yt_dlp/__init__.py

   1 #!/usr/bin/env python3
   2 f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp'  # noqa: F541
   3
   4 __license__ = 'Public Domain'
   5
   6 import itertools
   7 import os
   8 import random
   9 import re
  10 import sys
  11
  12 from .compat import compat_getpass, compat_os_name, compat_shlex_quote
  13 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  14 from .downloader import FileDownloader
  15 from .extractor import gen_extractors, list_extractors
  16 from .extractor.adobepass import MSO_INFO
  17 from .extractor.common import InfoExtractor
  18 from .options import parseOpts
  19 from .postprocessor import (
  20     FFmpegExtractAudioPP,
  21     FFmpegSubtitlesConvertorPP,
  22     FFmpegThumbnailsConvertorPP,
  23     FFmpegVideoConvertorPP,
  24     FFmpegVideoRemuxerPP,
  25     MetadataFromFieldPP,
  26     MetadataParserPP,
  27 )
  28 from .update import run_update
  29 from .utils import (
  30     NO_DEFAULT,
  31     DateRange,
  32     DownloadCancelled,
  33     DownloadError,
  34     GeoUtils,
  35     SameFileError,
  36     decodeOption,
  37     expand_path,
  38     float_or_none,
  39     int_or_none,
  40     match_filter_func,
  41     parse_duration,
  42     preferredencoding,
  43     read_batch_urls,
  44     render_table,
  45     setproctitle,
  46     std_headers,
  47     traverse_obj,
  48     write_string,
  49 )
  50 from .YoutubeDL import YoutubeDL
  51
  52
  53 def get_urls(urls, batchfile, verbose):
  54     # Batch file verification
  55     batch_urls = []
  56     if batchfile is not None:
  57         try:
  58             if batchfile == '-':
  59                 write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
  60                     'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
  61                 batchfd = sys.stdin
  62             else:
  63                 batchfd = open(
  64                     expand_path(batchfile), encoding='utf-8', errors='ignore')
  65             batch_urls = read_batch_urls(batchfd)
  66             if verbose:
  67                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  68         except OSError:
  69             sys.exit('ERROR: batch file %s could not be read' % batchfile)
  70     _enc = preferredencoding()
  71     return [
  72         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  73         for url in batch_urls + urls]
  74
  75
  76 def print_extractor_information(opts, urls):
  77     if opts.list_extractors:
  78         for ie in list_extractors(opts.age_limit):
  79             write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n', out=sys.stdout)
  80             matchedUrls = [url for url in urls if ie.suitable(url)]
  81             for mu in matchedUrls:
  82                 write_string('  ' + mu + '\n', out=sys.stdout)
  83     elif opts.list_extractor_descriptions:
  84         for ie in list_extractors(opts.age_limit):
  85             if not ie.working():
  86                 continue
  87             if ie.IE_DESC is False:
  88                 continue
  89             desc = ie.IE_DESC or ie.IE_NAME
  90             if getattr(ie, 'SEARCH_KEY', None) is not None:
  91                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
  92                 _COUNTS = ('', '5', '10', 'all')
  93                 desc += f'; "{ie.SEARCH_KEY}:" prefix (Example: "{ie.SEARCH_KEY}{random.choice(_COUNTS)}:{random.choice(_SEARCHES)}")'
  94             write_string(desc + '\n', out=sys.stdout)
  95     elif opts.ap_list_mso:
  96         table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]
  97         write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout)
  98     else:
  99         return False
 100     return True
 101
 102
 103 def set_compat_opts(opts):
 104     def _unused_compat_opt(name):
 105         if name not in opts.compat_opts:
 106             return False
 107         opts.compat_opts.discard(name)
 108         opts.compat_opts.update(['*%s' % name])
 109         return True
 110
 111     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 112         attr = getattr(opts, opt_name)
 113         if compat_name in opts.compat_opts:
 114             if attr is None:
 115                 setattr(opts, opt_name, not default)
 116                 return True
 117             else:
 118                 if remove_compat:
 119                     _unused_compat_opt(compat_name)
 120                 return False
 121         elif attr is None:
 122             setattr(opts, opt_name, default)
 123         return None
 124
 125     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 126     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 127     set_default_compat('no-clean-infojson', 'clean_infojson')
 128     if 'no-attach-info-json' in opts.compat_opts:
 129         if opts.embed_infojson:
 130             _unused_compat_opt('no-attach-info-json')
 131         else:
 132             opts.embed_infojson = False
 133     if 'format-sort' in opts.compat_opts:
 134         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 135     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 136     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 137     if _video_multistreams_set is False and _audio_multistreams_set is False:
 138         _unused_compat_opt('multistreams')
 139     if 'filename' in opts.compat_opts:
 140         if opts.outtmpl.get('default') is None:
 141             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 142         else:
 143             _unused_compat_opt('filename')
 144
 145
 146 def validate_options(opts):
 147     def validate(cndn, name, value=None, msg=None):
 148         if cndn:
 149             return True
 150         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 151
 152     def validate_in(name, value, items, msg=None):
 153         return validate(value is None or value in items, name, value, msg)
 154
 155     def validate_regex(name, value, regex):
 156         return validate(value is None or re.match(regex, value), name, value)
 157
 158     def validate_positive(name, value, strict=False):
 159         return validate(value is None or value > 0 or (not strict and value == 0),
 160                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 161
 162     def validate_minmax(min_val, max_val, min_name, max_name=None):
 163         if max_val is None or min_val is None or max_val >= min_val:
 164             return
 165         if not max_name:
 166             min_name, max_name = f'min {min_name}', f'max {min_name}'
 167         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 168
 169     # Usernames and passwords
 170     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 171              '.netrc', msg='using {name} conflicts with giving username/password')
 172     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 173     validate(opts.ap_password is None or opts.ap_username is not None,
 174              'TV Provider account username', msg='{name} missing')
 175     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 176                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 177
 178     # Numbers
 179     validate_positive('autonumber start', opts.autonumber_start)
 180     validate_positive('autonumber size', opts.autonumber_size, True)
 181     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 182     validate_positive('playlist start', opts.playliststart, True)
 183     if opts.playlistend != -1:
 184         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 185
 186     # Time ranges
 187     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 188     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 189     validate_positive('sleep interval', opts.sleep_interval)
 190     validate_positive('max sleep interval', opts.max_sleep_interval)
 191     if opts.sleep_interval is None:
 192         validate(
 193             opts.max_sleep_interval is None, 'min sleep interval',
 194             msg='{name} must be specified; use --min-sleep-interval')
 195     elif opts.max_sleep_interval is None:
 196         opts.max_sleep_interval = opts.sleep_interval
 197     else:
 198         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 199
 200     if opts.wait_for_video is not None:
 201         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 202         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 203                  'time range to wait for video', opts.wait_for_video)
 204         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 205         opts.wait_for_video = (min_wait, max_wait)
 206
 207     # Format sort
 208     for f in opts.format_sort:
 209         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 210
 211     # Postprocessor formats
 212     validate_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
 213     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 214     validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
 215     if opts.recodevideo is not None:
 216         opts.recodevideo = opts.recodevideo.replace(' ', '')
 217         validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 218     if opts.remuxvideo is not None:
 219         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 220         validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 221     if opts.audioquality:
 222         opts.audioquality = opts.audioquality.strip('k').strip('K')
 223         # int_or_none prevents inf, nan
 224         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 225
 226     # Retries
 227     def parse_retries(name, value):
 228         if value is None:
 229             return None
 230         elif value in ('inf', 'infinite'):
 231             return float('inf')
 232         try:
 233             return int(value)
 234         except (TypeError, ValueError):
 235             validate(False, f'{name} retry count', value)
 236
 237     opts.retries = parse_retries('download', opts.retries)
 238     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 239     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 240     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 241
 242     # Bytes
 243     def parse_bytes(name, value):
 244         if value is None:
 245             return None
 246         numeric_limit = FileDownloader.parse_bytes(value)
 247         validate(numeric_limit is not None, 'rate limit', value)
 248         return numeric_limit
 249
 250     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 251     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 252     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 253     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 254     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 255     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 256
 257     # Output templates
 258     def validate_outtmpl(tmpl, msg):
 259         err = YoutubeDL.validate_outtmpl(tmpl)
 260         if err:
 261             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 262
 263     for k, tmpl in opts.outtmpl.items():
 264         validate_outtmpl(tmpl, f'{k} output template')
 265     for type_, tmpl_list in opts.forceprint.items():
 266         for tmpl in tmpl_list:
 267             validate_outtmpl(tmpl, f'{type_} print template')
 268     for type_, tmpl_list in opts.print_to_file.items():
 269         for tmpl, file in tmpl_list:
 270             validate_outtmpl(tmpl, f'{type_} print to file template')
 271             validate_outtmpl(file, f'{type_} print to file filename')
 272     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 273     for k, tmpl in opts.progress_template.items():
 274         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 275         validate_outtmpl(tmpl, f'{k} template')
 276
 277     outtmpl_default = opts.outtmpl.get('default')
 278     if outtmpl_default == '':
 279         opts.skip_download = None
 280         del opts.outtmpl['default']
 281     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 282         raise ValueError(
 283             'Cannot download a video and extract audio into the same file! '
 284             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 285
 286     # Remove chapters
 287     remove_chapters_patterns, opts.remove_ranges = [], []
 288     for regex in opts.remove_chapters or []:
 289         if regex.startswith('*'):
 290             dur = list(map(parse_duration, regex[1:].split('-')))
 291             if len(dur) == 2 and all(t is not None for t in dur):
 292                 opts.remove_ranges.append(tuple(dur))
 293                 continue
 294             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 295         try:
 296             remove_chapters_patterns.append(re.compile(regex))
 297         except re.error as err:
 298             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 299     opts.remove_chapters = remove_chapters_patterns
 300
 301     # Cookies from browser
 302     if opts.cookiesfrombrowser:
 303         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 304         if mobj is None:
 305             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 306         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 307         browser_name = browser_name.lower()
 308         if browser_name not in SUPPORTED_BROWSERS:
 309             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 310                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 311         if keyring is not None:
 312             keyring = keyring.upper()
 313             if keyring not in SUPPORTED_KEYRINGS:
 314                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 315                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 316         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 317
 318     # MetadataParser
 319     def metadataparser_actions(f):
 320         if isinstance(f, str):
 321             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 322             try:
 323                 actions = [MetadataFromFieldPP.to_action(f)]
 324             except Exception as err:
 325                 raise ValueError(f'{cmd} is invalid; {err}')
 326         else:
 327             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 328             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 329
 330         for action in actions:
 331             try:
 332                 MetadataParserPP.validate_action(*action)
 333             except Exception as err:
 334                 raise ValueError(f'{cmd} is invalid; {err}')
 335             yield action
 336
 337     parse_metadata = opts.parse_metadata or []
 338     if opts.metafromtitle is not None:
 339         parse_metadata.append('title:%s' % opts.metafromtitle)
 340     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 341
 342     # Other options
 343     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 344     if geo_bypass_code is not None:
 345         try:
 346             GeoUtils.random_ipv4(geo_bypass_code)
 347         except Exception:
 348             raise ValueError('unsupported geo-bypass country or ip-block')
 349
 350     opts.match_filter = match_filter_func(opts.match_filter)
 351
 352     if opts.download_archive is not None:
 353         opts.download_archive = expand_path(opts.download_archive)
 354
 355     if opts.user_agent is not None:
 356         opts.headers.setdefault('User-Agent', opts.user_agent)
 357     if opts.referer is not None:
 358         opts.headers.setdefault('Referer', opts.referer)
 359
 360     if opts.no_sponsorblock:
 361         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 362
 363     warnings, deprecation_warnings = [], []
 364
 365     # Common mistake: -f best
 366     if opts.format == 'best':
 367         warnings.append('.\n         '.join((
 368             '"-f best" selects the best pre-merged format which is often not the best option',
 369             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 370             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 371
 372     # --(postprocessor/downloader)-args without name
 373     def report_args_compat(name, value, key1, key2=None):
 374         if key1 in value and key2 not in value:
 375             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 376             return True
 377         return False
 378
 379     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 380     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 381         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 382         opts.postprocessor_args.setdefault('sponskrub', [])
 383
 384     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 385                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 386         if val2 is NO_DEFAULT:
 387             val2 = getattr(opts, opt2)
 388         if not val2:
 389             return
 390
 391         if val1 is NO_DEFAULT:
 392             val1 = getattr(opts, opt1)
 393         if val1:
 394             warnings.append(f'{arg1} is ignored since {arg2} was given')
 395         setattr(opts, opt1, default)
 396
 397     # Conflicting options
 398     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 399     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 400     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 401                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 402     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 403     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 404     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 405     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 406     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 407     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters', val1=opts.sponskrub and opts.sponskrub_cut)
 408
 409     # Conflicts with --allow-unplayable-formats
 410     report_conflict('--add-metadata', 'addmetadata')
 411     report_conflict('--embed-chapters', 'addchapters')
 412     report_conflict('--embed-info-json', 'embed_infojson')
 413     report_conflict('--embed-subs', 'embedsubtitles')
 414     report_conflict('--embed-thumbnail', 'embedthumbnail')
 415     report_conflict('--extract-audio', 'extractaudio')
 416     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 417     report_conflict('--recode-video', 'recodevideo')
 418     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 419     report_conflict('--remux-video', 'remuxvideo')
 420     report_conflict('--sponskrub', 'sponskrub')
 421     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 422     report_conflict('--xattrs', 'xattrs')
 423
 424     # Fully deprecated options
 425     def report_deprecation(val, old, new=None):
 426         if not val:
 427             return
 428         deprecation_warnings.append(
 429             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 430             else f'{old} is deprecated and may not work as expected')
 431
 432     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 433     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 434     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 435     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 436     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 437
 438     # Dependent options
 439     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 440
 441     if opts.exec_before_dl_cmd:
 442         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 443
 444     if opts.useid:  # --id is not deprecated in youtube-dl
 445         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 446
 447     if opts.overwrites:  # --force-overwrites implies --no-continue
 448         opts.continue_dl = False
 449
 450     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 451         # Add chapters when adding metadata or marking sponsors
 452         opts.addchapters = True
 453
 454     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 455         # Do not unnecessarily download audio
 456         opts.format = 'bestaudio/best'
 457
 458     if opts.getcomments and opts.writeinfojson is None:
 459         # If JSON is not printed anywhere, but comments are requested, save it to file
 460         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 461             opts.writeinfojson = True
 462
 463     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 464         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 465         opts.writesubtitles = True
 466
 467     if opts.addmetadata and opts.embed_infojson is None:
 468         # If embedding metadata and infojson is present, embed it
 469         opts.embed_infojson = 'if_exists'
 470
 471     # Ask for passwords
 472     if opts.username is not None and opts.password is None:
 473         opts.password = compat_getpass('Type account password and press [Return]: ')
 474     if opts.ap_username is not None and opts.ap_password is None:
 475         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 476
 477     return warnings, deprecation_warnings
 478
 479
 480 def get_postprocessors(opts):
 481     yield from opts.add_postprocessors
 482
 483     if opts.parse_metadata:
 484         yield {
 485             'key': 'MetadataParser',
 486             'actions': opts.parse_metadata,
 487             'when': 'pre_process'
 488         }
 489     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 490     if sponsorblock_query:
 491         yield {
 492             'key': 'SponsorBlock',
 493             'categories': sponsorblock_query,
 494             'api': opts.sponsorblock_api,
 495             'when': 'after_filter'
 496         }
 497     if opts.convertsubtitles:
 498         yield {
 499             'key': 'FFmpegSubtitlesConvertor',
 500             'format': opts.convertsubtitles,
 501             'when': 'before_dl'
 502         }
 503     if opts.convertthumbnails:
 504         yield {
 505             'key': 'FFmpegThumbnailsConvertor',
 506             'format': opts.convertthumbnails,
 507             'when': 'before_dl'
 508         }
 509     if opts.extractaudio:
 510         yield {
 511             'key': 'FFmpegExtractAudio',
 512             'preferredcodec': opts.audioformat,
 513             'preferredquality': opts.audioquality,
 514             'nopostoverwrites': opts.nopostoverwrites,
 515         }
 516     if opts.remuxvideo:
 517         yield {
 518             'key': 'FFmpegVideoRemuxer',
 519             'preferedformat': opts.remuxvideo,
 520         }
 521     if opts.recodevideo:
 522         yield {
 523             'key': 'FFmpegVideoConvertor',
 524             'preferedformat': opts.recodevideo,
 525         }
 526     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 527     if opts.embedsubtitles:
 528         keep_subs = 'no-keep-subs' not in opts.compat_opts
 529         yield {
 530             'key': 'FFmpegEmbedSubtitle',
 531             # already_have_subtitle = True prevents the file from being deleted after embedding
 532             'already_have_subtitle': opts.writesubtitles and keep_subs
 533         }
 534         if not opts.writeautomaticsub and keep_subs:
 535             opts.writesubtitles = True
 536
 537     # ModifyChapters must run before FFmpegMetadataPP
 538     if opts.remove_chapters or sponsorblock_query:
 539         yield {
 540             'key': 'ModifyChapters',
 541             'remove_chapters_patterns': opts.remove_chapters,
 542             'remove_sponsor_segments': opts.sponsorblock_remove,
 543             'remove_ranges': opts.remove_ranges,
 544             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 545             'force_keyframes': opts.force_keyframes_at_cuts
 546         }
 547     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 548     # FFmpegExtractAudioPP as containers before conversion may not support
 549     # metadata (3gp, webm, etc.)
 550     # By default ffmpeg preserves metadata applicable for both
 551     # source and target containers. From this point the container won't change,
 552     # so metadata can be added here.
 553     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 554         yield {
 555             'key': 'FFmpegMetadata',
 556             'add_chapters': opts.addchapters,
 557             'add_metadata': opts.addmetadata,
 558             'add_infojson': opts.embed_infojson,
 559         }
 560     # Deprecated
 561     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 562     # but must be below EmbedSubtitle and FFmpegMetadata
 563     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 564     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 565     if opts.sponskrub is not False:
 566         yield {
 567             'key': 'SponSkrub',
 568             'path': opts.sponskrub_path,
 569             'args': opts.sponskrub_args,
 570             'cut': opts.sponskrub_cut,
 571             'force': opts.sponskrub_force,
 572             'ignoreerror': opts.sponskrub is None,
 573             '_from_cli': True,
 574         }
 575     if opts.embedthumbnail:
 576         yield {
 577             'key': 'EmbedThumbnail',
 578             # already_have_thumbnail = True prevents the file from being deleted after embedding
 579             'already_have_thumbnail': opts.writethumbnail
 580         }
 581         if not opts.writethumbnail:
 582             opts.writethumbnail = True
 583             opts.outtmpl['pl_thumbnail'] = ''
 584     if opts.split_chapters:
 585         yield {
 586             'key': 'FFmpegSplitChapters',
 587             'force_keyframes': opts.force_keyframes_at_cuts,
 588         }
 589     # XAttrMetadataPP should be run after post-processors that may change file contents
 590     if opts.xattrs:
 591         yield {'key': 'XAttrMetadata'}
 592     if opts.concat_playlist != 'never':
 593         yield {
 594             'key': 'FFmpegConcat',
 595             'only_multi_video': opts.concat_playlist != 'always',
 596             'when': 'playlist',
 597         }
 598     # Exec must be the last PP of each category
 599     for when, exec_cmd in opts.exec_cmd.items():
 600         yield {
 601             'key': 'Exec',
 602             'exec_cmd': exec_cmd,
 603             'when': when,
 604         }
 605
 606
 607 def parse_options(argv=None):
 608     """ @returns (parser, opts, urls, ydl_opts) """
 609     parser, opts, urls = parseOpts(argv)
 610     urls = get_urls(urls, opts.batchfile, opts.verbose)
 611
 612     set_compat_opts(opts)
 613     try:
 614         warnings, deprecation_warnings = validate_options(opts)
 615     except ValueError as err:
 616         parser.error(f'{err}\n')
 617
 618     postprocessors = list(get_postprocessors(opts))
 619
 620     any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
 621                    or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
 622                    or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
 623
 624     any_printing = opts.print_json
 625
 626     final_ext = (
 627         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 628         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 629         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 630         else None)
 631
 632     return parser, opts, urls, {
 633         'usenetrc': opts.usenetrc,
 634         'netrc_location': opts.netrc_location,
 635         'username': opts.username,
 636         'password': opts.password,
 637         'twofactor': opts.twofactor,
 638         'videopassword': opts.videopassword,
 639         'ap_mso': opts.ap_mso,
 640         'ap_username': opts.ap_username,
 641         'ap_password': opts.ap_password,
 642         'quiet': (opts.quiet or any_getting or any_printing),
 643         'no_warnings': opts.no_warnings,
 644         'forceurl': opts.geturl,
 645         'forcetitle': opts.gettitle,
 646         'forceid': opts.getid,
 647         'forcethumbnail': opts.getthumbnail,
 648         'forcedescription': opts.getdescription,
 649         'forceduration': opts.getduration,
 650         'forcefilename': opts.getfilename,
 651         'forceformat': opts.getformat,
 652         'forceprint': opts.forceprint,
 653         'print_to_file': opts.print_to_file,
 654         'forcejson': opts.dumpjson or opts.print_json,
 655         'dump_single_json': opts.dump_single_json,
 656         'force_write_download_archive': opts.force_write_download_archive,
 657         'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
 658         'skip_download': opts.skip_download,
 659         'format': opts.format,
 660         'allow_unplayable_formats': opts.allow_unplayable_formats,
 661         'ignore_no_formats_error': opts.ignore_no_formats_error,
 662         'format_sort': opts.format_sort,
 663         'format_sort_force': opts.format_sort_force,
 664         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 665         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 666         'check_formats': opts.check_formats,
 667         'listformats': opts.listformats,
 668         'listformats_table': opts.listformats_table,
 669         'outtmpl': opts.outtmpl,
 670         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 671         'paths': opts.paths,
 672         'autonumber_size': opts.autonumber_size,
 673         'autonumber_start': opts.autonumber_start,
 674         'restrictfilenames': opts.restrictfilenames,
 675         'windowsfilenames': opts.windowsfilenames,
 676         'ignoreerrors': opts.ignoreerrors,
 677         'force_generic_extractor': opts.force_generic_extractor,
 678         'ratelimit': opts.ratelimit,
 679         'throttledratelimit': opts.throttledratelimit,
 680         'overwrites': opts.overwrites,
 681         'retries': opts.retries,
 682         'file_access_retries': opts.file_access_retries,
 683         'fragment_retries': opts.fragment_retries,
 684         'extractor_retries': opts.extractor_retries,
 685         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 686         'keep_fragments': opts.keep_fragments,
 687         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 688         'buffersize': opts.buffersize,
 689         'noresizebuffer': opts.noresizebuffer,
 690         'http_chunk_size': opts.http_chunk_size,
 691         'continuedl': opts.continue_dl,
 692         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 693         'progress_with_newline': opts.progress_with_newline,
 694         'progress_template': opts.progress_template,
 695         'playliststart': opts.playliststart,
 696         'playlistend': opts.playlistend,
 697         'playlistreverse': opts.playlist_reverse,
 698         'playlistrandom': opts.playlist_random,
 699         'noplaylist': opts.noplaylist,
 700         'logtostderr': opts.outtmpl.get('default') == '-',
 701         'consoletitle': opts.consoletitle,
 702         'nopart': opts.nopart,
 703         'updatetime': opts.updatetime,
 704         'writedescription': opts.writedescription,
 705         'writeannotations': opts.writeannotations,
 706         'writeinfojson': opts.writeinfojson,
 707         'allow_playlist_files': opts.allow_playlist_files,
 708         'clean_infojson': opts.clean_infojson,
 709         'getcomments': opts.getcomments,
 710         'writethumbnail': opts.writethumbnail is True,
 711         'write_all_thumbnails': opts.writethumbnail == 'all',
 712         'writelink': opts.writelink,
 713         'writeurllink': opts.writeurllink,
 714         'writewebloclink': opts.writewebloclink,
 715         'writedesktoplink': opts.writedesktoplink,
 716         'writesubtitles': opts.writesubtitles,
 717         'writeautomaticsub': opts.writeautomaticsub,
 718         'allsubtitles': opts.allsubtitles,
 719         'listsubtitles': opts.listsubtitles,
 720         'subtitlesformat': opts.subtitlesformat,
 721         'subtitleslangs': opts.subtitleslangs,
 722         'matchtitle': decodeOption(opts.matchtitle),
 723         'rejecttitle': decodeOption(opts.rejecttitle),
 724         'max_downloads': opts.max_downloads,
 725         'prefer_free_formats': opts.prefer_free_formats,
 726         'trim_file_name': opts.trim_file_name,
 727         'verbose': opts.verbose,
 728         'dump_intermediate_pages': opts.dump_intermediate_pages,
 729         'write_pages': opts.write_pages,
 730         'test': opts.test,
 731         'keepvideo': opts.keepvideo,
 732         'min_filesize': opts.min_filesize,
 733         'max_filesize': opts.max_filesize,
 734         'min_views': opts.min_views,
 735         'max_views': opts.max_views,
 736         'daterange': opts.date,
 737         'cachedir': opts.cachedir,
 738         'youtube_print_sig_code': opts.youtube_print_sig_code,
 739         'age_limit': opts.age_limit,
 740         'download_archive': opts.download_archive,
 741         'break_on_existing': opts.break_on_existing,
 742         'break_on_reject': opts.break_on_reject,
 743         'break_per_url': opts.break_per_url,
 744         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 745         'cookiefile': opts.cookiefile,
 746         'cookiesfrombrowser': opts.cookiesfrombrowser,
 747         'legacyserverconnect': opts.legacy_server_connect,
 748         'nocheckcertificate': opts.no_check_certificate,
 749         'prefer_insecure': opts.prefer_insecure,
 750         'http_headers': opts.headers,
 751         'proxy': opts.proxy,
 752         'socket_timeout': opts.socket_timeout,
 753         'bidi_workaround': opts.bidi_workaround,
 754         'debug_printtraffic': opts.debug_printtraffic,
 755         'prefer_ffmpeg': opts.prefer_ffmpeg,
 756         'include_ads': opts.include_ads,
 757         'default_search': opts.default_search,
 758         'dynamic_mpd': opts.dynamic_mpd,
 759         'extractor_args': opts.extractor_args,
 760         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 761         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 762         'encoding': opts.encoding,
 763         'extract_flat': opts.extract_flat,
 764         'live_from_start': opts.live_from_start,
 765         'wait_for_video': opts.wait_for_video,
 766         'mark_watched': opts.mark_watched,
 767         'merge_output_format': opts.merge_output_format,
 768         'final_ext': final_ext,
 769         'postprocessors': postprocessors,
 770         'fixup': opts.fixup,
 771         'source_address': opts.source_address,
 772         'call_home': opts.call_home,
 773         'sleep_interval_requests': opts.sleep_interval_requests,
 774         'sleep_interval': opts.sleep_interval,
 775         'max_sleep_interval': opts.max_sleep_interval,
 776         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 777         'external_downloader': opts.external_downloader,
 778         'list_thumbnails': opts.list_thumbnails,
 779         'playlist_items': opts.playlist_items,
 780         'xattr_set_filesize': opts.xattr_set_filesize,
 781         'match_filter': opts.match_filter,
 782         'no_color': opts.no_color,
 783         'ffmpeg_location': opts.ffmpeg_location,
 784         'hls_prefer_native': opts.hls_prefer_native,
 785         'hls_use_mpegts': opts.hls_use_mpegts,
 786         'hls_split_discontinuity': opts.hls_split_discontinuity,
 787         'external_downloader_args': opts.external_downloader_args,
 788         'postprocessor_args': opts.postprocessor_args,
 789         'cn_verification_proxy': opts.cn_verification_proxy,
 790         'geo_verification_proxy': opts.geo_verification_proxy,
 791         'geo_bypass': opts.geo_bypass,
 792         'geo_bypass_country': opts.geo_bypass_country,
 793         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 794         '_warnings': warnings,
 795         '_deprecation_warnings': deprecation_warnings,
 796         'compat_opts': opts.compat_opts,
 797     }
 798
 799
 800 def _real_main(argv=None):
 801     setproctitle('yt-dlp')
 802
 803     parser, opts, all_urls, ydl_opts = parse_options(argv)
 804
 805     # Dump user agent
 806     if opts.dump_user_agent:
 807         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 808         write_string(f'{ua}\n', out=sys.stdout)
 809         sys.exit(0)
 810
 811     if print_extractor_information(opts, all_urls):
 812         sys.exit(0)
 813
 814     with YoutubeDL(ydl_opts) as ydl:
 815         actual_use = all_urls or opts.load_info_filename
 816
 817         # Remove cache dir
 818         if opts.rm_cachedir:
 819             ydl.cache.remove()
 820
 821         # Update version
 822         if opts.update_self:
 823             # If updater returns True, exit. Required for windows
 824             if run_update(ydl):
 825                 if actual_use:
 826                     sys.exit('ERROR: The program must exit for the update to complete')
 827                 sys.exit()
 828
 829         # Maybe do nothing
 830         if not actual_use:
 831             if opts.update_self or opts.rm_cachedir:
 832                 sys.exit()
 833
 834             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 835             parser.error(
 836                 'You must provide at least one URL.\n'
 837                 'Type yt-dlp --help to see a list of all options.')
 838
 839         try:
 840             if opts.load_info_filename is not None:
 841                 retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
 842             else:
 843                 retcode = ydl.download(all_urls)
 844         except DownloadCancelled:
 845             ydl.to_screen('Aborting remaining downloads')
 846             retcode = 101
 847
 848     sys.exit(retcode)
 849
 850
 851 def main(argv=None):
 852     try:
 853         _real_main(argv)
 854     except DownloadError:
 855         sys.exit(1)
 856     except SameFileError as e:
 857         sys.exit(f'ERROR: {e}')
 858     except KeyboardInterrupt:
 859         sys.exit('\nERROR: Interrupted by user')
 860     except BrokenPipeError as e:
 861         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 862         devnull = os.open(os.devnull, os.O_WRONLY)
 863         os.dup2(devnull, sys.stdout.fileno())
 864         sys.exit(f'\nERROR: {e}')
 865
 866
 867 __all__ = [
 868     'main',
 869     'YoutubeDL',
 870     'parse_options',
 871     'gen_extractors',
 872     'list_extractors',
 873 ]