Skrypt do poprawiania napisów (MicroDVD, MPlayer2).

napraw napisy.py

#! python3
# -*- coding: UTF-8 -*-
#
# Skrypt do poprawiania napisów (microdvd {123}, mplayer [123]).
# (Dostosowuje napisy z Napiprojekt do MPC-HC)
#
# Zmienia rozszerzenie txt na sub.
# Usuwa komentarze napiprojektu, subedita itp.
# Usuwa linijki zawierające linki (z podglądem).
# W napisach MicroDVD poprawia ukośniki (kursywa/italic) na znaki kontrolne {y:i}.
# W napisach Mplayer poprawia znaki kontrolne na ukośniki i usuwa inne (np kolor).
# Usuwa puste linie na końcu pliku.
#
# Sposób użycia:
# Upuścić plik z napisami na ikonkę skryptu lub
# odpalić skrypt w folderze z napisami (muszą mieć rozszerzenie txt)
#

import os, sys
import re
from fnmatch import filter as flt
from locale import getdefaultlocale, getpreferredencoding
sysenc = getpreferredencoding()
#syslang, sysenc = getdefaultlocale()
#syslang = '.' + syslang[:2]
print('System encoding:', sysenc)
fsenc = sys.getfilesystemencoding()
print('File system encoding:', fsenc)
print('\n')


def askremove(match):
    print('', match.group(0))
    rem = input('Remove [T/n]?:').strip()
    if rem and rem[0] in 'Nn':
        return match.group(0)
    else:
        return ''


def replace(txt):
    chs = 0

    #info
    txt, num = re.subn(r'[[{]*\d+[][}{:]{1,2}\d+[]}:](?:\d+:)?(?:movie info:)?.*?(?:subedit b\.\d{4}|napiprojekt)[^\r\n]+[\r\n]+', r'', txt, flags=re.I)
    chs += num
    #WARNING: re.X - spaces removed from expression
    txt, num = re.subn(
            r'''[[{]*\d+[][}{:]{1,2}\d+[]}:](?:\d+:)?
                    [^\r\n]*
                        (?:
                           #hatak|
                            \bw\s*w\s*w\b|
                            \bh\s*t\s*t\s*p\b|
                            \.\s*pl\b|
                            \.\s*com\b|
                            \.\s*org\b
                        )
                    [^\r\n]*[\r\n]+
            ''', askremove, txt, flags=re.I|re.X)
    chs += num
    #linie na końcu
    txt, num = re.subn(r'\s+$', r'', txt)
    chs += num

    #/ -> {y:i}
    if re.search(r'^{\d+}{\d+}', txt):#microdvd
        txt, num = re.subn(r'([}|])\s*((?:-\s*)?)/\s*', r'\1{y:i}\2', txt)
        chs += num
    elif re.search(r'^\[\d+\]\[\d+\]', txt):#mpl2
        #{y:i} -> /
        txt, num = re.subn(r'\s*{[yY]:i}', r'/', txt)
        chs += num
        #usuń znaki kontrolne
        txt, num = re.subn(r'\s*{\w:[^}]+}', r'', txt)
        chs += num

    #końcowy /
    txt, num = re.subn(r'\s*/\s*(\||\r|\n|$)', r'\1', txt)
    chs += num


    return txt, chs


def rename(path):
    newp = os.path.splitext(path)[0]
    if newp.endswith(syslang):
        newp += '.sub'
    else:
        newp += syslang + '.sub'
    if os.path.exists(newp):
        print('Path exists:', newp)
        if os.path.isfile(newp):
            rem = input('Replace [T/n]?:').strip()
            if rem and rem[0] in 'Nn':
                return path
            else:
                try:
                    os.remove(newp)
                    return newp
                except OSError as e:
                    print('{message} ({number})'.format(number=e.errno, message=e.strerror))
                    return path
        else:
            print('Not file...')
            return path

    try:
        os.rename(path, newp)
    except OSError as e:
        print('{message} ({number})'.format(number=e.errno, message=e.strerror))
        return path

    return newp

def process(flist):
    for sub in flist:
        if not os.path.isfile(sub): continue
        print(sub)
        with open(sub, 'rb') as bomtest:
            bom = bomtest.read(3)
        if b'\xEF\xBB\xBF' == bom: #napiprojekt używa tylko utf-8-sig
            print('Czytany jako utf-8-sig')
            with open(sub, 'r', encoding='utf-8-sig', errors='replace') as file:
                lines = file.read()
        else:
            with open(sub, 'r', encoding=sysenc, errors='replace') as file:
                lines = file.read()
        if not lines:
            print('Empty file, skipped...')
            continue
        if re.search('[[{]*\d+[][}{:]{1,2}\d+[]}:](?:\d+:)?', lines):
            lines, changes = replace(lines)
            if changes:
                print('Changes:', changes)
                #xy-vobsub not detect text subtitles in .sub files
                #skipped for now
                #sub = rename(sub)
                with open(sub, 'w', encoding=sysenc, errors='replace') as file:
                    file.write(lines)
        else:
            print('Not subtitle, skipped...')

        print('\n')


def main():
    if len(sys.argv) > 1:
        flist = [os.path.normpath(p) for p in sys.argv[1:]]
        txt = flt(flist, '*.txt')
        sub = flt(flist, '*.sub')
        flist = txt + sub
    else:
        flist = os.listdir(os.getcwd())
        flist = flt(flist, '*.txt')

    if not flist: return
    process(flist)

    os.system('pause')


if __name__ == '__main__':
    try:
        main()
    except Exception:
        import traceback
        print("Unhandled exception:\n")
        traceback.print_exc()
        input('Press any key...')