericUri resuri; + resuri.allocator_ = allocator; + // Ensure enough space for combining paths + resuri.Allocate(GetStringLength() + baseuri.GetStringLength() + 1); // + 1 for joining slash + + if (!(GetSchemeStringLength() == 0)) { + // Use all of this URI + resuri.auth_ = CopyPart(resuri.scheme_, scheme_, GetSchemeStringLength()); + resuri.path_ = CopyPart(resuri.auth_, auth_, GetAuthStringLength()); + resuri.query_ = CopyPart(resuri.path_, path_, GetPathStringLength()); + resuri.frag_ = CopyPart(resuri.query_, query_, GetQueryStringLength()); + resuri.RemoveDotSegments(); + } else { + // Use the base scheme + resuri.auth_ = CopyPart(resuri.scheme_, baseuri.scheme_, baseuri.GetSchemeStringLength()); + if (!(GetAuthStringLength() == 0)) { + // Use this auth, path, query + resuri.path_ = CopyPart(resuri.auth_, auth_, GetAuthStringLength()); + resuri.query_ = CopyPart(resuri.path_, path_, GetPathStringLength()); + resuri.frag_ = CopyPart(resuri.query_, query_, GetQueryStringLength()); + resuri.RemoveDotSegments(); + } else { + // Use the base auth + resuri.path_ = CopyPart(resuri.auth_, baseuri.auth_, baseuri.GetAuthStringLength()); + if (GetPathStringLength() == 0) { + // Use the base path + resuri.query_ = CopyPart(resuri.path_, baseuri.path_, baseuri.GetPathStringLength()); + if (GetQueryStringLength() == 0) { + // Use the base query + resuri.frag_ = CopyPart(resuri.query_, baseuri.query_, baseuri.GetQueryStringLength()); + } else { + // Use this query + resuri.frag_ = CopyPart(resuri.query_, query_, GetQueryStringLength()); + } + } else { + if (path_[0] == '/') { + // Absolute path - use all of this path + resuri.query_ = CopyPart(resuri.path_, path_, GetPathStringLength()); + resuri.RemoveDotSegments(); + } else { + // Relative path - append this path to base path after base path's last slash + size_t pos = 0; + if (!(baseuri.GetAuthStringLength() == 0) && baseuri.GetPathStringLength() == 0) { + resuri.path_[pos] = '/'; + pos++; + } + size_t lastslashpos = baseuri.GetPathStringLength(); + while (lastslashpos > 0) { + if (baseuri.path_[lastslashpos - 1] == '/') break; + lastslashpos--; + } + std::memcpy(&resuri.path_[pos], baseuri.path_, lastslashpos * sizeof(Ch)); + pos += lastslashpos; + resuri.query_ = CopyPart(&resuri.path_[pos], path_, GetPathStringLength()); + resuri.RemoveDotSegments(); + } + // Use this query + resuri.frag_ = CopyPart(resuri.query_, query_, GetQueryStringLength()); + } + } + } + // Always use this frag + resuri.base_ = CopyPart(resuri.frag_, frag_, GetFragStringLength()); + + // Re-constitute base_ and uri_ + resuri.SetBase(); + resuri.uri_ = resuri.base_ + resuri.GetBaseStringLength() + 1; + resuri.SetUri(); + return resuri; + } + + //! Get the allocator of this GenericUri. + Allocator& GetAllocator() { return *allocator_; } + +private: + // Allocate memory for a URI + // Returns total amount allocated + std::size_t Allocate(std::size_t len) { + // Create own allocator if user did not supply. + if (!allocator_) + ownAllocator_ = allocator_ = RAPIDJSON_NEW(Allocator)(); + + // Allocate one block containing each part of the URI (5) plus base plus full URI, all null terminated. + // Order: scheme, auth, path, query, frag, base, uri + // Note need to set, increment, assign in 3 stages to avoid compiler warning bug. + size_t total = (3 * len + 7) * sizeof(Ch); + scheme_ = static_cast(allocator_->Malloc(total)); + *scheme_ = '\0'; + auth_ = scheme_; + auth_++; + *auth_ = '\0'; + path_ = auth_; + path_++; + *path_ = '\0'; + query_ = path_; + query_++; + *query_ = '\0'; + frag_ = query_; + frag_++; + *frag_ = '\0'; + base_ = frag_; + base_++; + *base_ = '\0'; + uri_ = base_; + uri_++; + *uri_ = '\0'; + return total; + } + + // Free memory for a URI + void Free() { + if (scheme_) { + Allocator::Free(scheme_); + scheme_ = 0; + } + } + + // Parse a URI into constituent scheme, authority, path, query, & fragment parts + // Supports URIs that match regex ^(([^:/?#]+):)?(//([^/?#]*))?([^?#]*)(\?([^#]*))?(#(.*))? as per + // https://tools.ietf.org/html/rfc3986 + void Parse(const Ch* uri, std::size_t len) { + std::size_t start = 0, pos1 = 0, pos2 = 0; + Allocate(len); + + // Look for scheme ([^:/?#]+):)? + if (start < len) { + while (pos1 < len) { + if (uri[pos1] == ':') break; + pos1++; + } + if (pos1 != len) { + while (pos2 < len) { + if (uri[pos2] == '/') break; + if (uri[pos2] == '?') break; + if (uri[pos2] == '#') break; + pos2++; + } + if (pos1 < pos2) { + pos1++; + std::memcpy(scheme_, &uri[start], pos1 * sizeof(Ch)); + scheme_[pos1] = '\0'; + start = pos1; + } + } + } + // Look for auth (//([^/?#]*))? + // Note need to set, increment, assign in 3 stages to avoid compiler warning bug. + auth_ = scheme_ + GetSchemeStringLength(); + auth_++; + *auth_ = '\0'; + if (start < len - 1 && uri[start] == '/' && uri[start + 1] == '/') { + pos2 = start + 2; + while (pos2 < len) { + if (uri[pos2] == '/') break; + if (uri[pos2] == '?') break; + if (uri[pos2] == '#') break; + pos2++; + } + std::memcpy(auth_, &uri[start], (pos2 - start) * sizeof(Ch)); + auth_[pos2 - start] = '\0'; + start = pos2; + } + // Look for path ([^?#]*) + // Note need to set, increment, assign in 3 stages to avoid compiler warning bug. + path_ = auth_ + GetAuthStringLength(); + path_++; + *path_ = '\0'; + if (start < len) { + pos2 = start; + while (pos2 < len) { + if (uri[pos2] == '?') break; + if (uri[pos2] == '#') break; + pos2++; + } + if (start != pos2) { + std::memcpy(path_, &uri[start], (pos2 - start) * sizeof(Ch)); + path_[pos2 - start] = '\0'; + if (path_[0] == '/') + RemoveDotSegments(); // absolute path - normalize + start = pos2; + } + } + // Look for query (\?([^#]*))? + // Note need to set, increment, assign in 3 stages to avoid compiler warning bug. + query_ = path_ + GetPathStringLength(); + query_++; + *query_ = '\0'; + if (start < len && uri[start] == '?') { + pos2 = start + 1; + while (pos2 < len) { + if (uri[pos2] == '#') break; + pos2++; + } + if (start != pos2) { + std::memcpy(query_, &uri[start], (pos2 - start) * sizeof(Ch)); + query_[pos2 - start] = '\0'; + start = pos2; + } + } + // Look for fragment (#(.*))? + // Note need to set, increment, assign in 3 stages to avoid compiler warning bug. + frag_ = query_ + GetQueryStringLength(); + frag_++; + *frag_ = '\0'; + if (start < len && uri[start] == '#') { + std::memcpy(frag_, &uri[start], (len - start) * sizeof(Ch)); + frag_[len - start] = '\0'; + } + + // Re-constitute base_ and uri_ + base_ = frag_ + GetFragStringLength() + 1; + SetBase(); + uri_ = base_ + GetBaseStringLength() + 1; + SetUri(); + } + + // Reconstitute base + void SetBase() { + Ch* next = base_; + std::memcpy(next, scheme_, GetSchemeStringLength() * sizeof(Ch)); + next+= GetSchemeStringLength(); + std::memcpy(next, auth_, GetAuthStringLength() * sizeof(Ch)); + next+= GetAuthStringLength(); + std::memcpy(next, path_, GetPathStringLength() * sizeof(Ch)); + next+= GetPathStringLength(); + std::memcpy(next, query_, GetQueryStringLength() * sizeof(Ch)); + next+= GetQueryStringLength(); + *next = '\0'; + } + + // Reconstitute uri + void SetUri() { + Ch* next = uri_; + std::memcpy(next, base_, GetBaseStringLength() * sizeof(Ch)); + next+= GetBaseStringLength(); + std::memcpy(next, frag_, GetFragStringLength() * sizeof(Ch)); + next+= GetFragStringLength(); + *next = '\0'; + } + + // Copy a part from one GenericUri to another + // Return the pointer to the next part to be copied to + Ch* CopyPart(Ch* to, Ch* from, std::size_t len) { + RAPIDJSON_ASSERT(to != 0); + RAPIDJSON_ASSERT(from != 0); + std::memcpy(to, from, len * sizeof(Ch)); + to[len] = '\0'; + Ch* next = to + len + 1; + return next; + } + + // Remove . and .. segments from the path_ member. + // https://tools.ietf.org/html/rfc3986 + // This is done in place as we are only removing segments. + void RemoveDotSegments() { + std::size_t pathlen = GetPathStringLength(); + std::size_t pathpos = 0; // Position in path_ + std::size_t newpos = 0; // Position in new path_ + + // Loop through each segment in original path_ + while (pathpos < pathlen) { + // Get next segment, bounded by '/' or end + size_t slashpos = 0; + while ((pathpos + slashpos) < pathlen) { + if (path_[pathpos + slashpos] == '/') break; + slashpos++; + } + // Check for .. and . segments + if (slashpos == 2 && path_[pathpos] == '.' && path_[pathpos + 1] == '.') { + // Backup a .. segment in the new path_ + // We expect to find a previously added slash at the end or nothing + RAPIDJSON_ASSERT(newpos == 0 || path_[newpos - 1] == '/'); + size_t lastslashpos = newpos; + // Make sure we don't go beyond the start segment + if (lastslashpos > 1) { + // Find the next to last slash and back up to it + lastslashpos--; + while (lastslashpos > 0) { + if (path_[lastslashpos - 1] == '/') break; + lastslashpos--; + } + // Set the new path_ position + newpos = lastslashpos; + } + } else if (slashpos == 1 && path_[pathpos] == '.') { + // Discard . segment, leaves new path_ unchanged + } else { + // Move any other kind of segment to the new path_ + RAPIDJSON_ASSERT(newpos <= pathpos); + std::memmove(&path_[newpos], &path_[pathpos], slashpos * sizeof(Ch)); + newpos += slashpos; + // Add slash if not at end + if ((pathpos + slashpos) < pathlen) { + path_[newpos] = '/'; + newpos++; + } + } + // Move to next segment + pathpos += slashpos + 1; + } + path_[newpos] = '\0'; + } + + Ch* uri_; // Everything + Ch* base_; // Everything except fragment + Ch* scheme_; // Includes the : + Ch* auth_; // Includes the // + Ch* path_; // Absolute if starts with / + Ch* query_; // Includes the ? + Ch* frag_; // Includes the # + + Allocator* allocator_; //!< The current allocator. It is either user-supplied or equal to ownAllocator_. + Allocator* ownAllocator_; //!< Allocator owned by this Uri. +}; + +//! GenericUri for Value (UTF-8, default allocator). +typedef GenericUri Uri; + +RAPIDJSON_NAMESPACE_END + +#if defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_URI_H_ diff --git a/include/rapidjson/writer.h b/include/rapidjson/writer.h new file mode 100644 index 0000000..632e02c --- /dev/null +++ b/include/rapidjson/writer.h @@ -0,0 +1,721 @@ +// Tencent is pleased to support the open source community by making RapidJSON available. +// +// Copyright (C) 2015 THL A29 Limited, a Tencent company, and Milo Yip. +// +// Licensed under the MIT License (the "License"); you may not use this file except +// in compliance with the License. You may obtain a copy of the License at +// +// http://opensource.org/licenses/MIT +// +// Unless required by applicable law or agreed to in writing, software distributed +// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR +// CONDITIONS OF ANY KIND, either express or implied. See the License for the +// specific language governing permissions and limitations under the License. + +#ifndef RAPIDJSON_WRITER_H_ +#define RAPIDJSON_WRITER_H_ + +#include "stream.h" +#include "internal/clzll.h" +#include "internal/meta.h" +#include "internal/stack.h" +#include "internal/strfunc.h" +#include "internal/dtoa.h" +#include "internal/itoa.h" +#include "stringbuffer.h" +#include // placement new + +#if defined(RAPIDJSON_SIMD) && defined(_MSC_VER) +#include +#pragma intrinsic(_BitScanForward) +#endif +#ifdef RAPIDJSON_SSE42 +#include +#elif defined(RAPIDJSON_SSE2) +#include +#elif defined(RAPIDJSON_NEON) +#include +#endif + +#ifdef __clang__ +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(padded) +RAPIDJSON_DIAG_OFF(unreachable-code) +RAPIDJSON_DIAG_OFF(c++98-compat) +#elif defined(_MSC_VER) +RAPIDJSON_DIAG_PUSH +RAPIDJSON_DIAG_OFF(4127) // conditional expression is constant +#endif + +RAPIDJSON_NAMESPACE_BEGIN + +/////////////////////////////////////////////////////////////////////////////// +// WriteFlag + +/*! \def RAPIDJSON_WRITE_DEFAULT_FLAGS + \ingroup RAPIDJSON_CONFIG + \brief User-defined kWriteDefaultFlags definition. + + User can define this as any \c WriteFlag combinations. +*/ +#ifndef RAPIDJSON_WRITE_DEFAULT_FLAGS +#define RAPIDJSON_WRITE_DEFAULT_FLAGS kWriteNoFlags +#endif + +//! Combination of writeFlags +enum WriteFlag { + kWriteNoFlags = 0, //!< No flags are set. + kWriteValidateEncodingFlag = 1, //!< Validate encoding of JSON strings. + kWriteNanAndInfFlag = 2, //!< Allow writing of Infinity, -Infinity and NaN. + kWriteNanAndInfNullFlag = 4, //!< Allow writing of Infinity, -Infinity and NaN as null. + kWriteDefaultFlags = RAPIDJSON_WRITE_DEFAULT_FLAGS //!< Default write flags. Can be customized by defining RAPIDJSON_WRITE_DEFAULT_FLAGS +}; + +//! JSON writer +/*! Writer implements the concept Handler. + It generates JSON text by events to an output os. + + User may programmatically calls the functions of a writer to generate JSON text. + + On the other side, a writer can also be passed to objects that generates events, + + for example Reader::Parse() and Document::Accept(). + + \tparam OutputStream Type of output stream. + \tparam SourceEncoding Encoding of source string. + \tparam TargetEncoding Encoding of output stream. + \tparam StackAllocator Type of allocator for allocating memory of stack. + \note implements Handler concept +*/ +template, typename TargetEncoding = UTF8<>, typename StackAllocator = CrtAllocator, unsigned writeFlags = kWriteDefaultFlags> +class Writer { +public: + typedef typename SourceEncoding::Ch Ch; + + static const int kDefaultMaxDecimalPlaces = 324; + + //! Constructor + /*! \param os Output stream. + \param stackAllocator User supplied allocator. If it is null, it will create a private one. + \param levelDepth Initial capacity of stack. + */ + explicit + Writer(OutputStream& os, StackAllocator* stackAllocator = 0, size_t levelDepth = kDefaultLevelDepth) : + os_(&os), level_stack_(stackAllocator, levelDepth * sizeof(Level)), maxDecimalPlaces_(kDefaultMaxDecimalPlaces), hasRoot_(false) {} + + explicit + Writer(StackAllocator* allocator = 0, size_t levelDepth = kDefaultLevelDepth) : + os_(0), level_stack_(allocator, levelDepth * sizeof(Level)), maxDecimalPlaces_(kDefaultMaxDecimalPlaces), hasRoot_(false) {} + +#if RAPIDJSON_HAS_CXX11_RVALUE_REFS + Writer(Writer&& rhs) : + os_(rhs.os_), level_stack_(std::move(rhs.level_stack_)), maxDecimalPlaces_(rhs.maxDecimalPlaces_), hasRoot_(rhs.hasRoot_) { + rhs.os_ = 0; + } +#endif + + //! Reset the writer with a new stream. + /*! + This function reset the writer with a new stream and default settings, + in order to make a Writer object reusable for output multiple JSONs. + + \param os New output stream. + \code + Writer writer(os1); + writer.StartObject(); + // ... + writer.EndObject(); + + writer.Reset(os2); + writer.StartObject(); + // ... + writer.EndObject(); + \endcode + */ + void Reset(OutputStream& os) { + os_ = &os; + hasRoot_ = false; + level_stack_.Clear(); + } + + //! Checks whether the output is a complete JSON. + /*! + A complete JSON has a complete root object or array. + */ + bool IsComplete() const { + return hasRoot_ && level_stack_.Empty(); + } + + int GetMaxDecimalPlaces() const { + return maxDecimalPlaces_; + } + + //! Sets the maximum number of decimal places for double output. + /*! + This setting truncates the output with specified number of decimal places. + + For example, + + \code + writer.SetMaxDecimalPlaces(3); + writer.StartArray(); + writer.Double(0.12345); // "0.123" + writer.Double(0.0001); // "0.0" + writer.Double(1.234567890123456e30); // "1.234567890123456e30" (do not truncate significand for positive exponent) + writer.Double(1.23e-4); // "0.0" (do truncate significand for negative exponent) + writer.EndArray(); + \endcode + + The default setting does not truncate any decimal places. You can restore to this setting by calling + \code + writer.SetMaxDecimalPlaces(Writer::kDefaultMaxDecimalPlaces); + \endcode + */ + void SetMaxDecimalPlaces(int maxDecimalPlaces) { + maxDecimalPlaces_ = maxDecimalPlaces; + } + + /*!@name Implementation of Handler + \see Handler + */ + //@{ + + bool Null() { Prefix(kNullType); return EndValue(WriteNull()); } + bool Bool(bool b) { Prefix(b ? kTrueType : kFalseType); return EndValue(WriteBool(b)); } + bool Int(int i) { Prefix(kNumberType); return EndValue(WriteInt(i)); } + bool Uint(unsigned u) { Prefix(kNumberType); return EndValue(WriteUint(u)); } + bool Int64(int64_t i64) { Prefix(kNumberType); return EndValue(WriteInt64(i64)); } + bool Uint64(uint64_t u64) { Prefix(kNumberType); return EndValue(WriteUint64(u64)); } + + //! Writes the given \c double value to the stream + /*! + \param d The value to be written. + \return Whether it is succeed. + */ + bool Double(double d) { Prefix(kNumberType); return EndValue(WriteDouble(d)); } + + bool RawNumber(const Ch* str, SizeType length, bool copy = false) { + RAPIDJSON_ASSERT(str != 0); + (void)copy; + Prefix(kNumberType); + return EndValue(WriteString(str, length)); + } + + bool String(const Ch* str, SizeType length, bool copy = false) { + RAPIDJSON_ASSERT(str != 0); + (void)copy; + Prefix(kStringType); + return EndValue(WriteString(str, length)); + } + +#if RAPIDJSON_HAS_STDSTRING + bool String(const std::basic_string& str) { + return String(str.data(), SizeType(str.size())); + } +#endif + + bool StartObject() { + Prefix(kObjectType); + new (level_stack_.template Push()) Level(false); + return WriteStartObject(); + } + + bool Key(const Ch* str, SizeType length, bool copy = false) { return String(str, length, copy); } + +#if RAPIDJSON_HAS_STDSTRING + bool Key(const std::basic_string& str) + { + return Key(str.data(), SizeType(str.size())); + } +#endif + + bool EndObject(SizeType memberCount = 0) { + (void)memberCount; + RAPIDJSON_ASSERT(level_stack_.GetSize() >= sizeof(Level)); // not inside an Object + RAPIDJSON_ASSERT(!level_stack_.template Top()->inArray); // currently inside an Array, not Object + RAPIDJSON_ASSERT(0 == level_stack_.template Top()->valueCount % 2); // Object has a Key without a Value + level_stack_.template Pop(1); + return EndValue(WriteEndObject()); + } + + bool StartArray() { + Prefix(kArrayType); + new (level_stack_.template Push()) Level(true); + return WriteStartArray(); + } + + bool EndArray(SizeType elementCount = 0) { + (void)elementCount; + RAPIDJSON_ASSERT(level_stack_.GetSize() >= sizeof(Level)); + RAPIDJSON_ASSERT(level_stack_.template Top()->inArray); + level_stack_.template Pop(1); + return EndValue(WriteEndArray()); + } + //@} + + /*! @name Convenience extensions */ + //@{ + + //! Simpler but slower overload. + bool String(const Ch* const& str) { return String(str, internal::StrLen(str)); } + bool Key(const Ch* const& str) { return Key(str, internal::StrLen(str)); } + + //@} + + //! Write a raw JSON value. + /*! + For user to write a stringified JSON as a value. + + \param json A well-formed JSON value. It should not contain null character within [0, length - 1] range. + \param length Length of the json. + \param type Type of the root of json. + */ + bool RawValue(const Ch* json, size_t length, Type type) { + RAPIDJSON_ASSERT(json != 0); + Prefix(type); + return EndValue(WriteRawValue(json, length)); + } + + //! Flush the output stream. + /*! + Allows the user to flush the output stream immediately. + */ + void Flush() { + os_->Flush(); + } + + static const size_t kDefaultLevelDepth = 32; + +protected: + //! Information for each nested level + struct Level { + Level(bool inArray_) : valueCount(0), inArray(inArray_) {} + size_t valueCount; //!< number of values in this level + bool inArray; //!< true if in array, otherwise in object + }; + + bool WriteNull() { + PutReserve(*os_, 4); + PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'u'); PutUnsafe(*os_, 'l'); PutUnsafe(*os_, 'l'); return true; + } + + bool WriteBool(bool b) { + if (b) { + PutReserve(*os_, 4); + PutUnsafe(*os_, 't'); PutUnsafe(*os_, 'r'); PutUnsafe(*os_, 'u'); PutUnsafe(*os_, 'e'); + } + else { + PutReserve(*os_, 5); + PutUnsafe(*os_, 'f'); PutUnsafe(*os_, 'a'); PutUnsafe(*os_, 'l'); PutUnsafe(*os_, 's'); PutUnsafe(*os_, 'e'); + } + return true; + } + + bool WriteInt(int i) { + char buffer[11]; + const char* end = internal::i32toa(i, buffer); + PutReserve(*os_, static_cast(end - buffer)); + for (const char* p = buffer; p != end; ++p) + PutUnsafe(*os_, static_cast(*p)); + return true; + } + + bool WriteUint(unsigned u) { + char buffer[10]; + const char* end = internal::u32toa(u, buffer); + PutReserve(*os_, static_cast(end - buffer)); + for (const char* p = buffer; p != end; ++p) + PutUnsafe(*os_, static_cast(*p)); + return true; + } + + bool WriteInt64(int64_t i64) { + char buffer[21]; + const char* end = internal::i64toa(i64, buffer); + PutReserve(*os_, static_cast(end - buffer)); + for (const char* p = buffer; p != end; ++p) + PutUnsafe(*os_, static_cast(*p)); + return true; + } + + bool WriteUint64(uint64_t u64) { + char buffer[20]; + char* end = internal::u64toa(u64, buffer); + PutReserve(*os_, static_cast(end - buffer)); + for (char* p = buffer; p != end; ++p) + PutUnsafe(*os_, static_cast(*p)); + return true; + } + + bool WriteDouble(double d) { + if (internal::Double(d).IsNanOrInf()) { + if (!(writeFlags & kWriteNanAndInfFlag) && !(writeFlags & kWriteNanAndInfNullFlag)) + return false; + if (writeFlags & kWriteNanAndInfNullFlag) { + PutReserve(*os_, 4); + PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'u'); PutUnsafe(*os_, 'l'); PutUnsafe(*os_, 'l'); + return true; + } + if (internal::Double(d).IsNan()) { + PutReserve(*os_, 3); + PutUnsafe(*os_, 'N'); PutUnsafe(*os_, 'a'); PutUnsafe(*os_, 'N'); + return true; + } + if (internal::Double(d).Sign()) { + PutReserve(*os_, 9); + PutUnsafe(*os_, '-'); + } + else + PutReserve(*os_, 8); + PutUnsafe(*os_, 'I'); PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'f'); + PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 't'); PutUnsafe(*os_, 'y'); + return true; + } + + char buffer[25]; + char* end = internal::dtoa(d, buffer, maxDecimalPlaces_); + PutReserve(*os_, static_cast(end - buffer)); + for (char* p = buffer; p != end; ++p) + PutUnsafe(*os_, static_cast(*p)); + return true; + } + + bool WriteString(const Ch* str, SizeType length) { + static const typename OutputStream::Ch hexDigits[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' }; + static const char escape[256] = { +#define Z16 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 + //0 1 2 3 4 5 6 7 8 9 A B C D E F + 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'b', 't', 'n', 'u', 'f', 'r', 'u', 'u', // 00 + 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', 'u', // 10 + 0, 0, '"', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 20 + Z16, Z16, // 30~4F + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,'\\', 0, 0, 0, // 50 + Z16, Z16, Z16, Z16, Z16, Z16, Z16, Z16, Z16, Z16 // 60~FF +#undef Z16 + }; + + if (TargetEncoding::supportUnicode) + PutReserve(*os_, 2 + length * 6); // "\uxxxx..." + else + PutReserve(*os_, 2 + length * 12); // "\uxxxx\uyyyy..." + + PutUnsafe(*os_, '\"'); + GenericStringStream is(str); + while (ScanWriteUnescapedString(is, length)) { + const Ch c = is.Peek(); + if (!TargetEncoding::supportUnicode && static_cast(c) >= 0x80) { + // Unicode escaping + unsigned codepoint; + if (RAPIDJSON_UNLIKELY(!SourceEncoding::Decode(is, &codepoint))) + return false; + PutUnsafe(*os_, '\\'); + PutUnsafe(*os_, 'u'); + if (codepoint <= 0xD7FF || (codepoint >= 0xE000 && codepoint <= 0xFFFF)) { + PutUnsafe(*os_, hexDigits[(codepoint >> 12) & 15]); + PutUnsafe(*os_, hexDigits[(codepoint >> 8) & 15]); + PutUnsafe(*os_, hexDigits[(codepoint >> 4) & 15]); + PutUnsafe(*os_, hexDigits[(codepoint ) & 15]); + } + else { + RAPIDJSON_ASSERT(codepoint >= 0x010000 && codepoint <= 0x10FFFF); + // Surrogate pair + unsigned s = codepoint - 0x010000; + unsigned lead = (s >> 10) + 0xD800; + unsigned trail = (s & 0x3FF) + 0xDC00; + PutUnsafe(*os_, hexDigits[(lead >> 12) & 15]); + PutUnsafe(*os_, hexDigits[(lead >> 8) & 15]); + PutUnsafe(*os_, hexDigits[(lead >> 4) & 15]); + PutUnsafe(*os_, hexDigits[(lead ) & 15]); + PutUnsafe(*os_, '\\'); + PutUnsafe(*os_, 'u'); + PutUnsafe(*os_, hexDigits[(trail >> 12) & 15]); + PutUnsafe(*os_, hexDigits[(trail >> 8) & 15]); + PutUnsafe(*os_, hexDigits[(trail >> 4) & 15]); + PutUnsafe(*os_, hexDigits[(trail ) & 15]); + } + } + else if ((sizeof(Ch) == 1 || static_cast(c) < 256) && RAPIDJSON_UNLIKELY(escape[static_cast(c)])) { + is.Take(); + PutUnsafe(*os_, '\\'); + PutUnsafe(*os_, static_cast(escape[static_cast(c)])); + if (escape[static_cast(c)] == 'u') { + PutUnsafe(*os_, '0'); + PutUnsafe(*os_, '0'); + PutUnsafe(*os_, hexDigits[static_cast(c) >> 4]); + PutUnsafe(*os_, hexDigits[static_cast(c) & 0xF]); + } + } + else if (RAPIDJSON_UNLIKELY(!(writeFlags & kWriteValidateEncodingFlag ? + Transcoder::Validate(is, *os_) : + Transcoder::TranscodeUnsafe(is, *os_)))) + return false; + } + PutUnsafe(*os_, '\"'); + return true; + } + + bool ScanWriteUnescapedString(GenericStringStream& is, size_t length) { + return RAPIDJSON_LIKELY(is.Tell() < length); + } + + bool WriteStartObject() { os_->Put('{'); return true; } + bool WriteEndObject() { os_->Put('}'); return true; } + bool WriteStartArray() { os_->Put('['); return true; } + bool WriteEndArray() { os_->Put(']'); return true; } + + bool WriteRawValue(const Ch* json, size_t length) { + PutReserve(*os_, length); + GenericStringStream is(json); + while (RAPIDJSON_LIKELY(is.Tell() < length)) { + RAPIDJSON_ASSERT(is.Peek() != '\0'); + if (RAPIDJSON_UNLIKELY(!(writeFlags & kWriteValidateEncodingFlag ? + Transcoder::Validate(is, *os_) : + Transcoder::TranscodeUnsafe(is, *os_)))) + return false; + } + return true; + } + + void Prefix(Type type) { + (void)type; + if (RAPIDJSON_LIKELY(level_stack_.GetSize() != 0)) { // this value is not at root + Level* level = level_stack_.template Top(); + if (level->valueCount > 0) { + if (level->inArray) + os_->Put(','); // add comma if it is not the first element in array + else // in object + os_->Put((level->valueCount % 2 == 0) ? ',' : ':'); + } + if (!level->inArray && level->valueCount % 2 == 0) + RAPIDJSON_ASSERT(type == kStringType); // if it's in object, then even number should be a name + level->valueCount++; + } + else { + RAPIDJSON_ASSERT(!hasRoot_); // Should only has one and only one root. + hasRoot_ = true; + } + } + + // Flush the value if it is the top level one. + bool EndValue(bool ret) { + if (RAPIDJSON_UNLIKELY(level_stack_.Empty())) // end of json text + Flush(); + return ret; + } + + OutputStream* os_; + internal::Stack level_stack_; + int maxDecimalPlaces_; + bool hasRoot_; + +private: + // Prohibit copy constructor & assignment operator. + Writer(const Writer&); + Writer& operator=(const Writer&); +}; + +// Full specialization for StringStream to prevent memory copying + +template<> +inline bool Writer::WriteInt(int i) { + char *buffer = os_->Push(11); + const char* end = internal::i32toa(i, buffer); + os_->Pop(static_cast(11 - (end - buffer))); + return true; +} + +template<> +inline bool Writer::WriteUint(unsigned u) { + char *buffer = os_->Push(10); + const char* end = internal::u32toa(u, buffer); + os_->Pop(static_cast(10 - (end - buffer))); + return true; +} + +template<> +inline bool Writer::WriteInt64(int64_t i64) { + char *buffer = os_->Push(21); + const char* end = internal::i64toa(i64, buffer); + os_->Pop(static_cast(21 - (end - buffer))); + return true; +} + +template<> +inline bool Writer::WriteUint64(uint64_t u) { + char *buffer = os_->Push(20); + const char* end = internal::u64toa(u, buffer); + os_->Pop(static_cast(20 - (end - buffer))); + return true; +} + +template<> +inline bool Writer::WriteDouble(double d) { + if (internal::Double(d).IsNanOrInf()) { + // Note: This code path can only be reached if (RAPIDJSON_WRITE_DEFAULT_FLAGS & kWriteNanAndInfFlag). + if (!(kWriteDefaultFlags & kWriteNanAndInfFlag)) + return false; + if (kWriteDefaultFlags & kWriteNanAndInfNullFlag) { + PutReserve(*os_, 4); + PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'u'); PutUnsafe(*os_, 'l'); PutUnsafe(*os_, 'l'); + return true; + } + if (internal::Double(d).IsNan()) { + PutReserve(*os_, 3); + PutUnsafe(*os_, 'N'); PutUnsafe(*os_, 'a'); PutUnsafe(*os_, 'N'); + return true; + } + if (internal::Double(d).Sign()) { + PutReserve(*os_, 9); + PutUnsafe(*os_, '-'); + } + else + PutReserve(*os_, 8); + PutUnsafe(*os_, 'I'); PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'f'); + PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 'n'); PutUnsafe(*os_, 'i'); PutUnsafe(*os_, 't'); PutUnsafe(*os_, 'y'); + return true; + } + + char *buffer = os_->Push(25); + char* end = internal::dtoa(d, buffer, maxDecimalPlaces_); + os_->Pop(static_cast(25 - (end - buffer))); + return true; +} + +#if defined(RAPIDJSON_SSE2) || defined(RAPIDJSON_SSE42) +template<> +inline bool Writer::ScanWriteUnescapedString(StringStream& is, size_t length) { + if (length < 16) + return RAPIDJSON_LIKELY(is.Tell() < length); + + if (!RAPIDJSON_LIKELY(is.Tell() < length)) + return false; + + const char* p = is.src_; + const char* end = is.head_ + length; + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + const char* endAligned = reinterpret_cast(reinterpret_cast(end) & static_cast(~15)); + if (nextAligned > end) + return true; + + while (p != nextAligned) + if (*p < 0x20 || *p == '\"' || *p == '\\') { + is.src_ = p; + return RAPIDJSON_LIKELY(is.Tell() < length); + } + else + os_->PutUnsafe(*p++); + + // The rest of string using SIMD + static const char dquote[16] = { '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"', '\"' }; + static const char bslash[16] = { '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\', '\\' }; + static const char space[16] = { 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F, 0x1F }; + const __m128i dq = _mm_loadu_si128(reinterpret_cast(&dquote[0])); + const __m128i bs = _mm_loadu_si128(reinterpret_cast(&bslash[0])); + const __m128i sp = _mm_loadu_si128(reinterpret_cast(&space[0])); + + for (; p != endAligned; p += 16) { + const __m128i s = _mm_load_si128(reinterpret_cast(p)); + const __m128i t1 = _mm_cmpeq_epi8(s, dq); + const __m128i t2 = _mm_cmpeq_epi8(s, bs); + const __m128i t3 = _mm_cmpeq_epi8(_mm_max_epu8(s, sp), sp); // s < 0x20 <=> max(s, 0x1F) == 0x1F + const __m128i x = _mm_or_si128(_mm_or_si128(t1, t2), t3); + unsigned short r = static_cast(_mm_movemask_epi8(x)); + if (RAPIDJSON_UNLIKELY(r != 0)) { // some of characters is escaped + SizeType len; +#ifdef _MSC_VER // Find the index of first escaped + unsigned long offset; + _BitScanForward(&offset, r); + len = offset; +#else + len = static_cast(__builtin_ffs(r) - 1); +#endif + char* q = reinterpret_cast(os_->PushUnsafe(len)); + for (size_t i = 0; i < len; i++) + q[i] = p[i]; + + p += len; + break; + } + _mm_storeu_si128(reinterpret_cast<__m128i *>(os_->PushUnsafe(16)), s); + } + + is.src_ = p; + return RAPIDJSON_LIKELY(is.Tell() < length); +} +#elif defined(RAPIDJSON_NEON) +template<> +inline bool Writer::ScanWriteUnescapedString(StringStream& is, size_t length) { + if (length < 16) + return RAPIDJSON_LIKELY(is.Tell() < length); + + if (!RAPIDJSON_LIKELY(is.Tell() < length)) + return false; + + const char* p = is.src_; + const char* end = is.head_ + length; + const char* nextAligned = reinterpret_cast((reinterpret_cast(p) + 15) & static_cast(~15)); + const char* endAligned = reinterpret_cast(reinterpret_cast(end) & static_cast(~15)); + if (nextAligned > end) + return true; + + while (p != nextAligned) + if (*p < 0x20 || *p == '\"' || *p == '\\') { + is.src_ = p; + return RAPIDJSON_LIKELY(is.Tell() < length); + } + else + os_->PutUnsafe(*p++); + + // The rest of string using SIMD + const uint8x16_t s0 = vmovq_n_u8('"'); + const uint8x16_t s1 = vmovq_n_u8('\\'); + const uint8x16_t s2 = vmovq_n_u8('\b'); + const uint8x16_t s3 = vmovq_n_u8(32); + + for (; p != endAligned; p += 16) { + const uint8x16_t s = vld1q_u8(reinterpret_cast(p)); + uint8x16_t x = vceqq_u8(s, s0); + x = vorrq_u8(x, vceqq_u8(s, s1)); + x = vorrq_u8(x, vceqq_u8(s, s2)); + x = vorrq_u8(x, vcltq_u8(s, s3)); + + x = vrev64q_u8(x); // Rev in 64 + uint64_t low = vgetq_lane_u64(vreinterpretq_u64_u8(x), 0); // extract + uint64_t high = vgetq_lane_u64(vreinterpretq_u64_u8(x), 1); // extract + + SizeType len = 0; + bool escaped = false; + if (low == 0) { + if (high != 0) { + uint32_t lz = internal::clzll(high); + len = 8 + (lz >> 3); + escaped = true; + } + } else { + uint32_t lz = internal::clzll(low); + len = lz >> 3; + escaped = true; + } + if (RAPIDJSON_UNLIKELY(escaped)) { // some of characters is escaped + char* q = reinterpret_cast(os_->PushUnsafe(len)); + for (size_t i = 0; i < len; i++) + q[i] = p[i]; + + p += len; + break; + } + vst1q_u8(reinterpret_cast(os_->PushUnsafe(16)), s); + } + + is.src_ = p; + return RAPIDJSON_LIKELY(is.Tell() < length); +} +#endif // RAPIDJSON_NEON + +RAPIDJSON_NAMESPACE_END + +#if defined(_MSC_VER) || defined(__clang__) +RAPIDJSON_DIAG_POP +#endif + +#endif // RAPIDJSON_RAPIDJSON_H_ diff --git a/include/syelog.h b/include/syelog.h new file mode 100644 index 0000000..7cfa9f3 --- /dev/null +++ b/include/syelog.h @@ -0,0 +1,89 @@ +////////////////////////////////////////////////////////////////////////////// +// +// Detours Test Program (syelog.h of syelog.lib) +// +// Microsoft Research Detours Package +// +// Copyright (c) Microsoft Corporation. All rights reserved. +// +#pragma once +#ifndef _SYELOGD_H_ +#define _SYELOGD_H_ +#include + +#pragma pack(push, 1) +#pragma warning(push) +#pragma warning(disable: 4200) + +////////////////////////////////////////////////////////////////////////////// +// +// +#define SYELOG_PIPE_NAMEA "\\\\.\\pipe\\syelog" +#define SYELOG_PIPE_NAMEW L"\\\\.\\pipe\\syelog" +#ifdef UNICODE +#define SYELOG_PIPE_NAME SYELOG_PIPE_NAMEW +#else +#define SYELOG_PIPE_NAME SYELOG_PIPE_NAMEA +#endif + +////////////////////////////////////////////////////////////////////////////// +// +#define SYELOG_MAXIMUM_MESSAGE 4086 // 4096 - sizeof(header stuff) + +typedef struct _SYELOG_MESSAGE +{ + USHORT nBytes; + BYTE nFacility; + BYTE nSeverity; + DWORD nProcessId; + FILETIME ftOccurance; + BOOL fTerminate; + CHAR szMessage[SYELOG_MAXIMUM_MESSAGE]; +} SYELOG_MESSAGE, *PSYELOG_MESSAGE; + + +// Facility Codes. +// +#define SYELOG_FACILITY_KERNEL 0x10 // OS Kernel +#define SYELOG_FACILITY_SECURITY 0x20 // OS Security +#define SYELOG_FACILITY_LOGGING 0x30 // OS Logging-internal +#define SYELOG_FACILITY_SERVICE 0x40 // User-mode system daemon +#define SYELOG_FACILITY_APPLICATION 0x50 // User-mode application +#define SYELOG_FACILITY_USER 0x60 // User self-generated. +#define SYELOG_FACILITY_LOCAL0 0x70 // Locally defined. +#define SYELOG_FACILITY_LOCAL1 0x71 // Locally defined. +#define SYELOG_FACILITY_LOCAL2 0x72 // Locally defined. +#define SYELOG_FACILITY_LOCAL3 0x73 // Locally defined. +#define SYELOG_FACILITY_LOCAL4 0x74 // Locally defined. +#define SYELOG_FACILITY_LOCAL5 0x75 // Locally defined. +#define SYELOG_FACILITY_LOCAL6 0x76 // Locally defined. +#define SYELOG_FACILITY_LOCAL7 0x77 // Locally defined. +#define SYELOG_FACILITY_LOCAL8 0x78 // Locally defined. +#define SYELOG_FACILITY_LOCAL9 0x79 // Locally defined. + +// Severity Codes. +// +#define SYELOG_SEVERITY_FATAL 0x00 // System is dead. +#define SYELOG_SEVERITY_ALERT 0x10 // Take action immediately. +#define SYELOG_SEVERITY_CRITICAL 0x20 // Critical condition. +#define SYELOG_SEVERITY_ERROR 0x30 // Error +#define SYELOG_SEVERITY_WARNING 0x40 // Warning +#define SYELOG_SEVERITY_NOTICE 0x50 // Significant condition. +#define SYELOG_SEVERITY_INFORMATION 0x60 // Informational +#define SYELOG_SEVERITY_AUDIT_FAIL 0x66 // Audit Failed +#define SYELOG_SEVERITY_AUDIT_PASS 0x67 // Audit Succeeeded +#define SYELOG_SEVERITY_DEBUG 0x70 // Debugging + +// Logging Functions. +// +VOID SyelogOpen(PCSTR pszIdentifier, BYTE nFacility); +VOID Syelog(BYTE nSeverity, PCSTR pszMsgf, ...); +VOID SyelogV(BYTE nSeverity, PCSTR pszMsgf, va_list args); +VOID SyelogClose(BOOL fTerminate); + +#pragma warning(pop) +#pragma pack(pop) + +#endif // _SYELOGD_H_ +// +///////////////////////////////////////////////////////////////// End of File. diff --git a/include/zconf.h b/include/zconf.h new file mode 100644 index 0000000..1f1cab2 --- /dev/null +++ b/include/zconf.h @@ -0,0 +1,545 @@ +/* zconf.h -- configuration of the zlib compression library + * Copyright (C) 1995-2024 Jean-loup Gailly, Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id$ */ + +#ifndef ZCONF_H +#define ZCONF_H +/* #undef Z_PREFIX */ +/* #undef Z_HAVE_UNISTD_H */ + +/* + * If you *really* need a unique prefix for all types and library functions, + * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it. + * Even better than compiling with -DZ_PREFIX would be to use configure to set + * this permanently in zconf.h using "./configure --zprefix". + */ +#ifdef Z_PREFIX /* may be set to #if 1 by ./configure */ +# define Z_PREFIX_SET + +/* all linked symbols and init macros */ +# define _dist_code z__dist_code +# define _length_code z__length_code +# define _tr_align z__tr_align +# define _tr_flush_bits z__tr_flush_bits +# define _tr_flush_block z__tr_flush_block +# define _tr_init z__tr_init +# define _tr_stored_block z__tr_stored_block +# define _tr_tally z__tr_tally +# define adler32 z_adler32 +# define adler32_combine z_adler32_combine +# define adler32_combine64 z_adler32_combine64 +# define adler32_z z_adler32_z +# ifndef Z_SOLO +# define compress z_compress +# define compress2 z_compress2 +# define compressBound z_compressBound +# endif +# define crc32 z_crc32 +# define crc32_combine z_crc32_combine +# define crc32_combine64 z_crc32_combine64 +# define crc32_combine_gen z_crc32_combine_gen +# define crc32_combine_gen64 z_crc32_combine_gen64 +# define crc32_combine_op z_crc32_combine_op +# define crc32_z z_crc32_z +# define deflate z_deflate +# define deflateBound z_deflateBound +# define deflateCopy z_deflateCopy +# define deflateEnd z_deflateEnd +# define deflateGetDictionary z_deflateGetDictionary +# define deflateInit z_deflateInit +# define deflateInit2 z_deflateInit2 +# define deflateInit2_ z_deflateInit2_ +# define deflateInit_ z_deflateInit_ +# define deflateParams z_deflateParams +# define deflatePending z_deflatePending +# define deflatePrime z_deflatePrime +# define deflateReset z_deflateReset +# define deflateResetKeep z_deflateResetKeep +# define deflateSetDictionary z_deflateSetDictionary +# define deflateSetHeader z_deflateSetHeader +# define deflateTune z_deflateTune +# define deflate_copyright z_deflate_copyright +# define get_crc_table z_get_crc_table +# ifndef Z_SOLO +# define gz_error z_gz_error +# define gz_intmax z_gz_intmax +# define gz_strwinerror z_gz_strwinerror +# define gzbuffer z_gzbuffer +# define gzclearerr z_gzclearerr +# define gzclose z_gzclose +# define gzclose_r z_gzclose_r +# define gzclose_w z_gzclose_w +# define gzdirect z_gzdirect +# define gzdopen z_gzdopen +# define gzeof z_gzeof +# define gzerror z_gzerror +# define gzflush z_gzflush +# define gzfread z_gzfread +# define gzfwrite z_gzfwrite +# define gzgetc z_gzgetc +# define gzgetc_ z_gzgetc_ +# define gzgets z_gzgets +# define gzoffset z_gzoffset +# define gzoffset64 z_gzoffset64 +# define gzopen z_gzopen +# define gzopen64 z_gzopen64 +# ifdef _WIN32 +# define gzopen_w z_gzopen_w +# endif +# define gzprintf z_gzprintf +# define gzputc z_gzputc +# define gzputs z_gzputs +# define gzread z_gzread +# define gzrewind z_gzrewind +# define gzseek z_gzseek +# define gzseek64 z_gzseek64 +# define gzsetparams z_gzsetparams +# define gztell z_gztell +# define gztell64 z_gztell64 +# define gzungetc z_gzungetc +# define gzvprintf z_gzvprintf +# define gzwrite z_gzwrite +# endif +# define inflate z_inflate +# define inflateBack z_inflateBack +# define inflateBackEnd z_inflateBackEnd +# define inflateBackInit z_inflateBackInit +# define inflateBackInit_ z_inflateBackInit_ +# define inflateCodesUsed z_inflateCodesUsed +# define inflateCopy z_inflateCopy +# define inflateEnd z_inflateEnd +# define inflateGetDictionary z_inflateGetDictionary +# define inflateGetHeader z_inflateGetHeader +# define inflateInit z_inflateInit +# define inflateInit2 z_inflateInit2 +# define inflateInit2_ z_inflateInit2_ +# define inflateInit_ z_inflateInit_ +# define inflateMark z_inflateMark +# define inflatePrime z_inflatePrime +# define inflateReset z_inflateReset +# define inflateReset2 z_inflateReset2 +# define inflateResetKeep z_inflateResetKeep +# define inflateSetDictionary z_inflateSetDictionary +# define inflateSync z_inflateSync +# define inflateSyncPoint z_inflateSyncPoint +# define inflateUndermine z_inflateUndermine +# define inflateValidate z_inflateValidate +# define inflate_copyright z_inflate_copyright +# define inflate_fast z_inflate_fast +# define inflate_table z_inflate_table +# ifndef Z_SOLO +# define uncompress z_uncompress +# define uncompress2 z_uncompress2 +# endif +# define zError z_zError +# ifndef Z_SOLO +# define zcalloc z_zcalloc +# define zcfree z_zcfree +# endif +# define zlibCompileFlags z_zlibCompileFlags +# define zlibVersion z_zlibVersion + +/* all zlib typedefs in zlib.h and zconf.h */ +# define Byte z_Byte +# define Bytef z_Bytef +# define alloc_func z_alloc_func +# define charf z_charf +# define free_func z_free_func +# ifndef Z_SOLO +# define gzFile z_gzFile +# endif +# define gz_header z_gz_header +# define gz_headerp z_gz_headerp +# define in_func z_in_func +# define intf z_intf +# define out_func z_out_func +# define uInt z_uInt +# define uIntf z_uIntf +# define uLong z_uLong +# define uLongf z_uLongf +# define voidp z_voidp +# define voidpc z_voidpc +# define voidpf z_voidpf + +/* all zlib structs in zlib.h and zconf.h */ +# define gz_header_s z_gz_header_s +# define internal_state z_internal_state + +#endif + +#if defined(__MSDOS__) && !defined(MSDOS) +# define MSDOS +#endif +#if (defined(OS_2) || defined(__OS2__)) && !defined(OS2) +# define OS2 +#endif +#if defined(_WINDOWS) && !defined(WINDOWS) +# define WINDOWS +#endif +#if defined(_WIN32) || defined(_WIN32_WCE) || defined(__WIN32__) +# ifndef WIN32 +# define WIN32 +# endif +#endif +#if (defined(MSDOS) || defined(OS2) || defined(WINDOWS)) && !defined(WIN32) +# if !defined(__GNUC__) && !defined(__FLAT__) && !defined(__386__) +# ifndef SYS16BIT +# define SYS16BIT +# endif +# endif +#endif + +/* + * Compile with -DMAXSEG_64K if the alloc function cannot allocate more + * than 64k bytes at a time (needed on systems with 16-bit int). + */ +#ifdef SYS16BIT +# define MAXSEG_64K +#endif +#ifdef MSDOS +# define UNALIGNED_OK +#endif + +#ifdef __STDC_VERSION__ +# ifndef STDC +# define STDC +# endif +# if __STDC_VERSION__ >= 199901L +# ifndef STDC99 +# define STDC99 +# endif +# endif +#endif +#if !defined(STDC) && (defined(__STDC__) || defined(__cplusplus)) +# define STDC +#endif +#if !defined(STDC) && (defined(__GNUC__) || defined(__BORLANDC__)) +# define STDC +#endif +#if !defined(STDC) && (defined(MSDOS) || defined(WINDOWS) || defined(WIN32)) +# define STDC +#endif +#if !defined(STDC) && (defined(OS2) || defined(__HOS_AIX__)) +# define STDC +#endif + +#if defined(__OS400__) && !defined(STDC) /* iSeries (formerly AS/400). */ +# define STDC +#endif + +#ifndef STDC +# ifndef const /* cannot use !defined(STDC) && !defined(const) on Mac */ +# define const /* note: need a more gentle solution here */ +# endif +#endif + +#if defined(ZLIB_CONST) && !defined(z_const) +# define z_const const +#else +# define z_const +#endif + +#ifdef Z_SOLO +# ifdef _WIN64 + typedef unsigned long long z_size_t; +# else + typedef unsigned long z_size_t; +# endif +#else +# define z_longlong long long +# if defined(NO_SIZE_T) + typedef unsigned NO_SIZE_T z_size_t; +# elif defined(STDC) +# include + typedef size_t z_size_t; +# else + typedef unsigned long z_size_t; +# endif +# undef z_longlong +#endif + +/* Maximum value for memLevel in deflateInit2 */ +#ifndef MAX_MEM_LEVEL +# ifdef MAXSEG_64K +# define MAX_MEM_LEVEL 8 +# else +# define MAX_MEM_LEVEL 9 +# endif +#endif + +/* Maximum value for windowBits in deflateInit2 and inflateInit2. + * WARNING: reducing MAX_WBITS makes minigzip unable to extract .gz files + * created by gzip. (Files created by minigzip can still be extracted by + * gzip.) + */ +#ifndef MAX_WBITS +# define MAX_WBITS 15 /* 32K LZ77 window */ +#endif + +/* The memory requirements for deflate are (in bytes): + (1 << (windowBits+2)) + (1 << (memLevel+9)) + that is: 128K for windowBits=15 + 128K for memLevel = 8 (default values) + plus a few kilobytes for small objects. For example, if you want to reduce + the default memory requirements from 256K to 128K, compile with + make CFLAGS="-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7" + Of course this will generally degrade compression (there's no free lunch). + + The memory requirements for inflate are (in bytes) 1 << windowBits + that is, 32K for windowBits=15 (default value) plus about 7 kilobytes + for small objects. +*/ + + /* Type declarations */ + +#ifndef OF /* function prototypes */ +# ifdef STDC +# define OF(args) args +# else +# define OF(args) () +# endif +#endif + +/* The following definitions for FAR are needed only for MSDOS mixed + * model programming (small or medium model with some far allocations). + * This was tested only with MSC; for other MSDOS compilers you may have + * to define NO_MEMCPY in zutil.h. If you don't need the mixed model, + * just define FAR to be empty. + */ +#ifdef SYS16BIT +# if defined(M_I86SM) || defined(M_I86MM) + /* MSC small or medium model */ +# define SMALL_MEDIUM +# ifdef _MSC_VER +# define FAR _far +# else +# define FAR far +# endif +# endif +# if (defined(__SMALL__) || defined(__MEDIUM__)) + /* Turbo C small or medium model */ +# define SMALL_MEDIUM +# ifdef __BORLANDC__ +# define FAR _far +# else +# define FAR far +# endif +# endif +#endif + +#if defined(WINDOWS) || defined(WIN32) + /* If building or using zlib as a DLL, define ZLIB_DLL. + * This is not mandatory, but it offers a little performance increase. + */ +# ifdef ZLIB_DLL +# if defined(WIN32) && (!defined(__BORLANDC__) || (__BORLANDC__ >= 0x500)) +# ifdef ZLIB_INTERNAL +# define ZEXTERN extern __declspec(dllexport) +# else +# define ZEXTERN extern __declspec(dllimport) +# endif +# endif +# endif /* ZLIB_DLL */ + /* If building or using zlib with the WINAPI/WINAPIV calling convention, + * define ZLIB_WINAPI. + * Caution: the standard ZLIB1.DLL is NOT compiled using ZLIB_WINAPI. + */ +# ifdef ZLIB_WINAPI +# ifdef FAR +# undef FAR +# endif +# ifndef WIN32_LEAN_AND_MEAN +# define WIN32_LEAN_AND_MEAN +# endif +# include + /* No need for _export, use ZLIB.DEF instead. */ + /* For complete Windows compatibility, use WINAPI, not __stdcall. */ +# define ZEXPORT WINAPI +# ifdef WIN32 +# define ZEXPORTVA WINAPIV +# else +# define ZEXPORTVA FAR CDECL +# endif +# endif +#endif + +#if defined (__BEOS__) +# ifdef ZLIB_DLL +# ifdef ZLIB_INTERNAL +# define ZEXPORT __declspec(dllexport) +# define ZEXPORTVA __declspec(dllexport) +# else +# define ZEXPORT __declspec(dllimport) +# define ZEXPORTVA __declspec(dllimport) +# endif +# endif +#endif + +#ifndef ZEXTERN +# define ZEXTERN extern +#endif +#ifndef ZEXPORT +# define ZEXPORT +#endif +#ifndef ZEXPORTVA +# define ZEXPORTVA +#endif + +#ifndef FAR +# define FAR +#endif + +#if !defined(__MACTYPES__) +typedef unsigned char Byte; /* 8 bits */ +#endif +typedef unsigned int uInt; /* 16 bits or more */ +typedef unsigned long uLong; /* 32 bits or more */ + +#ifdef SMALL_MEDIUM + /* Borland C/C++ and some old MSC versions ignore FAR inside typedef */ +# define Bytef Byte FAR +#else + typedef Byte FAR Bytef; +#endif +typedef char FAR charf; +typedef int FAR intf; +typedef uInt FAR uIntf; +typedef uLong FAR uLongf; + +#ifdef STDC + typedef void const *voidpc; + typedef void FAR *voidpf; + typedef void *voidp; +#else + typedef Byte const *voidpc; + typedef Byte FAR *voidpf; + typedef Byte *voidp; +#endif + +#if !defined(Z_U4) && !defined(Z_SOLO) && defined(STDC) +# include +# if (UINT_MAX == 0xffffffffUL) +# define Z_U4 unsigned +# elif (ULONG_MAX == 0xffffffffUL) +# define Z_U4 unsigned long +# elif (USHRT_MAX == 0xffffffffUL) +# define Z_U4 unsigned short +# endif +#endif + +#ifdef Z_U4 + typedef Z_U4 z_crc_t; +#else + typedef unsigned long z_crc_t; +#endif + +#ifdef HAVE_UNISTD_H /* may be set to #if 1 by ./configure */ +# define Z_HAVE_UNISTD_H +#endif + +#ifdef HAVE_STDARG_H /* may be set to #if 1 by ./configure */ +# define Z_HAVE_STDARG_H +#endif + +#ifdef STDC +# ifndef Z_SOLO +# include /* for off_t */ +# endif +#endif + +#if defined(STDC) || defined(Z_HAVE_STDARG_H) +# ifndef Z_SOLO +# include /* for va_list */ +# endif +#endif + +#ifdef _WIN32 +# ifndef Z_SOLO +# include /* for wchar_t */ +# endif +#endif + +/* a little trick to accommodate both "#define _LARGEFILE64_SOURCE" and + * "#define _LARGEFILE64_SOURCE 1" as requesting 64-bit operations, (even + * though the former does not conform to the LFS document), but considering + * both "#undef _LARGEFILE64_SOURCE" and "#define _LARGEFILE64_SOURCE 0" as + * equivalently requesting no 64-bit operations + */ +#if defined(_LARGEFILE64_SOURCE) && -_LARGEFILE64_SOURCE - -1 == 1 +# undef _LARGEFILE64_SOURCE +#endif + +#ifndef Z_HAVE_UNISTD_H +# ifdef __WATCOMC__ +# define Z_HAVE_UNISTD_H +# endif +#endif +#ifndef Z_HAVE_UNISTD_H +# if defined(_LARGEFILE64_SOURCE) && !defined(_WIN32) +# define Z_HAVE_UNISTD_H +# endif +#endif +#ifndef Z_SOLO +# if defined(Z_HAVE_UNISTD_H) +//# include /* for SEEK_*, off_t, and _LFS64_LARGEFILE */ +# ifdef VMS +# include /* for off_t */ +# endif +# ifndef z_off_t +# define z_off_t off_t +# endif +# endif +#endif + +#if defined(_LFS64_LARGEFILE) && _LFS64_LARGEFILE-0 +# define Z_LFS64 +#endif + +#if defined(_LARGEFILE64_SOURCE) && defined(Z_LFS64) +# define Z_LARGE64 +#endif + +#if defined(_FILE_OFFSET_BITS) && _FILE_OFFSET_BITS-0 == 64 && defined(Z_LFS64) +# define Z_WANT64 +#endif + +#if !defined(SEEK_SET) && !defined(Z_SOLO) +# define SEEK_SET 0 /* Seek from beginning of file. */ +# define SEEK_CUR 1 /* Seek from current position. */ +# define SEEK_END 2 /* Set file pointer to EOF plus "offset" */ +#endif + +#ifndef z_off_t +# define z_off_t long +#endif + +#if !defined(_WIN32) && defined(Z_LARGE64) +# define z_off64_t off64_t +#else +# if defined(_WIN32) && !defined(__GNUC__) +# define z_off64_t __int64 +# else +# define z_off64_t z_off_t +# endif +#endif + +/* MVS linker does not support external names larger than 8 bytes */ +#if defined(__MVS__) + #pragma map(deflateInit_,"DEIN") + #pragma map(deflateInit2_,"DEIN2") + #pragma map(deflateEnd,"DEEND") + #pragma map(deflateBound,"DEBND") + #pragma map(inflateInit_,"ININ") + #pragma map(inflateInit2_,"ININ2") + #pragma map(inflateEnd,"INEND") + #pragma map(inflateSync,"INSY") + #pragma map(inflateSetDictionary,"INSEDI") + #pragma map(compressBound,"CMBND") + #pragma map(inflate_table,"INTABL") + #pragma map(inflate_fast,"INFA") + #pragma map(inflate_copyright,"INCOPY") +#endif + +#endif /* ZCONF_H */ diff --git a/include/zlib.h b/include/zlib.h new file mode 100644 index 0000000..8d4b932 --- /dev/null +++ b/include/zlib.h @@ -0,0 +1,1938 @@ +/* zlib.h -- interface of the 'zlib' general purpose compression library + version 1.3.1, January 22nd, 2024 + + Copyright (C) 1995-2024 Jean-loup Gailly and Mark Adler + + This software is provided 'as-is', without any express or implied + warranty. In no event will the authors be held liable for any damages + arising from the use of this software. + + Permission is granted to anyone to use this software for any purpose, + including commercial applications, and to alter it and redistribute it + freely, subject to the following restrictions: + + 1. The origin of this software must not be misrepresented; you must not + claim that you wrote the original software. If you use this software + in a product, an acknowledgment in the product documentation would be + appreciated but is not required. + 2. Altered source versions must be plainly marked as such, and must not be + misrepresented as being the original software. + 3. This notice may not be removed or altered from any source distribution. + + Jean-loup Gailly Mark Adler + jloup@gzip.org madler@alumni.caltech.edu + + + The data format used by the zlib library is described by RFCs (Request for + Comments) 1950 to 1952 in the files http://tools.ietf.org/html/rfc1950 + (zlib format), rfc1951 (deflate format) and rfc1952 (gzip format). +*/ + +#ifndef ZLIB_H +#define ZLIB_H + +#include "zconf.h" + +#ifdef __cplusplus +extern "C" { +#endif + +#define ZLIB_VERSION "1.3.1" +#define ZLIB_VERNUM 0x1310 +#define ZLIB_VER_MAJOR 1 +#define ZLIB_VER_MINOR 3 +#define ZLIB_VER_REVISION 1 +#define ZLIB_VER_SUBREVISION 0 + +/* + The 'zlib' compression library provides in-memory compression and + decompression functions, including integrity checks of the uncompressed data. + This version of the library supports only one compression method (deflation) + but other algorithms will be added later and will have the same stream + interface. + + Compression can be done in a single step if the buffers are large enough, + or can be done by repeated calls of the compression function. In the latter + case, the application must provide more input and/or consume the output + (providing more output space) before each call. + + The compressed data format used by default by the in-memory functions is + the zlib format, which is a zlib wrapper documented in RFC 1950, wrapped + around a deflate stream, which is itself documented in RFC 1951. + + The library also supports reading and writing files in gzip (.gz) format + with an interface similar to that of stdio using the functions that start + with "gz". The gzip format is different from the zlib format. gzip is a + gzip wrapper, documented in RFC 1952, wrapped around a deflate stream. + + This library can optionally read and write gzip and raw deflate streams in + memory as well. + + The zlib format was designed to be compact and fast for use in memory + and on communications channels. The gzip format was designed for single- + file compression on file systems, has a larger header than zlib to maintain + directory information, and uses a different, slower check method than zlib. + + The library does not install any signal handler. The decoder checks + the consistency of the compressed data, so the library should never crash + even in the case of corrupted input. +*/ + +typedef voidpf (*alloc_func)(voidpf opaque, uInt items, uInt size); +typedef void (*free_func)(voidpf opaque, voidpf address); + +struct internal_state; + +typedef struct z_stream_s { + z_const Bytef *next_in; /* next input byte */ + uInt avail_in; /* number of bytes available at next_in */ + uLong total_in; /* total number of input bytes read so far */ + + Bytef *next_out; /* next output byte will go here */ + uInt avail_out; /* remaining free space at next_out */ + uLong total_out; /* total number of bytes output so far */ + + z_const char *msg; /* last error message, NULL if no error */ + struct internal_state FAR *state; /* not visible by applications */ + + alloc_func zalloc; /* used to allocate the internal state */ + free_func zfree; /* used to free the internal state */ + voidpf opaque; /* private data object passed to zalloc and zfree */ + + int data_type; /* best guess about the data type: binary or text + for deflate, or the decoding state for inflate */ + uLong adler; /* Adler-32 or CRC-32 value of the uncompressed data */ + uLong reserved; /* reserved for future use */ +} z_stream; + +typedef z_stream FAR *z_streamp; + +/* + gzip header information passed to and from zlib routines. See RFC 1952 + for more details on the meanings of these fields. +*/ +typedef struct gz_header_s { + int text; /* true if compressed data believed to be text */ + uLong time; /* modification time */ + int xflags; /* extra flags (not used when writing a gzip file) */ + int os; /* operating system */ + Bytef *extra; /* pointer to extra field or Z_NULL if none */ + uInt extra_len; /* extra field length (valid if extra != Z_NULL) */ + uInt extra_max; /* space at extra (only when reading header) */ + Bytef *name; /* pointer to zero-terminated file name or Z_NULL */ + uInt name_max; /* space at name (only when reading header) */ + Bytef *comment; /* pointer to zero-terminated comment or Z_NULL */ + uInt comm_max; /* space at comment (only when reading header) */ + int hcrc; /* true if there was or will be a header crc */ + int done; /* true when done reading gzip header (not used + when writing a gzip file) */ +} gz_header; + +typedef gz_header FAR *gz_headerp; + +/* + The application must update next_in and avail_in when avail_in has dropped + to zero. It must update next_out and avail_out when avail_out has dropped + to zero. The application must initialize zalloc, zfree and opaque before + calling the init function. All other fields are set by the compression + library and must not be updated by the application. + + The opaque value provided by the application will be passed as the first + parameter for calls of zalloc and zfree. This can be useful for custom + memory management. The compression library attaches no meaning to the + opaque value. + + zalloc must return Z_NULL if there is not enough memory for the object. + If zlib is used in a multi-threaded application, zalloc and zfree must be + thread safe. In that case, zlib is thread-safe. When zalloc and zfree are + Z_NULL on entry to the initialization function, they are set to internal + routines that use the standard library functions malloc() and free(). + + On 16-bit systems, the functions zalloc and zfree must be able to allocate + exactly 65536 bytes, but will not be required to allocate more than this if + the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS, pointers + returned by zalloc for objects of exactly 65536 bytes *must* have their + offset normalized to zero. The default allocation function provided by this + library ensures this (see zutil.c). To reduce memory requirements and avoid + any allocation of 64K objects, at the expense of compression ratio, compile + the library with -DMAX_WBITS=14 (see zconf.h). + + The fields total_in and total_out can be used for statistics or progress + reports. After compression, total_in holds the total size of the + uncompressed data and may be saved for use by the decompressor (particularly + if the decompressor wants to decompress everything in a single step). +*/ + + /* constants */ + +#define Z_NO_FLUSH 0 +#define Z_PARTIAL_FLUSH 1 +#define Z_SYNC_FLUSH 2 +#define Z_FULL_FLUSH 3 +#define Z_FINISH 4 +#define Z_BLOCK 5 +#define Z_TREES 6 +/* Allowed flush values; see deflate() and inflate() below for details */ + +#define Z_OK 0 +#define Z_STREAM_END 1 +#define Z_NEED_DICT 2 +#define Z_ERRNO (-1) +#define Z_STREAM_ERROR (-2) +#define Z_DATA_ERROR (-3) +#define Z_MEM_ERROR (-4) +#define Z_BUF_ERROR (-5) +#define Z_VERSION_ERROR (-6) +/* Return codes for the compression/decompression functions. Negative values + * are errors, positive values are used for special but normal events. + */ + +#define Z_NO_COMPRESSION 0 +#define Z_BEST_SPEED 1 +#define Z_BEST_COMPRESSION 9 +#define Z_DEFAULT_COMPRESSION (-1) +/* compression levels */ + +#define Z_FILTERED 1 +#define Z_HUFFMAN_ONLY 2 +#define Z_RLE 3 +#define Z_FIXED 4 +#define Z_DEFAULT_STRATEGY 0 +/* compression strategy; see deflateInit2() below for details */ + +#define Z_BINARY 0 +#define Z_TEXT 1 +#define Z_ASCII Z_TEXT /* for compatibility with 1.2.2 and earlier */ +#define Z_UNKNOWN 2 +/* Possible values of the data_type field for deflate() */ + +#define Z_DEFLATED 8 +/* The deflate compression method (the only one supported in this version) */ + +#define Z_NULL 0 /* for initializing zalloc, zfree, opaque */ + +#define zlib_version zlibVersion() +/* for compatibility with versions < 1.0.2 */ + + + /* basic functions */ + +ZEXTERN const char * ZEXPORT zlibVersion(void); +/* The application can compare zlibVersion and ZLIB_VERSION for consistency. + If the first character differs, the library code actually used is not + compatible with the zlib.h header file used by the application. This check + is automatically made by deflateInit and inflateInit. + */ + +/* +ZEXTERN int ZEXPORT deflateInit(z_streamp strm, int level); + + Initializes the internal stream state for compression. The fields + zalloc, zfree and opaque must be initialized before by the caller. If + zalloc and zfree are set to Z_NULL, deflateInit updates them to use default + allocation functions. total_in, total_out, adler, and msg are initialized. + + The compression level must be Z_DEFAULT_COMPRESSION, or between 0 and 9: + 1 gives best speed, 9 gives best compression, 0 gives no compression at all + (the input data is simply copied a block at a time). Z_DEFAULT_COMPRESSION + requests a default compromise between speed and compression (currently + equivalent to level 6). + + deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_STREAM_ERROR if level is not a valid compression level, or + Z_VERSION_ERROR if the zlib library version (zlib_version) is incompatible + with the version assumed by the caller (ZLIB_VERSION). msg is set to null + if there is no error message. deflateInit does not perform any compression: + this will be done by deflate(). +*/ + + +ZEXTERN int ZEXPORT deflate(z_streamp strm, int flush); +/* + deflate compresses as much data as possible, and stops when the input + buffer becomes empty or the output buffer becomes full. It may introduce + some output latency (reading input without producing any output) except when + forced to flush. + + The detailed semantics are as follows. deflate performs one or both of the + following actions: + + - Compress more input starting at next_in and update next_in and avail_in + accordingly. If not all input can be processed (because there is not + enough room in the output buffer), next_in and avail_in are updated and + processing will resume at this point for the next call of deflate(). + + - Generate more output starting at next_out and update next_out and avail_out + accordingly. This action is forced if the parameter flush is non zero. + Forcing flush frequently degrades the compression ratio, so this parameter + should be set only when necessary. Some output may be provided even if + flush is zero. + + Before the call of deflate(), the application should ensure that at least + one of the actions is possible, by providing more input and/or consuming more + output, and updating avail_in or avail_out accordingly; avail_out should + never be zero before the call. The application can consume the compressed + output when it wants, for example when the output buffer is full (avail_out + == 0), or after each call of deflate(). If deflate returns Z_OK and with + zero avail_out, it must be called again after making room in the output + buffer because there might be more output pending. See deflatePending(), + which can be used if desired to determine whether or not there is more output + in that case. + + Normally the parameter flush is set to Z_NO_FLUSH, which allows deflate to + decide how much data to accumulate before producing output, in order to + maximize compression. + + If the parameter flush is set to Z_SYNC_FLUSH, all pending output is + flushed to the output buffer and the output is aligned on a byte boundary, so + that the decompressor can get all input data available so far. (In + particular avail_in is zero after the call if enough output space has been + provided before the call.) Flushing may degrade compression for some + compression algorithms and so it should be used only when necessary. This + completes the current deflate block and follows it with an empty stored block + that is three bits plus filler bits to the next byte, followed by four bytes + (00 00 ff ff). + + If flush is set to Z_PARTIAL_FLUSH, all pending output is flushed to the + output buffer, but the output is not aligned to a byte boundary. All of the + input data so far will be available to the decompressor, as for Z_SYNC_FLUSH. + This completes the current deflate block and follows it with an empty fixed + codes block that is 10 bits long. This assures that enough bytes are output + in order for the decompressor to finish the block before the empty fixed + codes block. + + If flush is set to Z_BLOCK, a deflate block is completed and emitted, as + for Z_SYNC_FLUSH, but the output is not aligned on a byte boundary, and up to + seven bits of the current block are held to be written as the next byte after + the next deflate block is completed. In this case, the decompressor may not + be provided enough bits at this point in order to complete decompression of + the data provided so far to the compressor. It may need to wait for the next + block to be emitted. This is for advanced applications that need to control + the emission of deflate blocks. + + If flush is set to Z_FULL_FLUSH, all output is flushed as with + Z_SYNC_FLUSH, and the compression state is reset so that decompression can + restart from this point if previous compressed data has been damaged or if + random access is desired. Using Z_FULL_FLUSH too often can seriously degrade + compression. + + If deflate returns with avail_out == 0, this function must be called again + with the same value of the flush parameter and more output space (updated + avail_out), until the flush is complete (deflate returns with non-zero + avail_out). In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that + avail_out is greater than six when the flush marker begins, in order to avoid + repeated flush markers upon calling deflate() again when avail_out == 0. + + If the parameter flush is set to Z_FINISH, pending input is processed, + pending output is flushed and deflate returns with Z_STREAM_END if there was + enough output space. If deflate returns with Z_OK or Z_BUF_ERROR, this + function must be called again with Z_FINISH and more output space (updated + avail_out) but no more input data, until it returns with Z_STREAM_END or an + error. After deflate has returned Z_STREAM_END, the only possible operations + on the stream are deflateReset or deflateEnd. + + Z_FINISH can be used in the first deflate call after deflateInit if all the + compression is to be done in a single step. In order to complete in one + call, avail_out must be at least the value returned by deflateBound (see + below). Then deflate is guaranteed to return Z_STREAM_END. If not enough + output space is provided, deflate will not return Z_STREAM_END, and it must + be called again as described above. + + deflate() sets strm->adler to the Adler-32 checksum of all input read + so far (that is, total_in bytes). If a gzip stream is being generated, then + strm->adler will be the CRC-32 checksum of the input read so far. (See + deflateInit2 below.) + + deflate() may update strm->data_type if it can make a good guess about + the input data type (Z_BINARY or Z_TEXT). If in doubt, the data is + considered binary. This field is only for information purposes and does not + affect the compression algorithm in any manner. + + deflate() returns Z_OK if some progress has been made (more input + processed or more output produced), Z_STREAM_END if all input has been + consumed and all output has been produced (only when flush is set to + Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example + if next_in or next_out was Z_NULL or the state was inadvertently written over + by the application), or Z_BUF_ERROR if no progress is possible (for example + avail_in or avail_out was zero). Note that Z_BUF_ERROR is not fatal, and + deflate() can be called again with more input and more output space to + continue compressing. +*/ + + +ZEXTERN int ZEXPORT deflateEnd(z_streamp strm); +/* + All dynamically allocated data structures for this stream are freed. + This function discards any unprocessed input and does not flush any pending + output. + + deflateEnd returns Z_OK if success, Z_STREAM_ERROR if the + stream state was inconsistent, Z_DATA_ERROR if the stream was freed + prematurely (some input or output was discarded). In the error case, msg + may be set but then points to a static string (which must not be + deallocated). +*/ + + +/* +ZEXTERN int ZEXPORT inflateInit(z_streamp strm); + + Initializes the internal stream state for decompression. The fields + next_in, avail_in, zalloc, zfree and opaque must be initialized before by + the caller. In the current version of inflate, the provided input is not + read or consumed. The allocation of a sliding window will be deferred to + the first call of inflate (if the decompression does not complete on the + first call). If zalloc and zfree are set to Z_NULL, inflateInit updates + them to use default allocation functions. total_in, total_out, adler, and + msg are initialized. + + inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_VERSION_ERROR if the zlib library version is incompatible with the + version assumed by the caller, or Z_STREAM_ERROR if the parameters are + invalid, such as a null pointer to the structure. msg is set to null if + there is no error message. inflateInit does not perform any decompression. + Actual decompression will be done by inflate(). So next_in, and avail_in, + next_out, and avail_out are unused and unchanged. The current + implementation of inflateInit() does not process any header information -- + that is deferred until inflate() is called. +*/ + + +ZEXTERN int ZEXPORT inflate(z_streamp strm, int flush); +/* + inflate decompresses as much data as possible, and stops when the input + buffer becomes empty or the output buffer becomes full. It may introduce + some output latency (reading input without producing any output) except when + forced to flush. + + The detailed semantics are as follows. inflate performs one or both of the + following actions: + + - Decompress more input starting at next_in and update next_in and avail_in + accordingly. If not all input can be processed (because there is not + enough room in the output buffer), then next_in and avail_in are updated + accordingly, and processing will resume at this point for the next call of + inflate(). + + - Generate more output starting at next_out and update next_out and avail_out + accordingly. inflate() provides as much output as possible, until there is + no more input data or no more space in the output buffer (see below about + the flush parameter). + + Before the call of inflate(), the application should ensure that at least + one of the actions is possible, by providing more input and/or consuming more + output, and updating the next_* and avail_* values accordingly. If the + caller of inflate() does not provide both available input and available + output space, it is possible that there will be no progress made. The + application can consume the uncompressed output when it wants, for example + when the output buffer is full (avail_out == 0), or after each call of + inflate(). If inflate returns Z_OK and with zero avail_out, it must be + called again after making room in the output buffer because there might be + more output pending. + + The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH, Z_FINISH, + Z_BLOCK, or Z_TREES. Z_SYNC_FLUSH requests that inflate() flush as much + output as possible to the output buffer. Z_BLOCK requests that inflate() + stop if and when it gets to the next deflate block boundary. When decoding + the zlib or gzip format, this will cause inflate() to return immediately + after the header and before the first block. When doing a raw inflate, + inflate() will go ahead and process the first block, and will return when it + gets to the end of that block, or when it runs out of data. + + The Z_BLOCK option assists in appending to or combining deflate streams. + To assist in this, on return inflate() always sets strm->data_type to the + number of unused bits in the last byte taken from strm->next_in, plus 64 if + inflate() is currently decoding the last block in the deflate stream, plus + 128 if inflate() returned immediately after decoding an end-of-block code or + decoding the complete header up to just before the first byte of the deflate + stream. The end-of-block will not be indicated until all of the uncompressed + data from that block has been written to strm->next_out. The number of + unused bits may in general be greater than seven, except when bit 7 of + data_type is set, in which case the number of unused bits will be less than + eight. data_type is set as noted here every time inflate() returns for all + flush options, and so can be used to determine the amount of currently + consumed input in bits. + + The Z_TREES option behaves as Z_BLOCK does, but it also returns when the + end of each deflate block header is reached, before any actual data in that + block is decoded. This allows the caller to determine the length of the + deflate block header for later use in random access within a deflate block. + 256 is added to the value of strm->data_type when inflate() returns + immediately after reaching the end of the deflate block header. + + inflate() should normally be called until it returns Z_STREAM_END or an + error. However if all decompression is to be performed in a single step (a + single call of inflate), the parameter flush should be set to Z_FINISH. In + this case all pending input is processed and all pending output is flushed; + avail_out must be large enough to hold all of the uncompressed data for the + operation to complete. (The size of the uncompressed data may have been + saved by the compressor for this purpose.) The use of Z_FINISH is not + required to perform an inflation in one step. However it may be used to + inform inflate that a faster approach can be used for the single inflate() + call. Z_FINISH also informs inflate to not maintain a sliding window if the + stream completes, which reduces inflate's memory footprint. If the stream + does not complete, either because not all of the stream is provided or not + enough output space is provided, then a sliding window will be allocated and + inflate() can be called again to continue the operation as if Z_NO_FLUSH had + been used. + + In this implementation, inflate() always flushes as much output as + possible to the output buffer, and always uses the faster approach on the + first call. So the effects of the flush parameter in this implementation are + on the return value of inflate() as noted below, when inflate() returns early + when Z_BLOCK or Z_TREES is used, and when inflate() avoids the allocation of + memory for a sliding window when Z_FINISH is used. + + If a preset dictionary is needed after this call (see inflateSetDictionary + below), inflate sets strm->adler to the Adler-32 checksum of the dictionary + chosen by the compressor and returns Z_NEED_DICT; otherwise it sets + strm->adler to the Adler-32 checksum of all output produced so far (that is, + total_out bytes) and returns Z_OK, Z_STREAM_END or an error code as described + below. At the end of the stream, inflate() checks that its computed Adler-32 + checksum is equal to that saved by the compressor and returns Z_STREAM_END + only if the checksum is correct. + + inflate() can decompress and check either zlib-wrapped or gzip-wrapped + deflate data. The header type is detected automatically, if requested when + initializing with inflateInit2(). Any information contained in the gzip + header is not retained unless inflateGetHeader() is used. When processing + gzip-wrapped deflate data, strm->adler32 is set to the CRC-32 of the output + produced so far. The CRC-32 is checked against the gzip trailer, as is the + uncompressed length, modulo 2^32. + + inflate() returns Z_OK if some progress has been made (more input processed + or more output produced), Z_STREAM_END if the end of the compressed data has + been reached and all uncompressed output has been produced, Z_NEED_DICT if a + preset dictionary is needed at this point, Z_DATA_ERROR if the input data was + corrupted (input stream not conforming to the zlib format or incorrect check + value, in which case strm->msg points to a string with a more specific + error), Z_STREAM_ERROR if the stream structure was inconsistent (for example + next_in or next_out was Z_NULL, or the state was inadvertently written over + by the application), Z_MEM_ERROR if there was not enough memory, Z_BUF_ERROR + if no progress was possible or if there was not enough room in the output + buffer when Z_FINISH is used. Note that Z_BUF_ERROR is not fatal, and + inflate() can be called again with more input and more output space to + continue decompressing. If Z_DATA_ERROR is returned, the application may + then call inflateSync() to look for a good compression block if a partial + recovery of the data is to be attempted. +*/ + + +ZEXTERN int ZEXPORT inflateEnd(z_streamp strm); +/* + All dynamically allocated data structures for this stream are freed. + This function discards any unprocessed input and does not flush any pending + output. + + inflateEnd returns Z_OK if success, or Z_STREAM_ERROR if the stream state + was inconsistent. +*/ + + + /* Advanced functions */ + +/* + The following functions are needed only in some special applications. +*/ + +/* +ZEXTERN int ZEXPORT deflateInit2(z_streamp strm, + int level, + int method, + int windowBits, + int memLevel, + int strategy); + + This is another version of deflateInit with more compression options. The + fields zalloc, zfree and opaque must be initialized before by the caller. + + The method parameter is the compression method. It must be Z_DEFLATED in + this version of the library. + + The windowBits parameter is the base two logarithm of the window size + (the size of the history buffer). It should be in the range 8..15 for this + version of the library. Larger values of this parameter result in better + compression at the expense of memory usage. The default value is 15 if + deflateInit is used instead. + + For the current implementation of deflate(), a windowBits value of 8 (a + window size of 256 bytes) is not supported. As a result, a request for 8 + will result in 9 (a 512-byte window). In that case, providing 8 to + inflateInit2() will result in an error when the zlib header with 9 is + checked against the initialization of inflate(). The remedy is to not use 8 + with deflateInit2() with this initialization, or at least in that case use 9 + with inflateInit2(). + + windowBits can also be -8..-15 for raw deflate. In this case, -windowBits + determines the window size. deflate() will then generate raw deflate data + with no zlib header or trailer, and will not compute a check value. + + windowBits can also be greater than 15 for optional gzip encoding. Add + 16 to windowBits to write a simple gzip header and trailer around the + compressed data instead of a zlib wrapper. The gzip header will have no + file name, no extra data, no comment, no modification time (set to zero), no + header crc, and the operating system will be set to the appropriate value, + if the operating system was determined at compile time. If a gzip stream is + being written, strm->adler is a CRC-32 instead of an Adler-32. + + For raw deflate or gzip encoding, a request for a 256-byte window is + rejected as invalid, since only the zlib header provides a means of + transmitting the window size to the decompressor. + + The memLevel parameter specifies how much memory should be allocated + for the internal compression state. memLevel=1 uses minimum memory but is + slow and reduces compression ratio; memLevel=9 uses maximum memory for + optimal speed. The default value is 8. See zconf.h for total memory usage + as a function of windowBits and memLevel. + + The strategy parameter is used to tune the compression algorithm. Use the + value Z_DEFAULT_STRATEGY for normal data, Z_FILTERED for data produced by a + filter (or predictor), Z_HUFFMAN_ONLY to force Huffman encoding only (no + string match), or Z_RLE to limit match distances to one (run-length + encoding). Filtered data consists mostly of small values with a somewhat + random distribution. In this case, the compression algorithm is tuned to + compress them better. The effect of Z_FILTERED is to force more Huffman + coding and less string matching; it is somewhat intermediate between + Z_DEFAULT_STRATEGY and Z_HUFFMAN_ONLY. Z_RLE is designed to be almost as + fast as Z_HUFFMAN_ONLY, but give better compression for PNG image data. The + strategy parameter only affects the compression ratio but not the + correctness of the compressed output even if it is not set appropriately. + Z_FIXED prevents the use of dynamic Huffman codes, allowing for a simpler + decoder for special applications. + + deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_STREAM_ERROR if any parameter is invalid (such as an invalid + method), or Z_VERSION_ERROR if the zlib library version (zlib_version) is + incompatible with the version assumed by the caller (ZLIB_VERSION). msg is + set to null if there is no error message. deflateInit2 does not perform any + compression: this will be done by deflate(). +*/ + +ZEXTERN int ZEXPORT deflateSetDictionary(z_streamp strm, + const Bytef *dictionary, + uInt dictLength); +/* + Initializes the compression dictionary from the given byte sequence + without producing any compressed output. When using the zlib format, this + function must be called immediately after deflateInit, deflateInit2 or + deflateReset, and before any call of deflate. When doing raw deflate, this + function must be called either before any call of deflate, or immediately + after the completion of a deflate block, i.e. after all input has been + consumed and all output has been delivered when using any of the flush + options Z_BLOCK, Z_PARTIAL_FLUSH, Z_SYNC_FLUSH, or Z_FULL_FLUSH. The + compressor and decompressor must use exactly the same dictionary (see + inflateSetDictionary). + + The dictionary should consist of strings (byte sequences) that are likely + to be encountered later in the data to be compressed, with the most commonly + used strings preferably put towards the end of the dictionary. Using a + dictionary is most useful when the data to be compressed is short and can be + predicted with good accuracy; the data can then be compressed better than + with the default empty dictionary. + + Depending on the size of the compression data structures selected by + deflateInit or deflateInit2, a part of the dictionary may in effect be + discarded, for example if the dictionary is larger than the window size + provided in deflateInit or deflateInit2. Thus the strings most likely to be + useful should be put at the end of the dictionary, not at the front. In + addition, the current implementation of deflate will use at most the window + size minus 262 bytes of the provided dictionary. + + Upon return of this function, strm->adler is set to the Adler-32 value + of the dictionary; the decompressor may later use this value to determine + which dictionary has been used by the compressor. (The Adler-32 value + applies to the whole dictionary even if only a subset of the dictionary is + actually used by the compressor.) If a raw deflate was requested, then the + Adler-32 value is not computed and strm->adler is not set. + + deflateSetDictionary returns Z_OK if success, or Z_STREAM_ERROR if a + parameter is invalid (e.g. dictionary being Z_NULL) or the stream state is + inconsistent (for example if deflate has already been called for this stream + or if not at a block boundary for raw deflate). deflateSetDictionary does + not perform any compression: this will be done by deflate(). +*/ + +ZEXTERN int ZEXPORT deflateGetDictionary(z_streamp strm, + Bytef *dictionary, + uInt *dictLength); +/* + Returns the sliding dictionary being maintained by deflate. dictLength is + set to the number of bytes in the dictionary, and that many bytes are copied + to dictionary. dictionary must have enough space, where 32768 bytes is + always enough. If deflateGetDictionary() is called with dictionary equal to + Z_NULL, then only the dictionary length is returned, and nothing is copied. + Similarly, if dictLength is Z_NULL, then it is not set. + + deflateGetDictionary() may return a length less than the window size, even + when more than the window size in input has been provided. It may return up + to 258 bytes less in that case, due to how zlib's implementation of deflate + manages the sliding window and lookahead for matches, where matches can be + up to 258 bytes long. If the application needs the last window-size bytes of + input, then that would need to be saved by the application outside of zlib. + + deflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the + stream state is inconsistent. +*/ + +ZEXTERN int ZEXPORT deflateCopy(z_streamp dest, + z_streamp source); +/* + Sets the destination stream as a complete copy of the source stream. + + This function can be useful when several compression strategies will be + tried, for example when there are several ways of pre-processing the input + data with a filter. The streams that will be discarded should then be freed + by calling deflateEnd. Note that deflateCopy duplicates the internal + compression state which can be quite large, so this strategy is slow and can + consume lots of memory. + + deflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not + enough memory, Z_STREAM_ERROR if the source stream state was inconsistent + (such as zalloc being Z_NULL). msg is left unchanged in both source and + destination. +*/ + +ZEXTERN int ZEXPORT deflateReset(z_streamp strm); +/* + This function is equivalent to deflateEnd followed by deflateInit, but + does not free and reallocate the internal compression state. The stream + will leave the compression level and any other attributes that may have been + set unchanged. total_in, total_out, adler, and msg are initialized. + + deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent (such as zalloc or state being Z_NULL). +*/ + +ZEXTERN int ZEXPORT deflateParams(z_streamp strm, + int level, + int strategy); +/* + Dynamically update the compression level and compression strategy. The + interpretation of level and strategy is as in deflateInit2(). This can be + used to switch between compression and straight copy of the input data, or + to switch to a different kind of input data requiring a different strategy. + If the compression approach (which is a function of the level) or the + strategy is changed, and if there have been any deflate() calls since the + state was initialized or reset, then the input available so far is + compressed with the old level and strategy using deflate(strm, Z_BLOCK). + There are three approaches for the compression levels 0, 1..3, and 4..9 + respectively. The new level and strategy will take effect at the next call + of deflate(). + + If a deflate(strm, Z_BLOCK) is performed by deflateParams(), and it does + not have enough output space to complete, then the parameter change will not + take effect. In this case, deflateParams() can be called again with the + same parameters and more output space to try again. + + In order to assure a change in the parameters on the first try, the + deflate stream should be flushed using deflate() with Z_BLOCK or other flush + request until strm.avail_out is not zero, before calling deflateParams(). + Then no more input data should be provided before the deflateParams() call. + If this is done, the old level and strategy will be applied to the data + compressed before deflateParams(), and the new level and strategy will be + applied to the data compressed after deflateParams(). + + deflateParams returns Z_OK on success, Z_STREAM_ERROR if the source stream + state was inconsistent or if a parameter was invalid, or Z_BUF_ERROR if + there was not enough output space to complete the compression of the + available input data before a change in the strategy or approach. Note that + in the case of a Z_BUF_ERROR, the parameters are not changed. A return + value of Z_BUF_ERROR is not fatal, in which case deflateParams() can be + retried with more output space. +*/ + +ZEXTERN int ZEXPORT deflateTune(z_streamp strm, + int good_length, + int max_lazy, + int nice_length, + int max_chain); +/* + Fine tune deflate's internal compression parameters. This should only be + used by someone who understands the algorithm used by zlib's deflate for + searching for the best matching string, and even then only by the most + fanatic optimizer trying to squeeze out the last compressed bit for their + specific input data. Read the deflate.c source code for the meaning of the + max_lazy, good_length, nice_length, and max_chain parameters. + + deflateTune() can be called after deflateInit() or deflateInit2(), and + returns Z_OK on success, or Z_STREAM_ERROR for an invalid deflate stream. + */ + +ZEXTERN uLong ZEXPORT deflateBound(z_streamp strm, + uLong sourceLen); +/* + deflateBound() returns an upper bound on the compressed size after + deflation of sourceLen bytes. It must be called after deflateInit() or + deflateInit2(), and after deflateSetHeader(), if used. This would be used + to allocate an output buffer for deflation in a single pass, and so would be + called before deflate(). If that first deflate() call is provided the + sourceLen input bytes, an output buffer allocated to the size returned by + deflateBound(), and the flush value Z_FINISH, then deflate() is guaranteed + to return Z_STREAM_END. Note that it is possible for the compressed size to + be larger than the value returned by deflateBound() if flush options other + than Z_FINISH or Z_NO_FLUSH are used. +*/ + +ZEXTERN int ZEXPORT deflatePending(z_streamp strm, + unsigned *pending, + int *bits); +/* + deflatePending() returns the number of bytes and bits of output that have + been generated, but not yet provided in the available output. The bytes not + provided would be due to the available output space having being consumed. + The number of bits of output not provided are between 0 and 7, where they + await more bits to join them in order to fill out a full byte. If pending + or bits are Z_NULL, then those values are not set. + + deflatePending returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent. + */ + +ZEXTERN int ZEXPORT deflatePrime(z_streamp strm, + int bits, + int value); +/* + deflatePrime() inserts bits in the deflate output stream. The intent + is that this function is used to start off the deflate output with the bits + leftover from a previous deflate stream when appending to it. As such, this + function can only be used for raw deflate, and must be used before the first + deflate() call after a deflateInit2() or deflateReset(). bits must be less + than or equal to 16, and that many of the least significant bits of value + will be inserted in the output. + + deflatePrime returns Z_OK if success, Z_BUF_ERROR if there was not enough + room in the internal buffer to insert the bits, or Z_STREAM_ERROR if the + source stream state was inconsistent. +*/ + +ZEXTERN int ZEXPORT deflateSetHeader(z_streamp strm, + gz_headerp head); +/* + deflateSetHeader() provides gzip header information for when a gzip + stream is requested by deflateInit2(). deflateSetHeader() may be called + after deflateInit2() or deflateReset() and before the first call of + deflate(). The text, time, os, extra field, name, and comment information + in the provided gz_header structure are written to the gzip header (xflag is + ignored -- the extra flags are set according to the compression level). The + caller must assure that, if not Z_NULL, name and comment are terminated with + a zero byte, and that if extra is not Z_NULL, that extra_len bytes are + available there. If hcrc is true, a gzip header crc is included. Note that + the current versions of the command-line version of gzip (up through version + 1.3.x) do not support header crc's, and will report that it is a "multi-part + gzip file" and give up. + + If deflateSetHeader is not used, the default gzip header has text false, + the time set to zero, and os set to the current operating system, with no + extra, name, or comment fields. The gzip header is returned to the default + state by deflateReset(). + + deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent. +*/ + +/* +ZEXTERN int ZEXPORT inflateInit2(z_streamp strm, + int windowBits); + + This is another version of inflateInit with an extra parameter. The + fields next_in, avail_in, zalloc, zfree and opaque must be initialized + before by the caller. + + The windowBits parameter is the base two logarithm of the maximum window + size (the size of the history buffer). It should be in the range 8..15 for + this version of the library. The default value is 15 if inflateInit is used + instead. windowBits must be greater than or equal to the windowBits value + provided to deflateInit2() while compressing, or it must be equal to 15 if + deflateInit2() was not used. If a compressed stream with a larger window + size is given as input, inflate() will return with the error code + Z_DATA_ERROR instead of trying to allocate a larger window. + + windowBits can also be zero to request that inflate use the window size in + the zlib header of the compressed stream. + + windowBits can also be -8..-15 for raw inflate. In this case, -windowBits + determines the window size. inflate() will then process raw deflate data, + not looking for a zlib or gzip header, not generating a check value, and not + looking for any check values for comparison at the end of the stream. This + is for use with other formats that use the deflate compressed data format + such as zip. Those formats provide their own check values. If a custom + format is developed using the raw deflate format for compressed data, it is + recommended that a check value such as an Adler-32 or a CRC-32 be applied to + the uncompressed data as is done in the zlib, gzip, and zip formats. For + most applications, the zlib format should be used as is. Note that comments + above on the use in deflateInit2() applies to the magnitude of windowBits. + + windowBits can also be greater than 15 for optional gzip decoding. Add + 32 to windowBits to enable zlib and gzip decoding with automatic header + detection, or add 16 to decode only the gzip format (the zlib format will + return a Z_DATA_ERROR). If a gzip stream is being decoded, strm->adler is a + CRC-32 instead of an Adler-32. Unlike the gunzip utility and gzread() (see + below), inflate() will *not* automatically decode concatenated gzip members. + inflate() will return Z_STREAM_END at the end of the gzip member. The state + would need to be reset to continue decoding a subsequent gzip member. This + *must* be done if there is more data after a gzip member, in order for the + decompression to be compliant with the gzip standard (RFC 1952). + + inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_VERSION_ERROR if the zlib library version is incompatible with the + version assumed by the caller, or Z_STREAM_ERROR if the parameters are + invalid, such as a null pointer to the structure. msg is set to null if + there is no error message. inflateInit2 does not perform any decompression + apart from possibly reading the zlib header if present: actual decompression + will be done by inflate(). (So next_in and avail_in may be modified, but + next_out and avail_out are unused and unchanged.) The current implementation + of inflateInit2() does not process any header information -- that is + deferred until inflate() is called. +*/ + +ZEXTERN int ZEXPORT inflateSetDictionary(z_streamp strm, + const Bytef *dictionary, + uInt dictLength); +/* + Initializes the decompression dictionary from the given uncompressed byte + sequence. This function must be called immediately after a call of inflate, + if that call returned Z_NEED_DICT. The dictionary chosen by the compressor + can be determined from the Adler-32 value returned by that call of inflate. + The compressor and decompressor must use exactly the same dictionary (see + deflateSetDictionary). For raw inflate, this function can be called at any + time to set the dictionary. If the provided dictionary is smaller than the + window and there is already data in the window, then the provided dictionary + will amend what's there. The application must insure that the dictionary + that was used for compression is provided. + + inflateSetDictionary returns Z_OK if success, Z_STREAM_ERROR if a + parameter is invalid (e.g. dictionary being Z_NULL) or the stream state is + inconsistent, Z_DATA_ERROR if the given dictionary doesn't match the + expected one (incorrect Adler-32 value). inflateSetDictionary does not + perform any decompression: this will be done by subsequent calls of + inflate(). +*/ + +ZEXTERN int ZEXPORT inflateGetDictionary(z_streamp strm, + Bytef *dictionary, + uInt *dictLength); +/* + Returns the sliding dictionary being maintained by inflate. dictLength is + set to the number of bytes in the dictionary, and that many bytes are copied + to dictionary. dictionary must have enough space, where 32768 bytes is + always enough. If inflateGetDictionary() is called with dictionary equal to + Z_NULL, then only the dictionary length is returned, and nothing is copied. + Similarly, if dictLength is Z_NULL, then it is not set. + + inflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the + stream state is inconsistent. +*/ + +ZEXTERN int ZEXPORT inflateSync(z_streamp strm); +/* + Skips invalid compressed data until a possible full flush point (see above + for the description of deflate with Z_FULL_FLUSH) can be found, or until all + available input is skipped. No output is provided. + + inflateSync searches for a 00 00 FF FF pattern in the compressed data. + All full flush points have this pattern, but not all occurrences of this + pattern are full flush points. + + inflateSync returns Z_OK if a possible full flush point has been found, + Z_BUF_ERROR if no more input was provided, Z_DATA_ERROR if no flush point + has been found, or Z_STREAM_ERROR if the stream structure was inconsistent. + In the success case, the application may save the current value of total_in + which indicates where valid compressed data was found. In the error case, + the application may repeatedly call inflateSync, providing more input each + time, until success or end of the input data. +*/ + +ZEXTERN int ZEXPORT inflateCopy(z_streamp dest, + z_streamp source); +/* + Sets the destination stream as a complete copy of the source stream. + + This function can be useful when randomly accessing a large stream. The + first pass through the stream can periodically record the inflate state, + allowing restarting inflate at those points when randomly accessing the + stream. + + inflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not + enough memory, Z_STREAM_ERROR if the source stream state was inconsistent + (such as zalloc being Z_NULL). msg is left unchanged in both source and + destination. +*/ + +ZEXTERN int ZEXPORT inflateReset(z_streamp strm); +/* + This function is equivalent to inflateEnd followed by inflateInit, + but does not free and reallocate the internal decompression state. The + stream will keep attributes that may have been set by inflateInit2. + total_in, total_out, adler, and msg are initialized. + + inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent (such as zalloc or state being Z_NULL). +*/ + +ZEXTERN int ZEXPORT inflateReset2(z_streamp strm, + int windowBits); +/* + This function is the same as inflateReset, but it also permits changing + the wrap and window size requests. The windowBits parameter is interpreted + the same as it is for inflateInit2. If the window size is changed, then the + memory allocated for the window is freed, and the window will be reallocated + by inflate() if needed. + + inflateReset2 returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent (such as zalloc or state being Z_NULL), or if + the windowBits parameter is invalid. +*/ + +ZEXTERN int ZEXPORT inflatePrime(z_streamp strm, + int bits, + int value); +/* + This function inserts bits in the inflate input stream. The intent is + that this function is used to start inflating at a bit position in the + middle of a byte. The provided bits will be used before any bytes are used + from next_in. This function should only be used with raw inflate, and + should be used before the first inflate() call after inflateInit2() or + inflateReset(). bits must be less than or equal to 16, and that many of the + least significant bits of value will be inserted in the input. + + If bits is negative, then the input stream bit buffer is emptied. Then + inflatePrime() can be called again to put bits in the buffer. This is used + to clear out bits leftover after feeding inflate a block description prior + to feeding inflate codes. + + inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent. +*/ + +ZEXTERN long ZEXPORT inflateMark(z_streamp strm); +/* + This function returns two values, one in the lower 16 bits of the return + value, and the other in the remaining upper bits, obtained by shifting the + return value down 16 bits. If the upper value is -1 and the lower value is + zero, then inflate() is currently decoding information outside of a block. + If the upper value is -1 and the lower value is non-zero, then inflate is in + the middle of a stored block, with the lower value equaling the number of + bytes from the input remaining to copy. If the upper value is not -1, then + it is the number of bits back from the current bit position in the input of + the code (literal or length/distance pair) currently being processed. In + that case the lower value is the number of bytes already emitted for that + code. + + A code is being processed if inflate is waiting for more input to complete + decoding of the code, or if it has completed decoding but is waiting for + more output space to write the literal or match data. + + inflateMark() is used to mark locations in the input data for random + access, which may be at bit positions, and to note those cases where the + output of a code may span boundaries of random access blocks. The current + location in the input stream can be determined from avail_in and data_type + as noted in the description for the Z_BLOCK flush parameter for inflate. + + inflateMark returns the value noted above, or -65536 if the provided + source stream state was inconsistent. +*/ + +ZEXTERN int ZEXPORT inflateGetHeader(z_streamp strm, + gz_headerp head); +/* + inflateGetHeader() requests that gzip header information be stored in the + provided gz_header structure. inflateGetHeader() may be called after + inflateInit2() or inflateReset(), and before the first call of inflate(). + As inflate() processes the gzip stream, head->done is zero until the header + is completed, at which time head->done is set to one. If a zlib stream is + being decoded, then head->done is set to -1 to indicate that there will be + no gzip header information forthcoming. Note that Z_BLOCK or Z_TREES can be + used to force inflate() to return immediately after header processing is + complete and before any actual data is decompressed. + + The text, time, xflags, and os fields are filled in with the gzip header + contents. hcrc is set to true if there is a header CRC. (The header CRC + was valid if done is set to one.) If extra is not Z_NULL, then extra_max + contains the maximum number of bytes to write to extra. Once done is true, + extra_len contains the actual extra field length, and extra contains the + extra field, or that field truncated if extra_max is less than extra_len. + If name is not Z_NULL, then up to name_max characters are written there, + terminated with a zero unless the length is greater than name_max. If + comment is not Z_NULL, then up to comm_max characters are written there, + terminated with a zero unless the length is greater than comm_max. When any + of extra, name, or comment are not Z_NULL and the respective field is not + present in the header, then that field is set to Z_NULL to signal its + absence. This allows the use of deflateSetHeader() with the returned + structure to duplicate the header. However if those fields are set to + allocated memory, then the application will need to save those pointers + elsewhere so that they can be eventually freed. + + If inflateGetHeader is not used, then the header information is simply + discarded. The header is always checked for validity, including the header + CRC if present. inflateReset() will reset the process to discard the header + information. The application would need to call inflateGetHeader() again to + retrieve the header from the next gzip stream. + + inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source + stream state was inconsistent. +*/ + +/* +ZEXTERN int ZEXPORT inflateBackInit(z_streamp strm, int windowBits, + unsigned char FAR *window); + + Initialize the internal stream state for decompression using inflateBack() + calls. The fields zalloc, zfree and opaque in strm must be initialized + before the call. If zalloc and zfree are Z_NULL, then the default library- + derived memory allocation routines are used. windowBits is the base two + logarithm of the window size, in the range 8..15. window is a caller + supplied buffer of that size. Except for special applications where it is + assured that deflate was used with small window sizes, windowBits must be 15 + and a 32K byte window must be supplied to be able to decompress general + deflate streams. + + See inflateBack() for the usage of these routines. + + inflateBackInit will return Z_OK on success, Z_STREAM_ERROR if any of + the parameters are invalid, Z_MEM_ERROR if the internal state could not be + allocated, or Z_VERSION_ERROR if the version of the library does not match + the version of the header file. +*/ + +typedef unsigned (*in_func)(void FAR *, + z_const unsigned char FAR * FAR *); +typedef int (*out_func)(void FAR *, unsigned char FAR *, unsigned); + +ZEXTERN int ZEXPORT inflateBack(z_streamp strm, + in_func in, void FAR *in_desc, + out_func out, void FAR *out_desc); +/* + inflateBack() does a raw inflate with a single call using a call-back + interface for input and output. This is potentially more efficient than + inflate() for file i/o applications, in that it avoids copying between the + output and the sliding window by simply making the window itself the output + buffer. inflate() can be faster on modern CPUs when used with large + buffers. inflateBack() trusts the application to not change the output + buffer passed by the output function, at least until inflateBack() returns. + + inflateBackInit() must be called first to allocate the internal state + and to initialize the state with the user-provided window buffer. + inflateBack() may then be used multiple times to inflate a complete, raw + deflate stream with each call. inflateBackEnd() is then called to free the + allocated state. + + A raw deflate stream is one with no zlib or gzip header or trailer. + This routine would normally be used in a utility that reads zip or gzip + files and writes out uncompressed files. The utility would decode the + header and process the trailer on its own, hence this routine expects only + the raw deflate stream to decompress. This is different from the default + behavior of inflate(), which expects a zlib header and trailer around the + deflate stream. + + inflateBack() uses two subroutines supplied by the caller that are then + called by inflateBack() for input and output. inflateBack() calls those + routines until it reads a complete deflate stream and writes out all of the + uncompressed data, or until it encounters an error. The function's + parameters and return types are defined above in the in_func and out_func + typedefs. inflateBack() will call in(in_desc, &buf) which should return the + number of bytes of provided input, and a pointer to that input in buf. If + there is no input available, in() must return zero -- buf is ignored in that + case -- and inflateBack() will return a buffer error. inflateBack() will + call out(out_desc, buf, len) to write the uncompressed data buf[0..len-1]. + out() should return zero on success, or non-zero on failure. If out() + returns non-zero, inflateBack() will return with an error. Neither in() nor + out() are permitted to change the contents of the window provided to + inflateBackInit(), which is also the buffer that out() uses to write from. + The length written by out() will be at most the window size. Any non-zero + amount of input may be provided by in(). + + For convenience, inflateBack() can be provided input on the first call by + setting strm->next_in and strm->avail_in. If that input is exhausted, then + in() will be called. Therefore strm->next_in must be initialized before + calling inflateBack(). If strm->next_in is Z_NULL, then in() will be called + immediately for input. If strm->next_in is not Z_NULL, then strm->avail_in + must also be initialized, and then if strm->avail_in is not zero, input will + initially be taken from strm->next_in[0 .. strm->avail_in - 1]. + + The in_desc and out_desc parameters of inflateBack() is passed as the + first parameter of in() and out() respectively when they are called. These + descriptors can be optionally used to pass any information that the caller- + supplied in() and out() functions need to do their job. + + On return, inflateBack() will set strm->next_in and strm->avail_in to + pass back any unused input that was provided by the last in() call. The + return values of inflateBack() can be Z_STREAM_END on success, Z_BUF_ERROR + if in() or out() returned an error, Z_DATA_ERROR if there was a format error + in the deflate stream (in which case strm->msg is set to indicate the nature + of the error), or Z_STREAM_ERROR if the stream was not properly initialized. + In the case of Z_BUF_ERROR, an input or output error can be distinguished + using strm->next_in which will be Z_NULL only if in() returned an error. If + strm->next_in is not Z_NULL, then the Z_BUF_ERROR was due to out() returning + non-zero. (in() will always be called before out(), so strm->next_in is + assured to be defined if out() returns non-zero.) Note that inflateBack() + cannot return Z_OK. +*/ + +ZEXTERN int ZEXPORT inflateBackEnd(z_streamp strm); +/* + All memory allocated by inflateBackInit() is freed. + + inflateBackEnd() returns Z_OK on success, or Z_STREAM_ERROR if the stream + state was inconsistent. +*/ + +ZEXTERN uLong ZEXPORT zlibCompileFlags(void); +/* Return flags indicating compile-time options. + + Type sizes, two bits each, 00 = 16 bits, 01 = 32, 10 = 64, 11 = other: + 1.0: size of uInt + 3.2: size of uLong + 5.4: size of voidpf (pointer) + 7.6: size of z_off_t + + Compiler, assembler, and debug options: + 8: ZLIB_DEBUG + 9: ASMV or ASMINF -- use ASM code + 10: ZLIB_WINAPI -- exported functions use the WINAPI calling convention + 11: 0 (reserved) + + One-time table building (smaller code, but not thread-safe if true): + 12: BUILDFIXED -- build static block decoding tables when needed + 13: DYNAMIC_CRC_TABLE -- build CRC calculation tables when needed + 14,15: 0 (reserved) + + Library content (indicates missing functionality): + 16: NO_GZCOMPRESS -- gz* functions cannot compress (to avoid linking + deflate code when not needed) + 17: NO_GZIP -- deflate can't write gzip streams, and inflate can't detect + and decode gzip streams (to avoid linking crc code) + 18-19: 0 (reserved) + + Operation variations (changes in library functionality): + 20: PKZIP_BUG_WORKAROUND -- slightly more permissive inflate + 21: FASTEST -- deflate algorithm with only one, lowest compression level + 22,23: 0 (reserved) + + The sprintf variant used by gzprintf (zero is best): + 24: 0 = vs*, 1 = s* -- 1 means limited to 20 arguments after the format + 25: 0 = *nprintf, 1 = *printf -- 1 means gzprintf() not secure! + 26: 0 = returns value, 1 = void -- 1 means inferred string length returned + + Remainder: + 27-31: 0 (reserved) + */ + +#ifndef Z_SOLO + + /* utility functions */ + +/* + The following utility functions are implemented on top of the basic + stream-oriented functions. To simplify the interface, some default options + are assumed (compression level and memory usage, standard memory allocation + functions). The source code of these utility functions can be modified if + you need special options. +*/ + +ZEXTERN int ZEXPORT compress(Bytef *dest, uLongf *destLen, + const Bytef *source, uLong sourceLen); +/* + Compresses the source buffer into the destination buffer. sourceLen is + the byte length of the source buffer. Upon entry, destLen is the total size + of the destination buffer, which must be at least the value returned by + compressBound(sourceLen). Upon exit, destLen is the actual size of the + compressed data. compress() is equivalent to compress2() with a level + parameter of Z_DEFAULT_COMPRESSION. + + compress returns Z_OK if success, Z_MEM_ERROR if there was not + enough memory, Z_BUF_ERROR if there was not enough room in the output + buffer. +*/ + +ZEXTERN int ZEXPORT compress2(Bytef *dest, uLongf *destLen, + const Bytef *source, uLong sourceLen, + int level); +/* + Compresses the source buffer into the destination buffer. The level + parameter has the same meaning as in deflateInit. sourceLen is the byte + length of the source buffer. Upon entry, destLen is the total size of the + destination buffer, which must be at least the value returned by + compressBound(sourceLen). Upon exit, destLen is the actual size of the + compressed data. + + compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_BUF_ERROR if there was not enough room in the output buffer, + Z_STREAM_ERROR if the level parameter is invalid. +*/ + +ZEXTERN uLong ZEXPORT compressBound(uLong sourceLen); +/* + compressBound() returns an upper bound on the compressed size after + compress() or compress2() on sourceLen bytes. It would be used before a + compress() or compress2() call to allocate the destination buffer. +*/ + +ZEXTERN int ZEXPORT uncompress(Bytef *dest, uLongf *destLen, + const Bytef *source, uLong sourceLen); +/* + Decompresses the source buffer into the destination buffer. sourceLen is + the byte length of the source buffer. Upon entry, destLen is the total size + of the destination buffer, which must be large enough to hold the entire + uncompressed data. (The size of the uncompressed data must have been saved + previously by the compressor and transmitted to the decompressor by some + mechanism outside the scope of this compression library.) Upon exit, destLen + is the actual size of the uncompressed data. + + uncompress returns Z_OK if success, Z_MEM_ERROR if there was not + enough memory, Z_BUF_ERROR if there was not enough room in the output + buffer, or Z_DATA_ERROR if the input data was corrupted or incomplete. In + the case where there is not enough room, uncompress() will fill the output + buffer with the uncompressed data up to that point. +*/ + +ZEXTERN int ZEXPORT uncompress2(Bytef *dest, uLongf *destLen, + const Bytef *source, uLong *sourceLen); +/* + Same as uncompress, except that sourceLen is a pointer, where the + length of the source is *sourceLen. On return, *sourceLen is the number of + source bytes consumed. +*/ + + /* gzip file access functions */ + +/* + This library supports reading and writing files in gzip (.gz) format with + an interface similar to that of stdio, using the functions that start with + "gz". The gzip format is different from the zlib format. gzip is a gzip + wrapper, documented in RFC 1952, wrapped around a deflate stream. +*/ + +typedef struct gzFile_s *gzFile; /* semi-opaque gzip file descriptor */ + +/* +ZEXTERN gzFile ZEXPORT gzopen(const char *path, const char *mode); + + Open the gzip (.gz) file at path for reading and decompressing, or + compressing and writing. The mode parameter is as in fopen ("rb" or "wb") + but can also include a compression level ("wb9") or a strategy: 'f' for + filtered data as in "wb6f", 'h' for Huffman-only compression as in "wb1h", + 'R' for run-length encoding as in "wb1R", or 'F' for fixed code compression + as in "wb9F". (See the description of deflateInit2 for more information + about the strategy parameter.) 'T' will request transparent writing or + appending with no compression and not using the gzip format. + + "a" can be used instead of "w" to request that the gzip stream that will + be written be appended to the file. "+" will result in an error, since + reading and writing to the same gzip file is not supported. The addition of + "x" when writing will create the file exclusively, which fails if the file + already exists. On systems that support it, the addition of "e" when + reading or writing will set the flag to close the file on an execve() call. + + These functions, as well as gzip, will read and decode a sequence of gzip + streams in a file. The append function of gzopen() can be used to create + such a file. (Also see gzflush() for another way to do this.) When + appending, gzopen does not test whether the file begins with a gzip stream, + nor does it look for the end of the gzip streams to begin appending. gzopen + will simply append a gzip stream to the existing file. + + gzopen can be used to read a file which is not in gzip format; in this + case gzread will directly read from the file without decompression. When + reading, this will be detected automatically by looking for the magic two- + byte gzip header. + + gzopen returns NULL if the file could not be opened, if there was + insufficient memory to allocate the gzFile state, or if an invalid mode was + specified (an 'r', 'w', or 'a' was not provided, or '+' was provided). + errno can be checked to determine if the reason gzopen failed was that the + file could not be opened. +*/ + +ZEXTERN gzFile ZEXPORT gzdopen(int fd, const char *mode); +/* + Associate a gzFile with the file descriptor fd. File descriptors are + obtained from calls like open, dup, creat, pipe or fileno (if the file has + been previously opened with fopen). The mode parameter is as in gzopen. + + The next call of gzclose on the returned gzFile will also close the file + descriptor fd, just like fclose(fdopen(fd, mode)) closes the file descriptor + fd. If you want to keep fd open, use fd = dup(fd_keep); gz = gzdopen(fd, + mode);. The duplicated descriptor should be saved to avoid a leak, since + gzdopen does not close fd if it fails. If you are using fileno() to get the + file descriptor from a FILE *, then you will have to use dup() to avoid + double-close()ing the file descriptor. Both gzclose() and fclose() will + close the associated file descriptor, so they need to have different file + descriptors. + + gzdopen returns NULL if there was insufficient memory to allocate the + gzFile state, if an invalid mode was specified (an 'r', 'w', or 'a' was not + provided, or '+' was provided), or if fd is -1. The file descriptor is not + used until the next gz* read, write, seek, or close operation, so gzdopen + will not detect if fd is invalid (unless fd is -1). +*/ + +ZEXTERN int ZEXPORT gzbuffer(gzFile file, unsigned size); +/* + Set the internal buffer size used by this library's functions for file to + size. The default buffer size is 8192 bytes. This function must be called + after gzopen() or gzdopen(), and before any other calls that read or write + the file. The buffer memory allocation is always deferred to the first read + or write. Three times that size in buffer space is allocated. A larger + buffer size of, for example, 64K or 128K bytes will noticeably increase the + speed of decompression (reading). + + The new buffer size also affects the maximum length for gzprintf(). + + gzbuffer() returns 0 on success, or -1 on failure, such as being called + too late. +*/ + +ZEXTERN int ZEXPORT gzsetparams(gzFile file, int level, int strategy); +/* + Dynamically update the compression level and strategy for file. See the + description of deflateInit2 for the meaning of these parameters. Previously + provided data is flushed before applying the parameter changes. + + gzsetparams returns Z_OK if success, Z_STREAM_ERROR if the file was not + opened for writing, Z_ERRNO if there is an error writing the flushed data, + or Z_MEM_ERROR if there is a memory allocation error. +*/ + +ZEXTERN int ZEXPORT gzread(gzFile file, voidp buf, unsigned len); +/* + Read and decompress up to len uncompressed bytes from file into buf. If + the input file is not in gzip format, gzread copies the given number of + bytes into the buffer directly from the file. + + After reaching the end of a gzip stream in the input, gzread will continue + to read, looking for another gzip stream. Any number of gzip streams may be + concatenated in the input file, and will all be decompressed by gzread(). + If something other than a gzip stream is encountered after a gzip stream, + that remaining trailing garbage is ignored (and no error is returned). + + gzread can be used to read a gzip file that is being concurrently written. + Upon reaching the end of the input, gzread will return with the available + data. If the error code returned by gzerror is Z_OK or Z_BUF_ERROR, then + gzclearerr can be used to clear the end of file indicator in order to permit + gzread to be tried again. Z_OK indicates that a gzip stream was completed + on the last gzread. Z_BUF_ERROR indicates that the input file ended in the + middle of a gzip stream. Note that gzread does not return -1 in the event + of an incomplete gzip stream. This error is deferred until gzclose(), which + will return Z_BUF_ERROR if the last gzread ended in the middle of a gzip + stream. Alternatively, gzerror can be used before gzclose to detect this + case. + + gzread returns the number of uncompressed bytes actually read, less than + len for end of file, or -1 for error. If len is too large to fit in an int, + then nothing is read, -1 is returned, and the error state is set to + Z_STREAM_ERROR. +*/ + +ZEXTERN z_size_t ZEXPORT gzfread(voidp buf, z_size_t size, z_size_t nitems, + gzFile file); +/* + Read and decompress up to nitems items of size size from file into buf, + otherwise operating as gzread() does. This duplicates the interface of + stdio's fread(), with size_t request and return types. If the library + defines size_t, then z_size_t is identical to size_t. If not, then z_size_t + is an unsigned integer type that can contain a pointer. + + gzfread() returns the number of full items read of size size, or zero if + the end of the file was reached and a full item could not be read, or if + there was an error. gzerror() must be consulted if zero is returned in + order to determine if there was an error. If the multiplication of size and + nitems overflows, i.e. the product does not fit in a z_size_t, then nothing + is read, zero is returned, and the error state is set to Z_STREAM_ERROR. + + In the event that the end of file is reached and only a partial item is + available at the end, i.e. the remaining uncompressed data length is not a + multiple of size, then the final partial item is nevertheless read into buf + and the end-of-file flag is set. The length of the partial item read is not + provided, but could be inferred from the result of gztell(). This behavior + is the same as the behavior of fread() implementations in common libraries, + but it prevents the direct use of gzfread() to read a concurrently written + file, resetting and retrying on end-of-file, when size is not 1. +*/ + +ZEXTERN int ZEXPORT gzwrite(gzFile file, voidpc buf, unsigned len); +/* + Compress and write the len uncompressed bytes at buf to file. gzwrite + returns the number of uncompressed bytes written or 0 in case of error. +*/ + +ZEXTERN z_size_t ZEXPORT gzfwrite(voidpc buf, z_size_t size, + z_size_t nitems, gzFile file); +/* + Compress and write nitems items of size size from buf to file, duplicating + the interface of stdio's fwrite(), with size_t request and return types. If + the library defines size_t, then z_size_t is identical to size_t. If not, + then z_size_t is an unsigned integer type that can contain a pointer. + + gzfwrite() returns the number of full items written of size size, or zero + if there was an error. If the multiplication of size and nitems overflows, + i.e. the product does not fit in a z_size_t, then nothing is written, zero + is returned, and the error state is set to Z_STREAM_ERROR. +*/ + +ZEXTERN int ZEXPORTVA gzprintf(gzFile file, const char *format, ...); +/* + Convert, format, compress, and write the arguments (...) to file under + control of the string format, as in fprintf. gzprintf returns the number of + uncompressed bytes actually written, or a negative zlib error code in case + of error. The number of uncompressed bytes written is limited to 8191, or + one less than the buffer size given to gzbuffer(). The caller should assure + that this limit is not exceeded. If it is exceeded, then gzprintf() will + return an error (0) with nothing written. In this case, there may also be a + buffer overflow with unpredictable consequences, which is possible only if + zlib was compiled with the insecure functions sprintf() or vsprintf(), + because the secure snprintf() or vsnprintf() functions were not available. + This can be determined using zlibCompileFlags(). +*/ + +ZEXTERN int ZEXPORT gzputs(gzFile file, const char *s); +/* + Compress and write the given null-terminated string s to file, excluding + the terminating null character. + + gzputs returns the number of characters written, or -1 in case of error. +*/ + +ZEXTERN char * ZEXPORT gzgets(gzFile file, char *buf, int len); +/* + Read and decompress bytes from file into buf, until len-1 characters are + read, or until a newline character is read and transferred to buf, or an + end-of-file condition is encountered. If any characters are read or if len + is one, the string is terminated with a null character. If no characters + are read due to an end-of-file or len is less than one, then the buffer is + left untouched. + + gzgets returns buf which is a null-terminated string, or it returns NULL + for end-of-file or in case of error. If there was an error, the contents at + buf are indeterminate. +*/ + +ZEXTERN int ZEXPORT gzputc(gzFile file, int c); +/* + Compress and write c, converted to an unsigned char, into file. gzputc + returns the value that was written, or -1 in case of error. +*/ + +ZEXTERN int ZEXPORT gzgetc(gzFile file); +/* + Read and decompress one byte from file. gzgetc returns this byte or -1 + in case of end of file or error. This is implemented as a macro for speed. + As such, it does not do all of the checking the other functions do. I.e. + it does not check to see if file is NULL, nor whether the structure file + points to has been clobbered or not. +*/ + +ZEXTERN int ZEXPORT gzungetc(int c, gzFile file); +/* + Push c back onto the stream for file to be read as the first character on + the next read. At least one character of push-back is always allowed. + gzungetc() returns the character pushed, or -1 on failure. gzungetc() will + fail if c is -1, and may fail if a character has been pushed but not read + yet. If gzungetc is used immediately after gzopen or gzdopen, at least the + output buffer size of pushed characters is allowed. (See gzbuffer above.) + The pushed character will be discarded if the stream is repositioned with + gzseek() or gzrewind(). +*/ + +ZEXTERN int ZEXPORT gzflush(gzFile file, int flush); +/* + Flush all pending output to file. The parameter flush is as in the + deflate() function. The return value is the zlib error number (see function + gzerror below). gzflush is only permitted when writing. + + If the flush parameter is Z_FINISH, the remaining data is written and the + gzip stream is completed in the output. If gzwrite() is called again, a new + gzip stream will be started in the output. gzread() is able to read such + concatenated gzip streams. + + gzflush should be called only when strictly necessary because it will + degrade compression if called too often. +*/ + +/* +ZEXTERN z_off_t ZEXPORT gzseek(gzFile file, + z_off_t offset, int whence); + + Set the starting position to offset relative to whence for the next gzread + or gzwrite on file. The offset represents a number of bytes in the + uncompressed data stream. The whence parameter is defined as in lseek(2); + the value SEEK_END is not supported. + + If the file is opened for reading, this function is emulated but can be + extremely slow. If the file is opened for writing, only forward seeks are + supported; gzseek then compresses a sequence of zeroes up to the new + starting position. + + gzseek returns the resulting offset location as measured in bytes from + the beginning of the uncompressed stream, or -1 in case of error, in + particular if the file is opened for writing and the new starting position + would be before the current position. +*/ + +ZEXTERN int ZEXPORT gzrewind(gzFile file); +/* + Rewind file. This function is supported only for reading. + + gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET). +*/ + +/* +ZEXTERN z_off_t ZEXPORT gztell(gzFile file); + + Return the starting position for the next gzread or gzwrite on file. + This position represents a number of bytes in the uncompressed data stream, + and is zero when starting, even if appending or reading a gzip stream from + the middle of a file using gzdopen(). + + gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR) +*/ + +/* +ZEXTERN z_off_t ZEXPORT gzoffset(gzFile file); + + Return the current compressed (actual) read or write offset of file. This + offset includes the count of bytes that precede the gzip stream, for example + when appending or when using gzdopen() for reading. When reading, the + offset does not include as yet unused buffered input. This information can + be used for a progress indicator. On error, gzoffset() returns -1. +*/ + +ZEXTERN int ZEXPORT gzeof(gzFile file); +/* + Return true (1) if the end-of-file indicator for file has been set while + reading, false (0) otherwise. Note that the end-of-file indicator is set + only if the read tried to go past the end of the input, but came up short. + Therefore, just like feof(), gzeof() may return false even if there is no + more data to read, in the event that the last read request was for the exact + number of bytes remaining in the input file. This will happen if the input + file size is an exact multiple of the buffer size. + + If gzeof() returns true, then the read functions will return no more data, + unless the end-of-file indicator is reset by gzclearerr() and the input file + has grown since the previous end of file was detected. +*/ + +ZEXTERN int ZEXPORT gzdirect(gzFile file); +/* + Return true (1) if file is being copied directly while reading, or false + (0) if file is a gzip stream being decompressed. + + If the input file is empty, gzdirect() will return true, since the input + does not contain a gzip stream. + + If gzdirect() is used immediately after gzopen() or gzdopen() it will + cause buffers to be allocated to allow reading the file to determine if it + is a gzip file. Therefore if gzbuffer() is used, it should be called before + gzdirect(). + + When writing, gzdirect() returns true (1) if transparent writing was + requested ("wT" for the gzopen() mode), or false (0) otherwise. (Note: + gzdirect() is not needed when writing. Transparent writing must be + explicitly requested, so the application already knows the answer. When + linking statically, using gzdirect() will include all of the zlib code for + gzip file reading and decompression, which may not be desired.) +*/ + +ZEXTERN int ZEXPORT gzclose(gzFile file); +/* + Flush all pending output for file, if necessary, close file and + deallocate the (de)compression state. Note that once file is closed, you + cannot call gzerror with file, since its structures have been deallocated. + gzclose must not be called more than once on the same file, just as free + must not be called more than once on the same allocation. + + gzclose will return Z_STREAM_ERROR if file is not valid, Z_ERRNO on a + file operation error, Z_MEM_ERROR if out of memory, Z_BUF_ERROR if the + last read ended in the middle of a gzip stream, or Z_OK on success. +*/ + +ZEXTERN int ZEXPORT gzclose_r(gzFile file); +ZEXTERN int ZEXPORT gzclose_w(gzFile file); +/* + Same as gzclose(), but gzclose_r() is only for use when reading, and + gzclose_w() is only for use when writing or appending. The advantage to + using these instead of gzclose() is that they avoid linking in zlib + compression or decompression code that is not used when only reading or only + writing respectively. If gzclose() is used, then both compression and + decompression code will be included the application when linking to a static + zlib library. +*/ + +ZEXTERN const char * ZEXPORT gzerror(gzFile file, int *errnum); +/* + Return the error message for the last error which occurred on file. + errnum is set to zlib error number. If an error occurred in the file system + and not in the compression library, errnum is set to Z_ERRNO and the + application may consult errno to get the exact error code. + + The application must not modify the returned string. Future calls to + this function may invalidate the previously returned string. If file is + closed, then the string previously returned by gzerror will no longer be + available. + + gzerror() should be used to distinguish errors from end-of-file for those + functions above that do not distinguish those cases in their return values. +*/ + +ZEXTERN void ZEXPORT gzclearerr(gzFile file); +/* + Clear the error and end-of-file flags for file. This is analogous to the + clearerr() function in stdio. This is useful for continuing to read a gzip + file that is being written concurrently. +*/ + +#endif /* !Z_SOLO */ + + /* checksum functions */ + +/* + These functions are not related to compression but are exported + anyway because they might be useful in applications using the compression + library. +*/ + +ZEXTERN uLong ZEXPORT adler32(uLong adler, const Bytef *buf, uInt len); +/* + Update a running Adler-32 checksum with the bytes buf[0..len-1] and + return the updated checksum. An Adler-32 value is in the range of a 32-bit + unsigned integer. If buf is Z_NULL, this function returns the required + initial value for the checksum. + + An Adler-32 checksum is almost as reliable as a CRC-32 but can be computed + much faster. + + Usage example: + + uLong adler = adler32(0L, Z_NULL, 0); + + while (read_buffer(buffer, length) != EOF) { + adler = adler32(adler, buffer, length); + } + if (adler != original_adler) error(); +*/ + +ZEXTERN uLong ZEXPORT adler32_z(uLong adler, const Bytef *buf, + z_size_t len); +/* + Same as adler32(), but with a size_t length. +*/ + +/* +ZEXTERN uLong ZEXPORT adler32_combine(uLong adler1, uLong adler2, + z_off_t len2); + + Combine two Adler-32 checksums into one. For two sequences of bytes, seq1 + and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for + each, adler1 and adler2. adler32_combine() returns the Adler-32 checksum of + seq1 and seq2 concatenated, requiring only adler1, adler2, and len2. Note + that the z_off_t type (like off_t) is a signed integer. If len2 is + negative, the result has no meaning or utility. +*/ + +ZEXTERN uLong ZEXPORT crc32(uLong crc, const Bytef *buf, uInt len); +/* + Update a running CRC-32 with the bytes buf[0..len-1] and return the + updated CRC-32. A CRC-32 value is in the range of a 32-bit unsigned integer. + If buf is Z_NULL, this function returns the required initial value for the + crc. Pre- and post-conditioning (one's complement) is performed within this + function so it shouldn't be done by the application. + + Usage example: + + uLong crc = crc32(0L, Z_NULL, 0); + + while (read_buffer(buffer, length) != EOF) { + crc = crc32(crc, buffer, length); + } + if (crc != original_crc) error(); +*/ + +ZEXTERN uLong ZEXPORT crc32_z(uLong crc, const Bytef *buf, + z_size_t len); +/* + Same as crc32(), but with a size_t length. +*/ + +/* +ZEXTERN uLong ZEXPORT crc32_combine(uLong crc1, uLong crc2, z_off_t len2); + + Combine two CRC-32 check values into one. For two sequences of bytes, + seq1 and seq2 with lengths len1 and len2, CRC-32 check values were + calculated for each, crc1 and crc2. crc32_combine() returns the CRC-32 + check value of seq1 and seq2 concatenated, requiring only crc1, crc2, and + len2. len2 must be non-negative. +*/ + +/* +ZEXTERN uLong ZEXPORT crc32_combine_gen(z_off_t len2); + + Return the operator corresponding to length len2, to be used with + crc32_combine_op(). len2 must be non-negative. +*/ + +ZEXTERN uLong ZEXPORT crc32_combine_op(uLong crc1, uLong crc2, uLong op); +/* + Give the same result as crc32_combine(), using op in place of len2. op is + is generated from len2 by crc32_combine_gen(). This will be faster than + crc32_combine() if the generated op is used more than once. +*/ + + + /* various hacks, don't look :) */ + +/* deflateInit and inflateInit are macros to allow checking the zlib version + * and the compiler's view of z_stream: + */ +ZEXTERN int ZEXPORT deflateInit_(z_streamp strm, int level, + const char *version, int stream_size); +ZEXTERN int ZEXPORT inflateInit_(z_streamp strm, + const char *version, int stream_size); +ZEXTERN int ZEXPORT deflateInit2_(z_streamp strm, int level, int method, + int windowBits, int memLevel, + int strategy, const char *version, + int stream_size); +ZEXTERN int ZEXPORT inflateInit2_(z_streamp strm, int windowBits, + const char *version, int stream_size); +ZEXTERN int ZEXPORT inflateBackInit_(z_streamp strm, int windowBits, + unsigned char FAR *window, + const char *version, + int stream_size); +#ifdef Z_PREFIX_SET +# define z_deflateInit(strm, level) \ + deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream)) +# define z_inflateInit(strm) \ + inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream)) +# define z_deflateInit2(strm, level, method, windowBits, memLevel, strategy) \ + deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\ + (strategy), ZLIB_VERSION, (int)sizeof(z_stream)) +# define z_inflateInit2(strm, windowBits) \ + inflateInit2_((strm), (windowBits), ZLIB_VERSION, \ + (int)sizeof(z_stream)) +# define z_inflateBackInit(strm, windowBits, window) \ + inflateBackInit_((strm), (windowBits), (window), \ + ZLIB_VERSION, (int)sizeof(z_stream)) +#else +# define deflateInit(strm, level) \ + deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream)) +# define inflateInit(strm) \ + inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream)) +# define deflateInit2(strm, level, method, windowBits, memLevel, strategy) \ + deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\ + (strategy), ZLIB_VERSION, (int)sizeof(z_stream)) +# define inflateInit2(strm, windowBits) \ + inflateInit2_((strm), (windowBits), ZLIB_VERSION, \ + (int)sizeof(z_stream)) +# define inflateBackInit(strm, windowBits, window) \ + inflateBackInit_((strm), (windowBits), (window), \ + ZLIB_VERSION, (int)sizeof(z_stream)) +#endif + +#ifndef Z_SOLO + +/* gzgetc() macro and its supporting function and exposed data structure. Note + * that the real internal state is much larger than the exposed structure. + * This abbreviated structure exposes just enough for the gzgetc() macro. The + * user should not mess with these exposed elements, since their names or + * behavior could change in the future, perhaps even capriciously. They can + * only be used by the gzgetc() macro. You have been warned. + */ +struct gzFile_s { + unsigned have; + unsigned char *next; + z_off64_t pos; +}; +ZEXTERN int ZEXPORT gzgetc_(gzFile file); /* backward compatibility */ +#ifdef Z_PREFIX_SET +# undef z_gzgetc +# define z_gzgetc(g) \ + ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g)) +#else +# define gzgetc(g) \ + ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g)) +#endif + +/* provide 64-bit offset functions if _LARGEFILE64_SOURCE defined, and/or + * change the regular functions to 64 bits if _FILE_OFFSET_BITS is 64 (if + * both are true, the application gets the *64 functions, and the regular + * functions are changed to 64 bits) -- in case these are set on systems + * without large file support, _LFS64_LARGEFILE must also be true + */ +#ifdef Z_LARGE64 + ZEXTERN gzFile ZEXPORT gzopen64(const char *, const char *); + ZEXTERN z_off64_t ZEXPORT gzseek64(gzFile, z_off64_t, int); + ZEXTERN z_off64_t ZEXPORT gztell64(gzFile); + ZEXTERN z_off64_t ZEXPORT gzoffset64(gzFile); + ZEXTERN uLong ZEXPORT adler32_combine64(uLong, uLong, z_off64_t); + ZEXTERN uLong ZEXPORT crc32_combine64(uLong, uLong, z_off64_t); + ZEXTERN uLong ZEXPORT crc32_combine_gen64(z_off64_t); +#endif + +#if !defined(ZLIB_INTERNAL) && defined(Z_WANT64) +# ifdef Z_PREFIX_SET +# define z_gzopen z_gzopen64 +# define z_gzseek z_gzseek64 +# define z_gztell z_gztell64 +# define z_gzoffset z_gzoffset64 +# define z_adler32_combine z_adler32_combine64 +# define z_crc32_combine z_crc32_combine64 +# define z_crc32_combine_gen z_crc32_combine_gen64 +# else +# define gzopen gzopen64 +# define gzseek gzseek64 +# define gztell gztell64 +# define gzoffset gzoffset64 +# define adler32_combine adler32_combine64 +# define crc32_combine crc32_combine64 +# define crc32_combine_gen crc32_combine_gen64 +# endif +# ifndef Z_LARGE64 + ZEXTERN gzFile ZEXPORT gzopen64(const char *, const char *); + ZEXTERN z_off_t ZEXPORT gzseek64(gzFile, z_off_t, int); + ZEXTERN z_off_t ZEXPORT gztell64(gzFile); + ZEXTERN z_off_t ZEXPORT gzoffset64(gzFile); + ZEXTERN uLong ZEXPORT adler32_combine64(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine64(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine_gen64(z_off_t); +# endif +#else + ZEXTERN gzFile ZEXPORT gzopen(const char *, const char *); + ZEXTERN z_off_t ZEXPORT gzseek(gzFile, z_off_t, int); + ZEXTERN z_off_t ZEXPORT gztell(gzFile); + ZEXTERN z_off_t ZEXPORT gzoffset(gzFile); + ZEXTERN uLong ZEXPORT adler32_combine(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine_gen(z_off_t); +#endif + +#else /* Z_SOLO */ + + ZEXTERN uLong ZEXPORT adler32_combine(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine(uLong, uLong, z_off_t); + ZEXTERN uLong ZEXPORT crc32_combine_gen(z_off_t); + +#endif /* !Z_SOLO */ + +/* undocumented functions */ +ZEXTERN const char * ZEXPORT zError(int); +ZEXTERN int ZEXPORT inflateSyncPoint(z_streamp); +ZEXTERN const z_crc_t FAR * ZEXPORT get_crc_table(void); +ZEXTERN int ZEXPORT inflateUndermine(z_streamp, int); +ZEXTERN int ZEXPORT inflateValidate(z_streamp, int); +ZEXTERN unsigned long ZEXPORT inflateCodesUsed(z_streamp); +ZEXTERN int ZEXPORT inflateResetKeep(z_streamp); +ZEXTERN int ZEXPORT deflateResetKeep(z_streamp); +#if defined(_WIN32) && !defined(Z_SOLO) +ZEXTERN gzFile ZEXPORT gzopen_w(const wchar_t *path, + const char *mode); +#endif +#if defined(STDC) || defined(Z_HAVE_STDARG_H) +# ifndef Z_SOLO +ZEXTERN int ZEXPORTVA gzvprintf(gzFile file, + const char *format, + va_list va); +# endif +#endif + +#ifdef __cplusplus +} +#endif + +#endif /* ZLIB_H */ diff --git a/jeweha-chs.exe.manifest b/jeweha-chs.exe.manifest new file mode 100644 index 0000000..2dd2182 --- /dev/null +++ b/jeweha-chs.exe.manifest @@ -0,0 +1,20 @@ + + + + + PerMonitorV2 + + + + + + + + + + + + + + + \ No newline at end of file diff --git a/lib/detours.lib b/lib/detours.lib new file mode 100644 index 0000000..ccb6017 Binary files /dev/null and b/lib/detours.lib differ diff --git a/lib/zlib.lib b/lib/zlib.lib new file mode 100644 index 0000000..0b156ef Binary files /dev/null and b/lib/zlib.lib differ diff --git a/libzip b/libzip new file mode 160000 index 0000000..e9083a7 --- /dev/null +++ b/libzip @@ -0,0 +1 @@ +Subproject commit e9083a7b41463489cd6641310c36f3b9197e1d27 diff --git a/main.cpp b/main.cpp new file mode 100644 index 0000000..970f56e --- /dev/null +++ b/main.cpp @@ -0,0 +1,70 @@ +#include +#include + +void ShowErrorMsg(LPCWSTR text) { + wchar_t* buf[1024]; + _swprintf((wchar_t *const)buf, L"%s%i", text, GetLastError()); + MessageBoxW(nullptr, (LPCWSTR)buf, L"错误消息", MB_OK); +} + +int WINAPI WinMain(HINSTANCE hInstance, HINSTANCE hPrevInstance, LPSTR lpCmdLine, int nCmdShow) { + // 要启动的进程名 + const wchar_t* processName = L"jeweha_.exe"; + // 要注入的 DLL 路径 + const wchar_t* dllPath = L"jeweha_CHS_1.0.dll"; + + // 启动进程 + STARTUPINFOW si; + PROCESS_INFORMATION pi; + ZeroMemory(&si, sizeof(si)); + ZeroMemory(&pi, sizeof(pi)); + + si.cb = sizeof(si); + + // 创建新进程 + if (!CreateProcessW((LPCWSTR)processName, nullptr, NULL, NULL, FALSE, CREATE_SUSPENDED, NULL, NULL, &si, &pi)) { + ShowErrorMsg(L"CreateProcessW failed: "); + return 1; + } + + size_t memSize = (wcslen(dllPath) + 1) * sizeof(wchar_t); + + // 在新进程中分配内存以存放 DLL 路径 + LPVOID pDllPath = VirtualAllocEx(pi.hProcess, NULL, memSize, MEM_COMMIT | MEM_RESERVE, PAGE_READWRITE); + if (!pDllPath) { + ShowErrorMsg(L"VirtualAllocEx failed: "); + CloseHandle(pi.hProcess); + CloseHandle(pi.hThread); + return 1; + } + + // 将 DLL 路径写入新进程的内存 + if (!WriteProcessMemory(pi.hProcess, pDllPath, (LPVOID)dllPath, memSize, NULL)) { + ShowErrorMsg(L"WriteProcessMemory failed: "); + VirtualFreeEx(pi.hProcess, pDllPath, 0, MEM_RELEASE); + CloseHandle(pi.hProcess); + CloseHandle(pi.hThread); + return 1; + } + + // 创建远程线程以加载 DLL + HANDLE hThread = CreateRemoteThread(pi.hProcess, NULL, 0, (LPTHREAD_START_ROUTINE)GetProcAddress(GetModuleHandleA("kernel32.dll"), "LoadLibraryW"), pDllPath, 0, NULL); + if (!hThread) { + ShowErrorMsg(L"CreateRemoteThread failed: "); + VirtualFreeEx(pi.hProcess, pDllPath, 0, MEM_RELEASE); + CloseHandle(pi.hProcess); + CloseHandle(pi.hThread); + return 1; + } + + // 等待线程完成 + WaitForSingleObject(hThread, INFINITE); + + // 清理 + VirtualFreeEx(pi.hProcess, pDllPath, 0, MEM_RELEASE); + CloseHandle(hThread); + ResumeThread(pi.hThread); // 恢复新进程的执行 + CloseHandle(pi.hProcess); + CloseHandle(pi.hThread); + return 0; +} diff --git a/string_replace_file.cpp b/string_replace_file.cpp new file mode 100644 index 0000000..72cf241 --- /dev/null +++ b/string_replace_file.cpp @@ -0,0 +1,43 @@ +#include "string_replace_file.hpp" +#include "rapidjson/document.h" +#include "rapidjson/writer.h" +#include "rapidjson/stringbuffer.h" +#include "fileop.h" +#include "wchar_util.h" +#include + +bool StringReplaceFile::Load(std::string file) { + size_t size; + if (!fileop::get_file_size(file, size)) { + return false; + } + std::wstring wfile; + if (!wchar_util::str_to_wstr(wfile, file, CP_UTF8)) { + return false; + } + HANDLE hFile = CreateFileW(wfile.c_str(), GENERIC_READ, FILE_SHARE_READ, NULL, OPEN_EXISTING, FILE_ATTRIBUTE_NORMAL, NULL); + if (hFile == INVALID_HANDLE_VALUE) { + return false; + } + char* buffer = new char[size]; + DWORD read; + if (!ReadFile(hFile, buffer, size, &read, NULL)) { + CloseHandle(hFile); + delete[] buffer; + return false; + } + CloseHandle(hFile); + rapidjson::Document doc; + doc.Parse(buffer); + delete[] buffer; + if (doc.IsObject()) { +#undef GetObject + for (auto& m : doc.GetObject()) { + if (m.value.IsString()) { + messages[m.name.GetString()] = m.value.GetString(); + } + } + } + return true; +} + diff --git a/string_replace_file.hpp b/string_replace_file.hpp new file mode 100644 index 0000000..3f35940 --- /dev/null +++ b/string_replace_file.hpp @@ -0,0 +1,8 @@ +#include +#include + +class StringReplaceFile { + public: + std::unordered_map messages; + bool Load(std::string file); +}; diff --git a/utils b/utils new file mode 160000 index 0000000..5fde5e6 --- /dev/null +++ b/utils @@ -0,0 +1 @@ +Subproject commit 5fde5e6e521bfbc74da3ec7144910e750b201b82 diff --git a/vfs.cpp b/vfs.cpp new file mode 100644 index 0000000..bdb0411 --- /dev/null +++ b/vfs.cpp @@ -0,0 +1,241 @@ +#include "vfs.hpp" +#include "wchar_util.h" +#include "str_util.h" +#include "fileop.h" +#include "shlwapi.h" + +VFS::VFS() { + WCHAR exePath[MAX_PATH]; + GetModuleFileNameW(NULL, exePath, MAX_PATH); + std::wstring path = exePath; + std::string pathStr; + if (!wchar_util::wstr_to_str(pathStr, path, CP_UTF8)) { + char buf[MAX_PATH]; + GetModuleFileNameA(NULL, buf, MAX_PATH); + pathStr = buf; + } + base_path = fileop::dirname(pathStr); + base_path = str_util::str_replace(base_path, "/", "\\"); +} + +VFS::~VFS() { + for (auto file : handles) { + zip_fclose((zip_file_t*)file.first); + } + for (auto archive : archives) { + zip_close(archive); + } +} + +bool VFS::AddArchive(std::string path) { + zip_t* archive = zip_open(path.c_str(), ZIP_RDONLY, nullptr); + if (!archive) return false; + archives.push_back(archive); + auto len = zip_get_num_entries(archive, 0); + for (zip_int64_t i = 0; i < len; i++) { + struct zip_stat st; + zip_stat_init(&st); + zip_stat_index(archive, i, 0, &st); + // Skip directories/folders (directory entries usually end with a '/') + if (st.name[strlen(st.name) - 1] == '/') { + continue; + } + std::string name = st.name; + name = str_util::str_replace(name, "/", "\\"); + files[name] = st.size; + } + return true; +} + +bool VFS::AddArchiveFromResource(HMODULE hModule, int resourceID) { + HRSRC hResInfo = FindResource(hModule, MAKEINTRESOURCE(resourceID), RT_RCDATA); + if (!hResInfo) return false; + HGLOBAL hResData = LoadResource(hModule, hResInfo); + if (!hResData) return false; + LPVOID lpResData = LockResource(hResData); + if (!lpResData) return false; + DWORD dwSize = SizeofResource(hModule, hResInfo); + if (!dwSize) return false; + auto re = zip_source_buffer_create(lpResData, dwSize, 0, nullptr); + if (!re) { + return false; + } + zip_t* archive = zip_open_from_source(re, ZIP_RDONLY, nullptr); + if (!archive) return false; + archives.push_back(archive); + auto len = zip_get_num_entries(archive, 0); + for (zip_int64_t i = 0; i < len; i++) { + struct zip_stat st; + zip_stat_init(&st); + zip_stat_index(archive, i, 0, &st); + // Skip directories/folders (directory entries usually end with a '/') + if (st.name[strlen(st.name) - 1] == '/') { + continue; + } + std::string name = st.name; + name = str_util::str_replace(name, "/", "\\"); + files[name] = st.size; + } + return true; +} + +void VFS::AddArchiveWithErrorMsg(std::string path) { + if (!AddArchive(path)) { + std::wstring wpath; + if (!wchar_util::str_to_wstr(wpath, path, CP_UTF8)) { + MessageBoxW(NULL, L"无法打开资源文件。请检查资源文件是否完整", L"错误", MB_ICONERROR); + ExitProcess(1); + return; + } + std::wstring wmsg = L"无法打开 " + wpath + L"。请检查文件是否存在"; + MessageBoxW(NULL, wmsg.c_str(), L"错误", MB_ICONERROR); + ExitProcess(1); + return; + } +} + +void VFS::AddArchiveFromResourceWithErrorMsg(HMODULE hModule, int resourceID) { + if (!AddArchiveFromResource(hModule, resourceID)) { + MessageBoxW(NULL, L"无法打开内置的资源文件。", L"错误", MB_ICONERROR); + ExitProcess(1); + return; + } +} + +bool VFS::ContainsFile(std::string path) { + path = str_util::str_replace(path, "/", "\\"); + if (fileop::isabs(path)) { + path = fileop::relpath(path, base_path); + } + return files.find(path) != files.end(); +} + +bool VFS::ContainsFile(std::wstring path) { + std::string str; + if (!wchar_util::wstr_to_str(str, path, CP_UTF8)) { + return false; + } + return ContainsFile(str); +} + +bool VFS::ContainsHandle(HANDLE hFile) { + return handles.find(hFile) != handles.end(); +} + +HANDLE VFS::CreateFileW(std::wstring path) { + std::string str; + if (!wchar_util::wstr_to_str(str, path, CP_UTF8)) { + SetLastError(ERROR_INVALID_PARAMETER); + return INVALID_HANDLE_VALUE; + } + str = fileop::relpath(str, base_path); + str = str_util::str_replace(str, "/", "\\"); + auto c = files.find(str); + if (c == files.end()) { + SetLastError(ERROR_FILE_NOT_FOUND); + return INVALID_HANDLE_VALUE; + } + str = (*c).first; + str = str_util::str_replace(str, "\\", "/"); + zip_t* archive = nullptr; + zip_uint64_t index = 0; + for (auto a : archives) { + if (zip_name_locate(a, str.c_str(), 0) != -1) { + archive = a; + break; + } + } + if (!archive) { + SetLastError(ERROR_FILE_NOT_FOUND); + return INVALID_HANDLE_VALUE; + } + index = zip_name_locate(archive, str.c_str(), 0); + zip_file_t* file = zip_fopen_index(archive, index, 0); + handles[(HANDLE)file] = str; + return (HANDLE)file; +} + +bool VFS::ReadFile(HANDLE hFile, LPVOID lpBuffer, DWORD nNumberOfBytesToRead, LPDWORD lpNumberOfBytesRead) { + if (!ContainsHandle(hFile)) { + SetLastError(ERROR_INVALID_HANDLE); + return false; + } + zip_file_t* file = (zip_file_t*)hFile; + if (!file) { + SetLastError(ERROR_INVALID_HANDLE); + return false; + } + zip_int64_t n = zip_fread(file, lpBuffer, nNumberOfBytesToRead); + if (n == -1) { + SetLastError(ERROR_INVALID_HANDLE); + return false; + } + if (lpNumberOfBytesRead) { + *lpNumberOfBytesRead = n; + } + return true; +} + +void VFS::CloseHandle(HANDLE hFile) { + if (!ContainsHandle(hFile)) { + SetLastError(ERROR_INVALID_HANDLE); + return; + } + zip_fclose((zip_file_t*)hFile); + handles.erase(hFile); +} + +DWORD VFS::GetFileSize(HANDLE hFile, LPDWORD lpFileSizeHigh) { + auto f = handles.find(hFile); + if (f == handles.end()) { + SetLastError(ERROR_INVALID_HANDLE); + return INVALID_FILE_SIZE; + } + auto data = *f; + auto name = data.second; + auto size = files[name]; + if (lpFileSizeHigh) { + *lpFileSizeHigh = size >> 32; + } + return size; +} + +BOOL VFS::GetFileSizeEx(HANDLE hFile, PLARGE_INTEGER lpFileSize) { + auto f = handles.find(hFile); + if (f == handles.end()) { + SetLastError(ERROR_INVALID_HANDLE); + return FALSE; + } + auto data = *f; + auto name = data.second; + auto size = files[name]; + lpFileSize->LowPart = size & 0xFFFFFFFF; + lpFileSize->HighPart = size >> 32; + return TRUE; +} + +DWORD VFS::SetFilePointer(HANDLE hFile, LONG lDistanceToMove, PLONG lpDistanceToMoveHigh, DWORD dwMoveMethod) { + if (!ContainsHandle(hFile)) { + SetLastError(ERROR_INVALID_HANDLE); + return INVALID_SET_FILE_POINTER; + } + zip_file_t* file = (zip_file_t*)hFile; + if (!file) { + SetLastError(ERROR_INVALID_HANDLE); + return INVALID_SET_FILE_POINTER; + } + zip_int64_t offset = lDistanceToMove; + if (lpDistanceToMoveHigh) { + offset |= ((zip_int64_t)*lpDistanceToMoveHigh) << 32; + } + zip_int64_t n = zip_fseek(file, offset, dwMoveMethod); + if (n == -1) { + SetLastError(ERROR_INVALID_HANDLE); + return INVALID_SET_FILE_POINTER; + } + return n; +} + +std::string VFS::GetBasePath() { + return base_path; +} diff --git a/vfs.hpp b/vfs.hpp new file mode 100644 index 0000000..43d588d --- /dev/null +++ b/vfs.hpp @@ -0,0 +1,52 @@ +#include "zip.h" +#include +#include +#include +#include +#include "str_util.h" + +struct CaseInsensitiveHash { + size_t operator()(const std::string& str) const { + // 创建字符串的小写副本 + std::string lowercaseStr = str_util::tolower(str); + + // 对小写字符串使用标准哈希函数 + return std::hash{}(lowercaseStr); + } +}; + +// 比较函数，忽略大小写 +struct CaseInsensitiveEqual { + bool operator()(const std::string& left, const std::string& right) const { + return left.size() == right.size() && + std::equal(left.begin(), left.end(), right.begin(), + [](unsigned char a, unsigned char b) { + return std::tolower(a) == std::tolower(b); + }); + } +}; + +class VFS { + public: + VFS(); + ~VFS(); + bool AddArchive(std::string path); + bool AddArchiveFromResource(HMODULE hModule, int resourceID); + void AddArchiveWithErrorMsg(std::string path); + void AddArchiveFromResourceWithErrorMsg(HMODULE hModule, int resourceID); + bool ContainsFile(std::string path); + bool ContainsFile(std::wstring path); + bool ContainsHandle(HANDLE hFile); + HANDLE CreateFileW(std::wstring path); + bool ReadFile(HANDLE hFile, LPVOID lpBuffer, DWORD nNumberOfBytesToRead, LPDWORD lpNumberOfBytesRead); + void CloseHandle(HANDLE hFile); + DWORD GetFileSize(HANDLE hFile, LPDWORD lpFileSizeHigh); + BOOL GetFileSizeEx(HANDLE hFile, PLARGE_INTEGER lpFileSize); + DWORD SetFilePointer(HANDLE hFile, LONG lDistanceToMove, PLONG lpDistanceToMoveHigh, DWORD dwMoveMethod); + std::unordered_map files; + std::string GetBasePath(); + private: + std::string base_path; + std::list archives; + std::unordered_map handles; +}; diff --git a/winres.rc b/winres.rc new file mode 100644 index 0000000..ce7324c Binary files /dev/null and b/winres.rc differ