youtubedl-subtitle-404-workaround.py

# https://github.com/ytdl-org/youtube-dl/blob/7947a1f7dbc6ba47a6f22ab67fd330e57c0ef87c/youtube_dl/YoutubeDL.py#L1811
# in process_info method of the YoutubeDL class in youtube_dl/YoutubeDL.py

# original as of writing this (latest commit fca6dba8b80286ae6d3ca0a60c4799c220a52650)
# youtube-dl version 2020.07.28
if subtitles_are_requested and info_dict.get('requested_subtitles'):
    # subtitles download errors are already managed as troubles in relevant IE
    # that way it will silently go on when used with unsupporting IE
    subtitles = info_dict['requested_subtitles']
    ie = self.get_info_extractor(info_dict['extractor_key'])
    for sub_lang, sub_info in subtitles.items():
        sub_format = sub_info['ext']
        sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
        if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
            self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
        else:
            self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
            if sub_info.get('data') is not None:
                try:
                    # Use newline='' to prevent conversion of newline characters
                    # See https://github.com/ytdl-org/youtube-dl/issues/10268
                    with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
                        subfile.write(sub_info['data'])
                except (OSError, IOError):
                    self.report_error('Cannot write subtitles file ' + sub_filename)
                    return
            else:
                try:
                    sub_data = ie._request_webpage(
                        sub_info['url'], info_dict['id'], note=False).read()
                    with io.open(encodeFilename(sub_filename), 'wb') as subfile:
                        subfile.write(sub_data)
                except (ExtractorError, IOError, OSError, ValueError) as err:
                    self.report_warning('Unable to download subtitle for "%s": %s' %
                                        (sub_lang, error_to_compat_str(err)))
                    continue

                   
# my changed code
if subtitles_are_requested and info_dict.get('requested_subtitles'):
    # subtitles download errors are already managed as troubles in relevant IE
    # that way it will silently go on when used with unsupporting IE
    subtitles = info_dict['requested_subtitles']
    ie = self.get_info_extractor(info_dict['extractor_key'])
    failed_sub_langs = [] #CHANGE: keep track of sub langs that fail to download
    for sub_lang, sub_info in subtitles.items():
        sub_format = sub_info['ext']
        sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
        if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
            self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
        else:
            self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
            if sub_info.get('data') is not None:
                try:
                    # Use newline='' to prevent conversion of newline characters
                    # See https://github.com/ytdl-org/youtube-dl/issues/10268
                    with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
                        subfile.write(sub_info['data'])
                except (OSError, IOError):
                    self.report_error('Cannot write subtitles file ' + sub_filename)
                    return
            else:
                try:
                    sub_data = ie._request_webpage(
                        sub_info['url'], info_dict['id'], note=False).read()
                    with io.open(encodeFilename(sub_filename), 'wb') as subfile:
                        subfile.write(sub_data)
                except (ExtractorError, IOError, OSError, ValueError) as err:
                    self.report_warning('Unable to download subtitle for "%s": %s' %
                                        (sub_lang, error_to_compat_str(err)))
                    failed_sub_langs.append(sub_lang) #CHANGE: add sub lang to the list if it fails
                    continue
    #CHANGE: delete failed sub langs from 'requested_subtitles' dictionary
    for sub_lang in failed_sub_langs:
        del info_dict['requested_subtitles'][sub_lang]