X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/2b4515201a54c38e34434ea2817d606be134a7c9..0bec2c9f91a3b4f834e00c8ea4790481e257d0a2:/youtube_dl/__init__.py diff --git a/youtube_dl/__init__.py b/youtube_dl/__init__.py index ad5f13d..ba435ea 100644 --- a/youtube_dl/__init__.py +++ b/youtube_dl/__init__.py @@ -16,7 +16,6 @@ from .options import ( parseOpts, ) from .compat import ( - compat_expanduser, compat_getpass, compat_shlex_split, workaround_optparse_bug9161, @@ -26,6 +25,7 @@ from .utils import ( decodeOption, DEFAULT_OUTTMPL, DownloadError, + expand_path, match_filter_func, MaxDownloadsReached, preferredencoding, @@ -88,7 +88,7 @@ def _real_main(argv=None): batchfd = sys.stdin else: batchfd = io.open( - compat_expanduser(opts.batchfile), + expand_path(opts.batchfile), 'r', encoding='utf-8', errors='ignore') batch_urls = read_batch_urls(batchfd) if opts.verbose: @@ -191,12 +191,17 @@ def _real_main(argv=None): if numeric_buffersize is None: parser.error('invalid buffer size specified') opts.buffersize = numeric_buffersize + if opts.http_chunk_size is not None: + numeric_chunksize = FileDownloader.parse_bytes(opts.http_chunk_size) + if not numeric_chunksize: + parser.error('invalid http chunk size specified') + opts.http_chunk_size = numeric_chunksize if opts.playliststart <= 0: raise ValueError('Playlist start must be positive') if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart: raise ValueError('Playlist end must be greater than playlist start') if opts.extractaudio: - if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']: + if opts.audioformat not in ['best', 'aac', 'flac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']: parser.error('invalid audio format specified') if opts.audioquality: opts.audioquality = opts.audioquality.strip('k').strip('K') @@ -206,7 +211,7 @@ def _real_main(argv=None): if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']: parser.error('invalid video recode format specified') if opts.convertsubtitles is not None: - if opts.convertsubtitles not in ['srt', 'vtt', 'ass']: + if opts.convertsubtitles not in ['srt', 'vtt', 'ass', 'lrc']: parser.error('invalid subtitle format specified') if opts.date is not None: @@ -238,7 +243,7 @@ def _real_main(argv=None): any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json any_printing = opts.print_json - download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive + download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive # PostProcessors postprocessors = [] @@ -259,6 +264,16 @@ def _real_main(argv=None): 'key': 'FFmpegVideoConvertor', 'preferedformat': opts.recodevideo, }) + # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and + # FFmpegExtractAudioPP as containers before conversion may not support + # metadata (3gp, webm, etc.) + # And this post-processor should be placed before other metadata + # manipulating post-processors (FFmpegEmbedSubtitle) to prevent loss of + # extra metadata. By default ffmpeg preserves metadata applicable for both + # source and target containers. From this point the container won't change, + # so metadata can be added here. + if opts.addmetadata: + postprocessors.append({'key': 'FFmpegMetadata'}) if opts.convertsubtitles: postprocessors.append({ 'key': 'FFmpegSubtitlesConvertor', @@ -276,11 +291,6 @@ def _real_main(argv=None): }) if not already_have_thumbnail: opts.writethumbnail = True - # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and - # FFmpegExtractAudioPP as containers before conversion may not support - # metadata (3gp, webm, etc.) - if opts.addmetadata: - postprocessors.append({'key': 'FFmpegMetadata'}) # XAttrMetadataPP should be run after post-processors that may change file # contents if opts.xattrs: @@ -338,8 +348,10 @@ def _real_main(argv=None): 'retries': opts.retries, 'fragment_retries': opts.fragment_retries, 'skip_unavailable_fragments': opts.skip_unavailable_fragments, + 'keep_fragments': opts.keep_fragments, 'buffersize': opts.buffersize, 'noresizebuffer': opts.noresizebuffer, + 'http_chunk_size': opts.http_chunk_size, 'continuedl': opts.continue_dl, 'noprogress': opts.noprogress, 'progress_with_newline': opts.progress_with_newline, @@ -418,6 +430,7 @@ def _real_main(argv=None): 'config_location': opts.config_location, 'geo_bypass': opts.geo_bypass, 'geo_bypass_country': opts.geo_bypass_country, + 'geo_bypass_ip_block': opts.geo_bypass_ip_block, # just for deprecation check 'autonumber': opts.autonumber if opts.autonumber is True else None, 'usetitle': opts.usetitle if opts.usetitle is True else None, @@ -444,7 +457,7 @@ def _real_main(argv=None): try: if opts.load_info_filename is not None: - retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename)) + retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename)) else: retcode = ydl.download(all_urls) except MaxDownloadsReached: