working script extraction

getting_dialog.py

@@ -16,9 +16,10 @@ def read_until_space(indx: int, data: tuple) -> int:
     end = indx
     while end + 1 < len(data):
         ## Read 2 bytes at a time until null found
-        if data[end:end+2] == b"\x4F\x00":
+        d = data[end:end+1]
+        if data[end:end+1] == b"\x00":
             break
-        end += 2
+        end += 1
     return end
 
 
@@ -73,6 +74,66 @@ e_name_data = Path(f"YGO_2020/main/howto_db/howtoplay_{language_code}.bin").read
 language_code = 'J'
 j_name_data = Path(f"YGO_2020/main/howto_db/howtoplay_{language_code}.bin").read_bytes() # Get all the bytes
 
+language_code = 'J'
+script_data = Path(f"YGO_2020/main/scriptdata_{language_code}.bin").read_bytes() # Get all the bytes
+
+
+offset = struct.unpack("<Q", script_data[0:8])[0] # 11914
+titlecount = struct.unpack("<I", script_data[8:12])[0] # 307
+scriptblockcount = struct.unpack("<I", script_data[12:16])[0] # 7340
+
+titleStartPos = list()
+titleEndPos = list()
+titleOffset = list()
+for p in range(titlecount):
+    i = p * 16
+    titleStartPos.append(struct.unpack("<I", script_data[i:i+4])[0])
+    titleEndPos.append(struct.unpack("<I", script_data[i+4:i+8])[0])
+    titleOffset.append(struct.unpack("<Q", script_data[i+8:i+16])[0])
+## 307 items? Is that good
+
+script_titles = list()
+for o in range(len(titleOffset)):
+    start = titleOffset[o]
+    end = read_until_space(start,script_data)
+    script_titles.append(script_data[start:end].decode("utf-8"))
+    print( script_data[start:end].decode("utf-8") )
+
+## Script Data extraction!
+scripts = {}
+for p in range(titlecount):
+    scripts[p] = {"title":script_titles[p], "blocks" : list()}
+    script_block = {}
+    blockCount = titleEndPos[p] - titleStartPos[p] #+1
+    #start = offset + titleOffset[p] + titleStartPos[p]
+    for j in range(blockCount):
+        seeker = offset + titleStartPos[p] * 32 + j * 32
+        start_char = struct.unpack("<Q", script_data[seeker:seeker+8])[0]
+        end = read_until_space(start_char, script_data)
+        script_block['character'] = script_data[start_char:end].decode('utf-8')
+
+        seeker = seeker + 8
+        start_arg1 = struct.unpack("<Q", script_data[seeker:seeker+8])[0]
+        end = read_until_space(start_arg1, script_data)
+        script_block['arg1'] = script_data[start_arg1:end].decode('utf-8')
+
+        seeker = seeker + 8
+        start_arg2 = struct.unpack("<Q", script_data[seeker:seeker+8])[0]
+        end = read_until_space(start_arg2, script_data)
+        script_block['arg2'] = script_data[start_arg2:end].decode('utf-8')
+
+        seeker = seeker + 8
+        start_text = struct.unpack("<Q", script_data[seeker:seeker+8])[0]
+        end = read_until_space(start_text, script_data) ## These strings seem to always start with /0x00
+        script_block['text'] = script_data[start_text:end].decode('utf-8')
+        
+        scripts[p]['blocks'].append(script_block.copy())
+        print(script_block)
+
+with open(f"scripts_dump_{language_code}.json", "w", encoding="utf-8") as outfile:
+   json.dump(scripts, outfile, ensure_ascii=False, indent=2)
+exit()
+
 print("now what?")
 
 count_of_items = int.from_bytes(e_name_data[0:4], byteorder='big') # 202
@@ -115,28 +176,6 @@ read_until_null(1624,name_data)
 name_data[1624:read_until_null(1624,name_data)].decode("utf-16be", errors="replace")
 
 
-language_code = 'E'
-script_data = Path(f"YGO_2020/main/scriptdata_{language_code}.bin").read_bytes() # Get all the bytes
-
-
-offset = struct.unpack("<Q", script_data[0:8])[0] # 11914
-titlecount = struct.unpack("<I", script_data[8:12])[0] # 307
-scriptblockcount = struct.unpack("<I", script_data[12:16])[0] # 7340
-
-titleStartPos = list()
-titleEndPos = list()
-titleOffset = list()
-for p in range(titlecount):
-    i = p * 16
-    titleStartPos.append(struct.unpack("<I", script_data[i:i+4])[0])
-    titleEndPos.append(struct.unpack("<I", script_data[i+4:i+8])[0])
-    titleOffset.append(struct.unpack("<Q", script_data[i+8:i+16])[0])
-## 307 items? Is that good
-
-for o in range(len(titleOffset)):
-    start = titleOffset[o]
-    end = read_until_space(start,script_data)
-    print( script_data[start:end].decode("utf-8") )
 
 ## Old attempts at getting dlg file extracted, leaving block for reference
 language_code = 'J'