Add support to extarct ._bp file

2025-04-20 21:48:59 +08:00
parent 7c2f3b2778
commit 8a2eea3636
2 changed files with 199 additions and 91 deletions
--- a/extract_bp.py
+++ b/extract_bp.py
@@ -0,0 +1,108 @@
+import struct
+import sys
+import json
+
+
+class BPScript:
+    def __init__(self, path: str):
+        self.path = path
+        with open(path, 'rb') as f:
+            self.data = bytearray(f.read())
+        self.len = len(self.data)
+        self.read_header()
+        self.iPos = self.header_size
+
+    def read_header(self):
+        self.header_size = struct.unpack('<L', self.data[0:4])[0]
+        self.instr_size = struct.unpack('<L', self.data[4:8])[0]
+        if self.len != self.header_size + self.instr_size:
+            raise ValueError("Invalid BPScript file size")
+        self.iPos = self.header_size
+        self.last_pos = 0
+        while self.iPos < self.len - 4:
+            d = struct.unpack('<L', self.data[self.iPos:self.iPos + 4])[0]
+            if d == 0x17:
+                self.last_pos = self.iPos
+            self.iPos += 4
+
+    def extract_string(self):
+        offset = struct.unpack('<h', self.data[self.iPos:self.iPos + 2])[0]
+        start_pos = self.iPos + offset - 1
+        self.iPos += 2
+        if start_pos < self.last_pos or start_pos >= self.len or ((start_pos > self.last_pos + 1) and self.data[start_pos-1] != 0):
+            self.iPos -= 2
+            return None
+        pos = start_pos
+        while True:
+            if self.data[pos] == 0x00:
+                break
+            pos += 1
+        data = self.data[start_pos:pos]
+        try:
+            data = data.decode('cp932')
+        except UnicodeDecodeError:
+            data = data.decode('utf-8')
+            data = "utf8:" + data
+        if not data:
+            return None
+        return start_pos, data
+
+    def extract_values(self):
+        strings = []
+        self.iPos = self.header_size
+        while self.iPos < self.last_pos:
+            ins = self.data[self.iPos]
+            self.iPos += 1
+            if ins == 0x5:
+                try:
+                    t = self.extract_string()
+                    if t is not None:
+                        strings.append(t)
+                except UnicodeDecodeError:
+                    self.iPos -= 2
+        return dict(strings)
+    
+    def extract_strings(self):
+        self.iPos = self.last_pos + 1
+        while self.data[self.iPos] == 0:
+            self.iPos += 1
+        strings = []
+        start_pos = self.iPos
+        while self.iPos < self.len:
+            if self.data[self.iPos] == 0:
+                data = self.data[start_pos:self.iPos]
+                try:
+                    data = data.decode('cp932')
+                except UnicodeDecodeError:
+                    data = data.decode('utf-8')
+                    data = "utf8:" + data
+                if data:
+                    strings.append((start_pos, data))
+                start_pos = self.iPos + 1
+            self.iPos += 1
+        return dict(strings)
+
+
+base = sys.argv[1]
+json_f = f"{base}.json"
+if len(sys.argv) > 2:
+    json_f = sys.argv[2]
+scr = BPScript(base)
+print(scr.header_size)
+print(scr.last_pos)
+a = scr.extract_values()
+print(a)
+b = scr.extract_strings()
+print(b)
+print(len(a), len(b))
+missing = {}
+for key in b.keys():
+    if key not in a:
+        missing[key] = b[key]
+for key in a.keys():
+    if key not in b:
+        missing[key] = a[key]
+print(missing)
+print(len(missing))
+with open(json_f, "w", encoding="utf-8") as f:
+    json.dump(b, f, ensure_ascii=False, indent=2)