MES file text extractor from Dead of the Brain (PC-9801, 1992)

mes-decrypt.py

#!/usr/bin/python

import re

filename = './MES/OPEN_1.MES'
with open(filename, 'rb') as f:
    content = f.read()
# get lines from bytes start marker 0xBA 0x23-25 to end marker 0xBA 0x26
# 0x23 == cole, 0x24 == doc, 0x25 == jack(?)
results = re.findall(br'(\xBA[\x23-\x25].*?)\xBA\x26', content)
if results:
    for r in results:
        # print repr(r)
        # r = results[0]
        sym = False
        control = False
        skip = False
        sub = ''
        for c in r:
            if c == '\xBA' and not skip:    # control byte
                control = True
            elif control and c == '\x23':   # dialog start cole
                sub += '\x83\x52\x81\x5B\x83\x8B\x81\x46'
                control = False
            elif control and c == '\x24':   # dialog start doc
                sub += '\x83\x4E\x81\x5B\x83\x4B\x81\x5B\x81\x46'
                control = False
            elif control and c == '\x25':   # dialog start jack?
                sub += '\x81\x46'
                control = False
            elif control and c == '\x26':   # dialog end
                control = False
            elif control and c == '\x28':   # symbol byte
                sym = True
                control = False
            elif sym and c == '\x0D':       # comma
                sub += '\x81\x41'
                sym = False
            elif sym and c == '\x0E':       # ..
                sub += '\x81\x64'
                sym = False
            elif sym and c == '\x0F':       # period
                sub += '\x81\x42'
                sym = False
            elif sym and c == '\x10':       # space
                sub += '\x81\x40'
                sym = False
            elif sym and c == '\x11':       # !
                sub += '\x81\x49'
                sym = False
            elif sym and c == '\x12':       # ?
                sub += '\x81\x48'
                sym = False
            elif sym and c == '\x13':       # newline for two characted dialog in one box
                sub += '\n'
                sym = False
            elif skip:
                sub += c
                skip = False
            elif re.match('[\x81-\x83,\x88-\x9F,\xE0-\xEA]', c):    # kana, kanji and some symbols - skip next byte
                sub += c
                skip = True
            elif sym or control:
                print "ERROR sym/control byte skipped"
                sym = False
                control = False
            else:
                # hiragana char, prefix 0x82 and shift by 0x74
                sub += '\x82' + chr(ord(c)+114)
        sub += '\n'
        print (sub.decode('shift-jis')).encode('utf-8', 'ignore')