yt_dlp/__init__.py

   1 try:
   2     import contextvars  # noqa: F401
   3 except Exception:
   4     raise Exception(
   5         f'You are using an unsupported version of Python. Only Python versions 3.7 and above are supported by yt-dlp')  # noqa: F541
   6
   7 __license__ = 'Public Domain'
   8
   9 import collections
  10 import getpass
  11 import itertools
  12 import optparse
  13 import os
  14 import re
  15 import sys
  16 import traceback
  17
  18 from .compat import compat_shlex_quote
  19 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  20 from .downloader.external import get_external_downloader
  21 from .extractor import list_extractor_classes
  22 from .extractor.adobepass import MSO_INFO
  23 from .options import parseOpts
  24 from .postprocessor import (
  25     FFmpegExtractAudioPP,
  26     FFmpegMergerPP,
  27     FFmpegPostProcessor,
  28     FFmpegSubtitlesConvertorPP,
  29     FFmpegThumbnailsConvertorPP,
  30     FFmpegVideoConvertorPP,
  31     FFmpegVideoRemuxerPP,
  32     MetadataFromFieldPP,
  33     MetadataParserPP,
  34 )
  35 from .update import Updater
  36 from .utils import (
  37     NO_DEFAULT,
  38     POSTPROCESS_WHEN,
  39     DateRange,
  40     DownloadCancelled,
  41     DownloadError,
  42     FormatSorter,
  43     GeoUtils,
  44     PlaylistEntries,
  45     SameFileError,
  46     decodeOption,
  47     download_range_func,
  48     expand_path,
  49     float_or_none,
  50     format_field,
  51     int_or_none,
  52     match_filter_func,
  53     parse_bytes,
  54     parse_duration,
  55     preferredencoding,
  56     read_batch_urls,
  57     read_stdin,
  58     render_table,
  59     setproctitle,
  60     std_headers,
  61     traverse_obj,
  62     variadic,
  63     write_string,
  64 )
  65 from .YoutubeDL import YoutubeDL
  66
  67 _IN_CLI = False
  68
  69
  70 def _exit(status=0, *args):
  71     for msg in args:
  72         sys.stderr.write(msg)
  73     raise SystemExit(status)
  74
  75
  76 def get_urls(urls, batchfile, verbose):
  77     # Batch file verification
  78     batch_urls = []
  79     if batchfile is not None:
  80         try:
  81             batch_urls = read_batch_urls(
  82                 read_stdin('URLs') if batchfile == '-'
  83                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  84             if verbose:
  85                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  86         except OSError:
  87             _exit(f'ERROR: batch file {batchfile} could not be read')
  88     _enc = preferredencoding()
  89     return [
  90         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  91         for url in batch_urls + urls]
  92
  93
  94 def print_extractor_information(opts, urls):
  95     out = ''
  96     if opts.list_extractors:
  97         # Importing GenericIE is currently slow since it imports YoutubeIE
  98         from .extractor.generic import GenericIE
  99
 100         urls = dict.fromkeys(urls, False)
 101         for ie in list_extractor_classes(opts.age_limit):
 102             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
 103             if ie == GenericIE:
 104                 matched_urls = [url for url, matched in urls.items() if not matched]
 105             else:
 106                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
 107                 urls.update(dict.fromkeys(matched_urls, True))
 108             out += ''.join(f'  {url}\n' for url in matched_urls)
 109     elif opts.list_extractor_descriptions:
 110         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 111         out = '\n'.join(
 112             ie.description(markdown=False, search_examples=_SEARCHES)
 113             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
 114     elif opts.ap_list_mso:
 115         out = 'Supported TV Providers:\n%s\n' % render_table(
 116             ['mso', 'mso name'],
 117             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 118     else:
 119         return False
 120     write_string(out, out=sys.stdout)
 121     return True
 122
 123
 124 def set_compat_opts(opts):
 125     def _unused_compat_opt(name):
 126         if name not in opts.compat_opts:
 127             return False
 128         opts.compat_opts.discard(name)
 129         opts.compat_opts.update(['*%s' % name])
 130         return True
 131
 132     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 133         attr = getattr(opts, opt_name)
 134         if compat_name in opts.compat_opts:
 135             if attr is None:
 136                 setattr(opts, opt_name, not default)
 137                 return True
 138             else:
 139                 if remove_compat:
 140                     _unused_compat_opt(compat_name)
 141                 return False
 142         elif attr is None:
 143             setattr(opts, opt_name, default)
 144         return None
 145
 146     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 147     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 148     set_default_compat('no-clean-infojson', 'clean_infojson')
 149     if 'no-attach-info-json' in opts.compat_opts:
 150         if opts.embed_infojson:
 151             _unused_compat_opt('no-attach-info-json')
 152         else:
 153             opts.embed_infojson = False
 154     if 'format-sort' in opts.compat_opts:
 155         opts.format_sort.extend(FormatSorter.ytdl_default)
 156     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 157     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 158     if _video_multistreams_set is False and _audio_multistreams_set is False:
 159         _unused_compat_opt('multistreams')
 160     if 'filename' in opts.compat_opts:
 161         if opts.outtmpl.get('default') is None:
 162             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 163         else:
 164             _unused_compat_opt('filename')
 165
 166
 167 def validate_options(opts):
 168     def validate(cndn, name, value=None, msg=None):
 169         if cndn:
 170             return True
 171         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 172
 173     def validate_in(name, value, items, msg=None):
 174         return validate(value is None or value in items, name, value, msg)
 175
 176     def validate_regex(name, value, regex):
 177         return validate(value is None or re.match(regex, value), name, value)
 178
 179     def validate_positive(name, value, strict=False):
 180         return validate(value is None or value > 0 or (not strict and value == 0),
 181                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 182
 183     def validate_minmax(min_val, max_val, min_name, max_name=None):
 184         if max_val is None or min_val is None or max_val >= min_val:
 185             return
 186         if not max_name:
 187             min_name, max_name = f'min {min_name}', f'max {min_name}'
 188         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 189
 190     # Usernames and passwords
 191     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 192              '.netrc', msg='using {name} conflicts with giving username/password')
 193     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 194     validate(opts.ap_password is None or opts.ap_username is not None,
 195              'TV Provider account username', msg='{name} missing')
 196     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 197                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 198
 199     # Numbers
 200     validate_positive('autonumber start', opts.autonumber_start)
 201     validate_positive('autonumber size', opts.autonumber_size, True)
 202     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 203     validate_positive('playlist start', opts.playliststart, True)
 204     if opts.playlistend != -1:
 205         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 206
 207     # Time ranges
 208     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 209     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 210     validate_positive('sleep interval', opts.sleep_interval)
 211     validate_positive('max sleep interval', opts.max_sleep_interval)
 212     if opts.sleep_interval is None:
 213         validate(
 214             opts.max_sleep_interval is None, 'min sleep interval',
 215             msg='{name} must be specified; use --min-sleep-interval')
 216     elif opts.max_sleep_interval is None:
 217         opts.max_sleep_interval = opts.sleep_interval
 218     else:
 219         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 220
 221     if opts.wait_for_video is not None:
 222         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 223         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 224                  'time range to wait for video', opts.wait_for_video)
 225         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 226         opts.wait_for_video = (min_wait, max_wait)
 227
 228     # Format sort
 229     for f in opts.format_sort:
 230         validate_regex('format sorting', f, FormatSorter.regex)
 231
 232     # Postprocessor formats
 233     validate_regex('merge output format', opts.merge_output_format,
 234                    r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
 235     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 236     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 237     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 238     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 239     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 240     if opts.audioquality:
 241         opts.audioquality = opts.audioquality.strip('k').strip('K')
 242         # int_or_none prevents inf, nan
 243         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 244
 245     # Retries
 246     def parse_retries(name, value):
 247         if value is None:
 248             return None
 249         elif value in ('inf', 'infinite'):
 250             return float('inf')
 251         try:
 252             return int(value)
 253         except (TypeError, ValueError):
 254             validate(False, f'{name} retry count', value)
 255
 256     opts.retries = parse_retries('download', opts.retries)
 257     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 258     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 259     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 260
 261     # Retry sleep function
 262     def parse_sleep_func(expr):
 263         NUMBER_RE = r'\d+(?:\.\d+)?'
 264         op, start, limit, step, *_ = tuple(re.fullmatch(
 265             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 266             expr.strip()).groups()) + (None, None)
 267
 268         if op == 'exp':
 269             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 270         else:
 271             default_step = start if op or limit else 0
 272             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 273
 274     for key, expr in opts.retry_sleep.items():
 275         if not expr:
 276             del opts.retry_sleep[key]
 277             continue
 278         try:
 279             opts.retry_sleep[key] = parse_sleep_func(expr)
 280         except AttributeError:
 281             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 282
 283     # Bytes
 284     def validate_bytes(name, value):
 285         if value is None:
 286             return None
 287         numeric_limit = parse_bytes(value)
 288         validate(numeric_limit is not None, 'rate limit', value)
 289         return numeric_limit
 290
 291     opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
 292     opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
 293     opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
 294     opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
 295     opts.buffersize = validate_bytes('buffer size', opts.buffersize)
 296     opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
 297
 298     # Output templates
 299     def validate_outtmpl(tmpl, msg):
 300         err = YoutubeDL.validate_outtmpl(tmpl)
 301         if err:
 302             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 303
 304     for k, tmpl in opts.outtmpl.items():
 305         validate_outtmpl(tmpl, f'{k} output template')
 306     for type_, tmpl_list in opts.forceprint.items():
 307         for tmpl in tmpl_list:
 308             validate_outtmpl(tmpl, f'{type_} print template')
 309     for type_, tmpl_list in opts.print_to_file.items():
 310         for tmpl, file in tmpl_list:
 311             validate_outtmpl(tmpl, f'{type_} print to file template')
 312             validate_outtmpl(file, f'{type_} print to file filename')
 313     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 314     for k, tmpl in opts.progress_template.items():
 315         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 316         validate_outtmpl(tmpl, f'{k} template')
 317
 318     outtmpl_default = opts.outtmpl.get('default')
 319     if outtmpl_default == '':
 320         opts.skip_download = None
 321         del opts.outtmpl['default']
 322
 323     def parse_chapters(name, value):
 324         chapters, ranges = [], []
 325         parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
 326         for regex in value or []:
 327             if regex.startswith('*'):
 328                 for range_ in map(str.strip, regex[1:].split(',')):
 329                     mobj = range_ != '-' and re.fullmatch(r'([^-]+)?\s*-\s*([^-]+)?', range_)
 330                     dur = mobj and (parse_timestamp(mobj.group(1) or '0'), parse_timestamp(mobj.group(2) or 'inf'))
 331                     if None in (dur or [None]):
 332                         raise ValueError(f'invalid {name} time range "{regex}". Must be of the form "*start-end"')
 333                     ranges.append(dur)
 334                 continue
 335             try:
 336                 chapters.append(re.compile(regex))
 337             except re.error as err:
 338                 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 339         return chapters, ranges
 340
 341     opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
 342     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
 343
 344     # Cookies from browser
 345     if opts.cookiesfrombrowser:
 346         container = None
 347         mobj = re.fullmatch(r'''(?x)
 348             (?P<name>[^+:]+)
 349             (?:\s*\+\s*(?P<keyring>[^:]+))?
 350             (?:\s*:\s*(?!:)(?P<profile>.+?))?
 351             (?:\s*::\s*(?P<container>.+))?
 352         ''', opts.cookiesfrombrowser)
 353         if mobj is None:
 354             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 355         browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
 356         browser_name = browser_name.lower()
 357         if browser_name not in SUPPORTED_BROWSERS:
 358             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 359                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 360         if keyring is not None:
 361             keyring = keyring.upper()
 362             if keyring not in SUPPORTED_KEYRINGS:
 363                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 364                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 365         opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
 366
 367     # MetadataParser
 368     def metadataparser_actions(f):
 369         if isinstance(f, str):
 370             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 371             try:
 372                 actions = [MetadataFromFieldPP.to_action(f)]
 373             except Exception as err:
 374                 raise ValueError(f'{cmd} is invalid; {err}')
 375         else:
 376             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 377             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 378
 379         for action in actions:
 380             try:
 381                 MetadataParserPP.validate_action(*action)
 382             except Exception as err:
 383                 raise ValueError(f'{cmd} is invalid; {err}')
 384             yield action
 385
 386     if opts.metafromtitle is not None:
 387         opts.parse_metadata.setdefault('pre_process', []).append('title:%s' % opts.metafromtitle)
 388     opts.parse_metadata = {
 389         k: list(itertools.chain(*map(metadataparser_actions, v)))
 390         for k, v in opts.parse_metadata.items()
 391     }
 392
 393     # Other options
 394     if opts.playlist_items is not None:
 395         try:
 396             tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
 397         except Exception as err:
 398             raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
 399
 400     opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
 401     if opts.geo_bypass.lower() not in ('default', 'never'):
 402         try:
 403             GeoUtils.random_ipv4(opts.geo_bypass)
 404         except Exception:
 405             raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
 406         if len(opts.geo_bypass) == 2:
 407             opts.geo_bypass_country = opts.geo_bypass
 408         else:
 409             opts.geo_bypass_ip_block = opts.geo_bypass
 410     opts.geo_bypass = opts.geo_bypass.lower() != 'never'
 411
 412     opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
 413
 414     if opts.download_archive is not None:
 415         opts.download_archive = expand_path(opts.download_archive)
 416
 417     if opts.ffmpeg_location is not None:
 418         opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
 419
 420     if opts.user_agent is not None:
 421         opts.headers.setdefault('User-Agent', opts.user_agent)
 422     if opts.referer is not None:
 423         opts.headers.setdefault('Referer', opts.referer)
 424
 425     if opts.no_sponsorblock:
 426         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 427
 428     default_downloader = None
 429     for proto, path in opts.external_downloader.items():
 430         if path == 'native':
 431             continue
 432         ed = get_external_downloader(path)
 433         if ed is None:
 434             raise ValueError(
 435                 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
 436         elif ed and proto == 'default':
 437             default_downloader = ed.get_basename()
 438
 439     for policy in opts.color.values():
 440         if policy not in ('always', 'auto', 'no_color', 'never'):
 441             raise ValueError(f'"{policy}" is not a valid color policy')
 442
 443     warnings, deprecation_warnings = [], []
 444
 445     # Common mistake: -f best
 446     if opts.format == 'best':
 447         warnings.append('.\n         '.join((
 448             '"-f best" selects the best pre-merged format which is often not the best option',
 449             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 450             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 451
 452     # --(postprocessor/downloader)-args without name
 453     def report_args_compat(name, value, key1, key2=None, where=None):
 454         if key1 in value and key2 not in value:
 455             warnings.append(f'{name.title()} arguments given without specifying name. '
 456                             f'The arguments will be given to {where or f"all {name}s"}')
 457             return True
 458         return False
 459
 460     if report_args_compat('external downloader', opts.external_downloader_args,
 461                           'default', where=default_downloader) and default_downloader:
 462         # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
 463         opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
 464
 465     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 466         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 467         opts.postprocessor_args.setdefault('sponskrub', [])
 468
 469     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 470                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 471         if val2 is NO_DEFAULT:
 472             val2 = getattr(opts, opt2)
 473         if not val2:
 474             return
 475
 476         if val1 is NO_DEFAULT:
 477             val1 = getattr(opts, opt1)
 478         if val1:
 479             warnings.append(f'{arg1} is ignored since {arg2} was given')
 480         setattr(opts, opt1, default)
 481
 482     # Conflicting options
 483     report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
 484     report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
 485     report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
 486     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 487     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 488     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 489                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 490     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 491     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 492     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 493     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 494     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 495     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 496                     val1=opts.sponskrub and opts.sponskrub_cut)
 497
 498     # Conflicts with --allow-unplayable-formats
 499     report_conflict('--embed-metadata', 'addmetadata')
 500     report_conflict('--embed-chapters', 'addchapters')
 501     report_conflict('--embed-info-json', 'embed_infojson')
 502     report_conflict('--embed-subs', 'embedsubtitles')
 503     report_conflict('--embed-thumbnail', 'embedthumbnail')
 504     report_conflict('--extract-audio', 'extractaudio')
 505     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 506     report_conflict('--recode-video', 'recodevideo')
 507     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 508     report_conflict('--remux-video', 'remuxvideo')
 509     report_conflict('--sponskrub', 'sponskrub')
 510     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 511     report_conflict('--xattrs', 'xattrs')
 512
 513     # Fully deprecated options
 514     def report_deprecation(val, old, new=None):
 515         if not val:
 516             return
 517         deprecation_warnings.append(
 518             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 519             else f'{old} is deprecated and may not work as expected')
 520
 521     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 522     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 523     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 524     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 525     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 526
 527     # Dependent options
 528     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 529
 530     if opts.exec_before_dl_cmd:
 531         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 532
 533     if opts.useid:  # --id is not deprecated in youtube-dl
 534         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 535
 536     if opts.overwrites:  # --force-overwrites implies --no-continue
 537         opts.continue_dl = False
 538
 539     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 540         # Add chapters when adding metadata or marking sponsors
 541         opts.addchapters = True
 542
 543     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 544         # Do not unnecessarily download audio
 545         opts.format = 'bestaudio/best'
 546
 547     if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
 548         # If JSON is not printed anywhere, but comments are requested, save it to file
 549         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 550             opts.writeinfojson = True
 551
 552     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 553         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 554         opts.writesubtitles = True
 555
 556     if opts.addmetadata and opts.embed_infojson is None:
 557         # If embedding metadata and infojson is present, embed it
 558         opts.embed_infojson = 'if_exists'
 559
 560     # Ask for passwords
 561     if opts.username is not None and opts.password is None:
 562         opts.password = getpass.getpass('Type account password and press [Return]: ')
 563     if opts.ap_username is not None and opts.ap_password is None:
 564         opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 565
 566     return warnings, deprecation_warnings
 567
 568
 569 def get_postprocessors(opts):
 570     yield from opts.add_postprocessors
 571
 572     for when, actions in opts.parse_metadata.items():
 573         yield {
 574             'key': 'MetadataParser',
 575             'actions': actions,
 576             'when': when
 577         }
 578     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 579     if sponsorblock_query:
 580         yield {
 581             'key': 'SponsorBlock',
 582             'categories': sponsorblock_query,
 583             'api': opts.sponsorblock_api,
 584             'when': 'after_filter'
 585         }
 586     if opts.convertsubtitles:
 587         yield {
 588             'key': 'FFmpegSubtitlesConvertor',
 589             'format': opts.convertsubtitles,
 590             'when': 'before_dl'
 591         }
 592     if opts.convertthumbnails:
 593         yield {
 594             'key': 'FFmpegThumbnailsConvertor',
 595             'format': opts.convertthumbnails,
 596             'when': 'before_dl'
 597         }
 598     if opts.extractaudio:
 599         yield {
 600             'key': 'FFmpegExtractAudio',
 601             'preferredcodec': opts.audioformat,
 602             'preferredquality': opts.audioquality,
 603             'nopostoverwrites': opts.nopostoverwrites,
 604         }
 605     if opts.remuxvideo:
 606         yield {
 607             'key': 'FFmpegVideoRemuxer',
 608             'preferedformat': opts.remuxvideo,
 609         }
 610     if opts.recodevideo:
 611         yield {
 612             'key': 'FFmpegVideoConvertor',
 613             'preferedformat': opts.recodevideo,
 614         }
 615     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 616     if opts.embedsubtitles:
 617         keep_subs = 'no-keep-subs' not in opts.compat_opts
 618         yield {
 619             'key': 'FFmpegEmbedSubtitle',
 620             # already_have_subtitle = True prevents the file from being deleted after embedding
 621             'already_have_subtitle': opts.writesubtitles and keep_subs
 622         }
 623         if not opts.writeautomaticsub and keep_subs:
 624             opts.writesubtitles = True
 625
 626     # ModifyChapters must run before FFmpegMetadataPP
 627     if opts.remove_chapters or sponsorblock_query:
 628         yield {
 629             'key': 'ModifyChapters',
 630             'remove_chapters_patterns': opts.remove_chapters,
 631             'remove_sponsor_segments': opts.sponsorblock_remove,
 632             'remove_ranges': opts.remove_ranges,
 633             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 634             'force_keyframes': opts.force_keyframes_at_cuts
 635         }
 636     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 637     # FFmpegExtractAudioPP as containers before conversion may not support
 638     # metadata (3gp, webm, etc.)
 639     # By default ffmpeg preserves metadata applicable for both
 640     # source and target containers. From this point the container won't change,
 641     # so metadata can be added here.
 642     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 643         yield {
 644             'key': 'FFmpegMetadata',
 645             'add_chapters': opts.addchapters,
 646             'add_metadata': opts.addmetadata,
 647             'add_infojson': opts.embed_infojson,
 648         }
 649     # Deprecated
 650     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 651     # but must be below EmbedSubtitle and FFmpegMetadata
 652     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 653     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 654     if opts.sponskrub is not False:
 655         yield {
 656             'key': 'SponSkrub',
 657             'path': opts.sponskrub_path,
 658             'args': opts.sponskrub_args,
 659             'cut': opts.sponskrub_cut,
 660             'force': opts.sponskrub_force,
 661             'ignoreerror': opts.sponskrub is None,
 662             '_from_cli': True,
 663         }
 664     if opts.embedthumbnail:
 665         yield {
 666             'key': 'EmbedThumbnail',
 667             # already_have_thumbnail = True prevents the file from being deleted after embedding
 668             'already_have_thumbnail': opts.writethumbnail
 669         }
 670         if not opts.writethumbnail:
 671             opts.writethumbnail = True
 672             opts.outtmpl['pl_thumbnail'] = ''
 673     if opts.split_chapters:
 674         yield {
 675             'key': 'FFmpegSplitChapters',
 676             'force_keyframes': opts.force_keyframes_at_cuts,
 677         }
 678     # XAttrMetadataPP should be run after post-processors that may change file contents
 679     if opts.xattrs:
 680         yield {'key': 'XAttrMetadata'}
 681     if opts.concat_playlist != 'never':
 682         yield {
 683             'key': 'FFmpegConcat',
 684             'only_multi_video': opts.concat_playlist != 'always',
 685             'when': 'playlist',
 686         }
 687     # Exec must be the last PP of each category
 688     for when, exec_cmd in opts.exec_cmd.items():
 689         yield {
 690             'key': 'Exec',
 691             'exec_cmd': exec_cmd,
 692             'when': when,
 693         }
 694
 695
 696 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
 697
 698
 699 def parse_options(argv=None):
 700     """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
 701     parser, opts, urls = parseOpts(argv)
 702     urls = get_urls(urls, opts.batchfile, opts.verbose)
 703
 704     set_compat_opts(opts)
 705     try:
 706         warnings, deprecation_warnings = validate_options(opts)
 707     except ValueError as err:
 708         parser.error(f'{err}\n')
 709
 710     postprocessors = list(get_postprocessors(opts))
 711
 712     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
 713     any_getting = any(getattr(opts, k) for k in (
 714         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 715         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 716     ))
 717     if opts.quiet is None:
 718         opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
 719
 720     playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
 721     write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
 722                                and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
 723     if not any((
 724         opts.extract_flat,
 725         opts.dump_single_json,
 726         opts.forceprint.get('playlist'),
 727         opts.print_to_file.get('playlist'),
 728         write_playlist_infojson,
 729     )):
 730         if not playlist_pps:
 731             opts.extract_flat = 'discard'
 732         elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
 733             opts.extract_flat = 'discard_in_playlist'
 734
 735     final_ext = (
 736         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 737         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 738         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 739         else None)
 740
 741     return ParsedOptions(parser, opts, urls, {
 742         'usenetrc': opts.usenetrc,
 743         'netrc_location': opts.netrc_location,
 744         'username': opts.username,
 745         'password': opts.password,
 746         'twofactor': opts.twofactor,
 747         'videopassword': opts.videopassword,
 748         'ap_mso': opts.ap_mso,
 749         'ap_username': opts.ap_username,
 750         'ap_password': opts.ap_password,
 751         'client_certificate': opts.client_certificate,
 752         'client_certificate_key': opts.client_certificate_key,
 753         'client_certificate_password': opts.client_certificate_password,
 754         'quiet': opts.quiet,
 755         'no_warnings': opts.no_warnings,
 756         'forceurl': opts.geturl,
 757         'forcetitle': opts.gettitle,
 758         'forceid': opts.getid,
 759         'forcethumbnail': opts.getthumbnail,
 760         'forcedescription': opts.getdescription,
 761         'forceduration': opts.getduration,
 762         'forcefilename': opts.getfilename,
 763         'forceformat': opts.getformat,
 764         'forceprint': opts.forceprint,
 765         'print_to_file': opts.print_to_file,
 766         'forcejson': opts.dumpjson or opts.print_json,
 767         'dump_single_json': opts.dump_single_json,
 768         'force_write_download_archive': opts.force_write_download_archive,
 769         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 770         'skip_download': opts.skip_download,
 771         'format': opts.format,
 772         'allow_unplayable_formats': opts.allow_unplayable_formats,
 773         'ignore_no_formats_error': opts.ignore_no_formats_error,
 774         'format_sort': opts.format_sort,
 775         'format_sort_force': opts.format_sort_force,
 776         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 777         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 778         'check_formats': opts.check_formats,
 779         'listformats': opts.listformats,
 780         'listformats_table': opts.listformats_table,
 781         'outtmpl': opts.outtmpl,
 782         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 783         'paths': opts.paths,
 784         'autonumber_size': opts.autonumber_size,
 785         'autonumber_start': opts.autonumber_start,
 786         'restrictfilenames': opts.restrictfilenames,
 787         'windowsfilenames': opts.windowsfilenames,
 788         'ignoreerrors': opts.ignoreerrors,
 789         'force_generic_extractor': opts.force_generic_extractor,
 790         'allowed_extractors': opts.allowed_extractors or ['default'],
 791         'ratelimit': opts.ratelimit,
 792         'throttledratelimit': opts.throttledratelimit,
 793         'overwrites': opts.overwrites,
 794         'retries': opts.retries,
 795         'file_access_retries': opts.file_access_retries,
 796         'fragment_retries': opts.fragment_retries,
 797         'extractor_retries': opts.extractor_retries,
 798         'retry_sleep_functions': opts.retry_sleep,
 799         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 800         'keep_fragments': opts.keep_fragments,
 801         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 802         'buffersize': opts.buffersize,
 803         'noresizebuffer': opts.noresizebuffer,
 804         'http_chunk_size': opts.http_chunk_size,
 805         'continuedl': opts.continue_dl,
 806         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 807         'progress_with_newline': opts.progress_with_newline,
 808         'progress_template': opts.progress_template,
 809         'playliststart': opts.playliststart,
 810         'playlistend': opts.playlistend,
 811         'playlistreverse': opts.playlist_reverse,
 812         'playlistrandom': opts.playlist_random,
 813         'lazy_playlist': opts.lazy_playlist,
 814         'noplaylist': opts.noplaylist,
 815         'logtostderr': opts.outtmpl.get('default') == '-',
 816         'consoletitle': opts.consoletitle,
 817         'nopart': opts.nopart,
 818         'updatetime': opts.updatetime,
 819         'writedescription': opts.writedescription,
 820         'writeannotations': opts.writeannotations,
 821         'writeinfojson': opts.writeinfojson,
 822         'allow_playlist_files': opts.allow_playlist_files,
 823         'clean_infojson': opts.clean_infojson,
 824         'getcomments': opts.getcomments,
 825         'writethumbnail': opts.writethumbnail is True,
 826         'write_all_thumbnails': opts.writethumbnail == 'all',
 827         'writelink': opts.writelink,
 828         'writeurllink': opts.writeurllink,
 829         'writewebloclink': opts.writewebloclink,
 830         'writedesktoplink': opts.writedesktoplink,
 831         'writesubtitles': opts.writesubtitles,
 832         'writeautomaticsub': opts.writeautomaticsub,
 833         'allsubtitles': opts.allsubtitles,
 834         'listsubtitles': opts.listsubtitles,
 835         'subtitlesformat': opts.subtitlesformat,
 836         'subtitleslangs': opts.subtitleslangs,
 837         'matchtitle': decodeOption(opts.matchtitle),
 838         'rejecttitle': decodeOption(opts.rejecttitle),
 839         'max_downloads': opts.max_downloads,
 840         'prefer_free_formats': opts.prefer_free_formats,
 841         'trim_file_name': opts.trim_file_name,
 842         'verbose': opts.verbose,
 843         'dump_intermediate_pages': opts.dump_intermediate_pages,
 844         'write_pages': opts.write_pages,
 845         'load_pages': opts.load_pages,
 846         'test': opts.test,
 847         'keepvideo': opts.keepvideo,
 848         'min_filesize': opts.min_filesize,
 849         'max_filesize': opts.max_filesize,
 850         'min_views': opts.min_views,
 851         'max_views': opts.max_views,
 852         'daterange': opts.date,
 853         'cachedir': opts.cachedir,
 854         'youtube_print_sig_code': opts.youtube_print_sig_code,
 855         'age_limit': opts.age_limit,
 856         'download_archive': opts.download_archive,
 857         'break_on_existing': opts.break_on_existing,
 858         'break_on_reject': opts.break_on_reject,
 859         'break_per_url': opts.break_per_url,
 860         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 861         'cookiefile': opts.cookiefile,
 862         'cookiesfrombrowser': opts.cookiesfrombrowser,
 863         'legacyserverconnect': opts.legacy_server_connect,
 864         'nocheckcertificate': opts.no_check_certificate,
 865         'prefer_insecure': opts.prefer_insecure,
 866         'enable_file_urls': opts.enable_file_urls,
 867         'http_headers': opts.headers,
 868         'proxy': opts.proxy,
 869         'socket_timeout': opts.socket_timeout,
 870         'bidi_workaround': opts.bidi_workaround,
 871         'debug_printtraffic': opts.debug_printtraffic,
 872         'prefer_ffmpeg': opts.prefer_ffmpeg,
 873         'include_ads': opts.include_ads,
 874         'default_search': opts.default_search,
 875         'dynamic_mpd': opts.dynamic_mpd,
 876         'extractor_args': opts.extractor_args,
 877         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 878         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 879         'encoding': opts.encoding,
 880         'extract_flat': opts.extract_flat,
 881         'live_from_start': opts.live_from_start,
 882         'wait_for_video': opts.wait_for_video,
 883         'mark_watched': opts.mark_watched,
 884         'merge_output_format': opts.merge_output_format,
 885         'final_ext': final_ext,
 886         'postprocessors': postprocessors,
 887         'fixup': opts.fixup,
 888         'source_address': opts.source_address,
 889         'call_home': opts.call_home,
 890         'sleep_interval_requests': opts.sleep_interval_requests,
 891         'sleep_interval': opts.sleep_interval,
 892         'max_sleep_interval': opts.max_sleep_interval,
 893         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 894         'external_downloader': opts.external_downloader,
 895         'download_ranges': opts.download_ranges,
 896         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 897         'list_thumbnails': opts.list_thumbnails,
 898         'playlist_items': opts.playlist_items,
 899         'xattr_set_filesize': opts.xattr_set_filesize,
 900         'match_filter': opts.match_filter,
 901         'color': opts.color,
 902         'ffmpeg_location': opts.ffmpeg_location,
 903         'hls_prefer_native': opts.hls_prefer_native,
 904         'hls_use_mpegts': opts.hls_use_mpegts,
 905         'hls_split_discontinuity': opts.hls_split_discontinuity,
 906         'external_downloader_args': opts.external_downloader_args,
 907         'postprocessor_args': opts.postprocessor_args,
 908         'cn_verification_proxy': opts.cn_verification_proxy,
 909         'geo_verification_proxy': opts.geo_verification_proxy,
 910         'geo_bypass': opts.geo_bypass,
 911         'geo_bypass_country': opts.geo_bypass_country,
 912         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 913         '_warnings': warnings,
 914         '_deprecation_warnings': deprecation_warnings,
 915         'compat_opts': opts.compat_opts,
 916     })
 917
 918
 919 def _real_main(argv=None):
 920     setproctitle('yt-dlp')
 921
 922     parser, opts, all_urls, ydl_opts = parse_options(argv)
 923
 924     # Dump user agent
 925     if opts.dump_user_agent:
 926         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 927         write_string(f'{ua}\n', out=sys.stdout)
 928         return
 929
 930     if print_extractor_information(opts, all_urls):
 931         return
 932
 933     # We may need ffmpeg_location without having access to the YoutubeDL instance
 934     # See https://github.com/yt-dlp/yt-dlp/issues/2191
 935     if opts.ffmpeg_location:
 936         FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
 937
 938     with YoutubeDL(ydl_opts) as ydl:
 939         pre_process = opts.update_self or opts.rm_cachedir
 940         actual_use = all_urls or opts.load_info_filename
 941
 942         if opts.rm_cachedir:
 943             ydl.cache.remove()
 944
 945         try:
 946             updater = Updater(ydl, opts.update_self)
 947             if opts.update_self and updater.update() and actual_use:
 948                 if updater.cmd:
 949                     return updater.restart()
 950                 # This code is reachable only for zip variant in py < 3.10
 951                 # It makes sense to exit here, but the old behavior is to continue
 952                 ydl.report_warning('Restart yt-dlp to use the updated version')
 953                 # return 100, 'ERROR: The program must exit for the update to complete'
 954         except Exception:
 955             traceback.print_exc()
 956             ydl._download_retcode = 100
 957
 958         if not actual_use:
 959             if pre_process:
 960                 return ydl._download_retcode
 961
 962             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 963             parser.error(
 964                 'You must provide at least one URL.\n'
 965                 'Type yt-dlp --help to see a list of all options.')
 966
 967         parser.destroy()
 968         try:
 969             if opts.load_info_filename is not None:
 970                 if all_urls:
 971                     ydl.report_warning('URLs are ignored due to --load-info-json')
 972                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 973             else:
 974                 return ydl.download(all_urls)
 975         except DownloadCancelled:
 976             ydl.to_screen('Aborting remaining downloads')
 977             return 101
 978
 979
 980 def main(argv=None):
 981     global _IN_CLI
 982     _IN_CLI = True
 983     try:
 984         _exit(*variadic(_real_main(argv)))
 985     except DownloadError:
 986         _exit(1)
 987     except SameFileError as e:
 988         _exit(f'ERROR: {e}')
 989     except KeyboardInterrupt:
 990         _exit('\nERROR: Interrupted by user')
 991     except BrokenPipeError as e:
 992         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 993         devnull = os.open(os.devnull, os.O_WRONLY)
 994         os.dup2(devnull, sys.stdout.fileno())
 995         _exit(f'\nERROR: {e}')
 996     except optparse.OptParseError as e:
 997         _exit(2, f'\n{e}')
 998
 999
1000 from .extractor import gen_extractors, list_extractors
1001
1002 __all__ = [
1003     'main',
1004     'YoutubeDL',
1005     'parse_options',
1006     'gen_extractors',
1007     'list_extractors',
1008 ]