switch from bms to wavescan script

.gitignore

@@ -2,4 +2,5 @@ output/
 temp/
 *.pck
 *.hdiff
-tools.zip
+tools.zip
+__pycache__/

extract.py

@@ -2,6 +2,7 @@ import os
 import shutil
 import zipfile
 import filecmp
+import wavescan
 import subprocess
 from halo import Halo
 from progress.bar import PixelBar
@@ -60,20 +61,10 @@ def main():
 			if alone:
 				output_path = "wem"
 
-			# prepare args
-			args = [
-				path("tools/quickbms/quickbms.exe"),
-				"-o",
-				"-Y",
-				path("tools/quickbms/wavescan.bms"),
-				path(f"temp/{file}"),
-				path(f"temp/{output_path}")
-			]
-
 			# update spinner and call program
 			spinner.text = f"[{curr}/{steps}] Extracting"
 			spinner.start()
-			call(args)
+			wavescan.extract(path(f"temp/{file}"), path(f"temp/{output_path}"))
 			spinner.stop()
 			print(f"[{curr}/{steps}] Extracting")
 
@@ -114,20 +105,10 @@ def main():
 			if not alone:
 				curr += 1
 
-				# prepare args
-				args = [
-					path("tools/quickbms/quickbms.exe"),
-					"-o",
-					"-Y",
-					path("tools/quickbms/wavescan.bms"),
-					path(f"temp/{file}"),
-					path(f"temp/patched_decoded")
-				]
-
 				# update spinner and call program
 				spinner.text = f"[{curr}/{steps}] Extracting patch"
 				spinner.start()
-				call(args)
+				wavescan.extract(path(f"temp/{file}"), path(f"temp/patched_decoded"))
 				spinner.stop()
 				print(f"[{curr}/{steps}] Extracting patch")
 

filereader.py

@@ -3,16 +3,15 @@ import struct
 
 class FileReader:
 	""" 
-	Simplified byte file reader with buffer 
+	Simplified byte file reader with buffer, it's not particularly optimised but good enough
-	Not particularly optimised, contains repetitive functions
+	In the scope of this project, not everything will be used in here
-	In the scope of this project, not everything will be used either
-
 	"""
-	def __init__(self, file, endianness: str):
+
+	def __init__(self, file, endianness:str):
 		self.stream = file
 		self.endianness = endianness
 
-	def _read(self, mode:str, bufferLength:int, endianness=None) -> bytes:
+	def _read(self, mode:str, bufferLength:int, endianness:str=None) -> bytes:
 		# endianness override
 		if endianness is None:
 			endianness = self.endianness
@@ -22,31 +21,31 @@ class FileReader:
 		return struct.unpack(f"{endianness}{mode}", bytearray(self.stream.read(bufferLength)))[0]
 
 	# read methods
-	def ReadInt16(self, endianness=None) -> int:
+	def ReadInt16(self, endianness:str=None) -> int:
 		return self._read("h", 2, endianness)
 	
-	def ReadUInt16(self, endianness=None) -> int:
+	def ReadUInt16(self, endianness:str=None) -> int:
 		return self._read("H", 2, endianness)
 	
-	def ReadInt32(self, endianness=None) -> int:
+	def ReadInt32(self, endianness:str=None) -> int:
 		return self._read("i", 4, endianness)
 	
-	def ReadUInt32(self, endianness=None) -> int:
+	def ReadUInt32(self, endianness:str=None) -> int:
 		return self._read("I", 4, endianness)
 
-	def ReadLong(self, endianness=None) -> int:
+	def ReadLong(self, endianness:str=None) -> int:
 		return self._read("l", 4, endianness)
 
-	def ReadULong(self, endianness=None) -> int:
+	def ReadULong(self, endianness:str=None) -> int:
 		return self._read("L", 4, endianness)
 
-	def ReadLongLong(self, endianness=None) -> int:
+	def ReadLongLong(self, endianness:str=None) -> int:
 		return self._read("q", 8, endianness)
 
-	def ReadULongLong(self, endianness=None) -> int:
+	def ReadULongLong(self, endianness:str=None) -> int:
 		return self._read("Q", 8, endianness)
 
-	def ReadBytes(self, length:int, endianness=None) -> bytes:
+	def ReadBytes(self, length:int, endianness:str=None) -> bytes:
 		return self._read(f"{str(length)}s", int(length), endianness)
 
 	# buffer utils

tools/quickbms/wavescan.bms

@@ -1,67 +0,0 @@
-# modified version to ignore filenames (and wav)
-#
-# scan data for wave files
-# RIFF and RIFX header supported
-# note: There are wave files with a wrong file size after RIFF/RIFX
-#       This script takes the stream size, adds the header size and writes the correct size after RIFF/RIFX
-# (c) 2012-06-26 by AlphaTwentyThree
-#
-# future update plans:
-# - option to also write data between found wave files to disk
-# - option to automatically transform the file to a playable or at least decodable format
-
-for i = 1                        # run through loop with count variable i
-   FindLoc OFFSET string "WAVE" 0 ""   # search for "WAVE", save position as variable OFFSET
-   if OFFSET == ""                  # when nothing is found
-      cleanexit                  # the script exits (e.g. at end of file)
-   endif
-   math OFFSET -= 8               # jump to possible
-   goto OFFSET                     # RIFF/RIFX file start
-   getDstring IDENT 4               # read string of 4 bytes, save variable as IDENT
-   if IDENT == "RIFX"               # differentiate between header possibilities
-      endian big                  # set endianness to big, if file has RIFX identifier
-      callfunction write 1         # see function section below
-   elif IDENT == "RIFF"            # endianness stays little
-      callfunction write 1         # also run function
-   else                        # string "WAVE" found, but doesn't belong to wave file
-      set SIZE 0xc               # do as if something with 0xc bytes was found to continue search from the right position
-   endif
-   set SEARCH OFFSET               # set marker to position from where to search next
-   math SEARCH += SIZE               # (that would be after the file that was found)
-   if SEARCH == FSIZE               # in case the last found file ends with the main file, we exit
-      cleanexit
-   endif
-   goto SEARCH                     # if we haven't exited the script above, we set out cursor to after the last found file
-next i
-
-startfunction write                  # function "write" starts here, is called when a wave file is found above
-   get NAME basename               # save name without extension under variable NAME
-   string NAME += "_"               # add underscore to the name
-   string NAME += i               # add the loop variable to the name
-   goto OFFSET                     # set cursor to the beginning of the found file
-   get DUMMY long                  # RIFF/RIFX identifier, not needed
-   get DUMMY long                  # riff size, not needed
-   get DUMMY long                  # "WAVE", not needed, we arrive at the "fmt " section
-   for                           # loop search for the "data" section at the start of the stream (get the stream size from there)
-      getDstring AREA_NAME 4         # name of area in header
-      get AREA_SIZE long            # size of area in header
-      savepos MYOFF               # save position we are at
-      if AREA_NAME == "data"         # when we arrive at the needed "data" area:
-         break                  # we exit the loop
-      else                     # otherwise:
-         math MYOFF += AREA_SIZE      # not reached "data" area -> adjust cursor position...
-         goto MYOFF               # ... and go there
-      endif
-   next                        # remember: the cursor is now directly at the stream start
-   set STREAMSIZE AREA_SIZE         # the last AREA_SIZE is the size we need (size of the audio stream)
-   set HEADERSIZE MYOFF            # 
-   math HEADERSIZE -= OFFSET         # calculate the size of the stream header (offset - offset = size)
-   set SIZE HEADERSIZE               # 
-   math SIZE += STREAMSIZE            # calculate complete file size (header + stream = file)
-   log MEMORY_FILE OFFSET SIZE         # write file to memory
-   math SIZE -= 8                  # subtract 8 bytes to get the riff size
-   putVarChr MEMORY_FILE 4 SIZE long    # write the correct riff size to the header inside the memory
-   string NAME += ".wem"            # add extension to the name (the name could contain the name of the first marker if the file has markers)
-   math SIZE += 8                  # add the subtracted 8 bytes again
-   log NAME 0 SIZE MEMORY_FILE         # write file in memory to disk
-endfunction                        # remember that we continue with our next i now!

wavescan.py

@@ -1,16 +1,18 @@
 # Custom rewrite of the Wwise AKPK packages extractor, original by Nicknine and bnnm
 # TODO: use extracted files id with the mapping table to restore file names
-# TODO: make this script callable by the main one
 from filereader import FileReader
 import os
 
-filename = "External0.pck"
+
-cwd = os.getcwd()
+reader = None
-reader = FileReader(open(filename, "rb"), "little") # defaults to little endian
 bank_version = 0
 
-def main():
+
+def extract(input_file, output_folder):
 	global bank_version
+	global reader
+
+	reader = FileReader(open(input_file, "rb"), "little") # defaults to little endian
 
 	# check file
 	if reader.ReadBytes(4) != b"AKPK":
@@ -50,7 +52,7 @@ def main():
 
 	# extract each sector
 	for sector in sectors:
-		extract_sector(*sector[1:], endianness, lang_array, bank_version)
+		extract_sector(*sector[1:], endianness, lang_array, bank_version, output_folder)
 
 		if sector[0] and bank_version == 0:
 			if externals_sector_size == 0:
@@ -109,7 +111,7 @@ def detect_bank_version(offset):
 
 	reader.SetBufferPos(current)
 
-def extract_sector(section_size, is_sounds, is_externals, ext, endianness, lang_array, bank_version, filter_bnk_only=0, filter_wem_only=0):
+def extract_sector(section_size, is_sounds, is_externals, ext, endianness, lang_array, bank_version, output_folder, filter_bnk_only=0, filter_wem_only=0):
 	# check sector validity
 	if section_size == 0:
 		return
@@ -193,20 +195,18 @@ def extract_sector(section_size, is_sounds, is_externals, ext, endianness, lang_
 		if filter_wem_only == 1 and ext != "wem":
 			continue
 
-		print(f"NAME - {name} | OFFSET - {offset} | SIZE - {size}")
+		# file infos
+		# print(f"NAME - {name} | OFFSET - {offset} | SIZE - {size}")
 
 		# save file into disk
 		current = reader.GetBufferPos()
 		reader.SetBufferPos(offset)
 		file_data = reader.ReadBytes(size)
 
-		os.makedirs(os.path.join(cwd, os.path.dirname(name)), exist_ok=True)
+		os.makedirs(output_folder, exist_ok=True)
 
-		with open(os.path.join(cwd, name), "wb+") as f:
+		with open(os.path.join(output_folder, os.path.basename(name)), "wb+") as f:
 			f.write(file_data)
 			f.close()
 
 		reader.SetBufferPos(current)
-
-if __name__ == "__main__":
-	main()