yt_dlp/__init__.py

   1 import sys
   2
   3 if sys.version_info < (3, 8):
   4     raise ImportError(
   5         f'You are using an unsupported version of Python. Only Python versions 3.8 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 traceback
  16
  17 from .compat import compat_shlex_quote
  18 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  19 from .downloader.external import get_external_downloader
  20 from .extractor import list_extractor_classes
  21 from .extractor.adobepass import MSO_INFO
  22 from .options import parseOpts
  23 from .postprocessor import (
  24     FFmpegExtractAudioPP,
  25     FFmpegMergerPP,
  26     FFmpegPostProcessor,
  27     FFmpegSubtitlesConvertorPP,
  28     FFmpegThumbnailsConvertorPP,
  29     FFmpegVideoConvertorPP,
  30     FFmpegVideoRemuxerPP,
  31     MetadataFromFieldPP,
  32     MetadataParserPP,
  33 )
  34 from .update import Updater
  35 from .utils import (
  36     NO_DEFAULT,
  37     POSTPROCESS_WHEN,
  38     DateRange,
  39     DownloadCancelled,
  40     DownloadError,
  41     FormatSorter,
  42     GeoUtils,
  43     PlaylistEntries,
  44     SameFileError,
  45     decodeOption,
  46     download_range_func,
  47     expand_path,
  48     float_or_none,
  49     format_field,
  50     int_or_none,
  51     match_filter_func,
  52     parse_bytes,
  53     parse_duration,
  54     preferredencoding,
  55     read_batch_urls,
  56     read_stdin,
  57     render_table,
  58     setproctitle,
  59     traverse_obj,
  60     variadic,
  61     write_string,
  62 )
  63 from .utils.networking import std_headers
  64 from .YoutubeDL import YoutubeDL
  65
  66 _IN_CLI = False
  67
  68
  69 def _exit(status=0, *args):
  70     for msg in args:
  71         sys.stderr.write(msg)
  72     raise SystemExit(status)
  73
  74
  75 def get_urls(urls, batchfile, verbose):
  76     # Batch file verification
  77     batch_urls = []
  78     if batchfile is not None:
  79         try:
  80             batch_urls = read_batch_urls(
  81                 read_stdin('URLs') if batchfile == '-'
  82                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  83             if verbose:
  84                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  85         except OSError:
  86             _exit(f'ERROR: batch file {batchfile} could not be read')
  87     _enc = preferredencoding()
  88     return [
  89         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  90         for url in batch_urls + urls]
  91
  92
  93 def print_extractor_information(opts, urls):
  94     out = ''
  95     if opts.list_extractors:
  96         # Importing GenericIE is currently slow since it imports YoutubeIE
  97         from .extractor.generic import GenericIE
  98
  99         urls = dict.fromkeys(urls, False)
 100         for ie in list_extractor_classes(opts.age_limit):
 101             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
 102             if ie == GenericIE:
 103                 matched_urls = [url for url, matched in urls.items() if not matched]
 104             else:
 105                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
 106                 urls.update(dict.fromkeys(matched_urls, True))
 107             out += ''.join(f'  {url}\n' for url in matched_urls)
 108     elif opts.list_extractor_descriptions:
 109         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 110         out = '\n'.join(
 111             ie.description(markdown=False, search_examples=_SEARCHES)
 112             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
 113     elif opts.ap_list_mso:
 114         out = 'Supported TV Providers:\n%s\n' % render_table(
 115             ['mso', 'mso name'],
 116             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 117     else:
 118         return False
 119     write_string(out, out=sys.stdout)
 120     return True
 121
 122
 123 def set_compat_opts(opts):
 124     def _unused_compat_opt(name):
 125         if name not in opts.compat_opts:
 126             return False
 127         opts.compat_opts.discard(name)
 128         opts.compat_opts.update(['*%s' % name])
 129         return True
 130
 131     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 132         attr = getattr(opts, opt_name)
 133         if compat_name in opts.compat_opts:
 134             if attr is None:
 135                 setattr(opts, opt_name, not default)
 136                 return True
 137             else:
 138                 if remove_compat:
 139                     _unused_compat_opt(compat_name)
 140                 return False
 141         elif attr is None:
 142             setattr(opts, opt_name, default)
 143         return None
 144
 145     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 146     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 147     set_default_compat('no-clean-infojson', 'clean_infojson')
 148     if 'no-attach-info-json' in opts.compat_opts:
 149         if opts.embed_infojson:
 150             _unused_compat_opt('no-attach-info-json')
 151         else:
 152             opts.embed_infojson = False
 153     if 'format-sort' in opts.compat_opts:
 154         opts.format_sort.extend(FormatSorter.ytdl_default)
 155     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 156     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 157     if _video_multistreams_set is False and _audio_multistreams_set is False:
 158         _unused_compat_opt('multistreams')
 159     if 'filename' in opts.compat_opts:
 160         if opts.outtmpl.get('default') is None:
 161             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 162         else:
 163             _unused_compat_opt('filename')
 164
 165
 166 def validate_options(opts):
 167     def validate(cndn, name, value=None, msg=None):
 168         if cndn:
 169             return True
 170         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 171
 172     def validate_in(name, value, items, msg=None):
 173         return validate(value is None or value in items, name, value, msg)
 174
 175     def validate_regex(name, value, regex):
 176         return validate(value is None or re.match(regex, value), name, value)
 177
 178     def validate_positive(name, value, strict=False):
 179         return validate(value is None or value > 0 or (not strict and value == 0),
 180                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 181
 182     def validate_minmax(min_val, max_val, min_name, max_name=None):
 183         if max_val is None or min_val is None or max_val >= min_val:
 184             return
 185         if not max_name:
 186             min_name, max_name = f'min {min_name}', f'max {min_name}'
 187         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 188
 189     # Usernames and passwords
 190     validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
 191              msg='{name}, netrc command and username/password are mutually exclusive options')
 192     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 193     validate(opts.ap_password is None or opts.ap_username is not None,
 194              'TV Provider account username', msg='{name} missing')
 195     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 196                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 197
 198     # Numbers
 199     validate_positive('autonumber start', opts.autonumber_start)
 200     validate_positive('autonumber size', opts.autonumber_size, True)
 201     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 202     validate_positive('playlist start', opts.playliststart, True)
 203     if opts.playlistend != -1:
 204         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 205
 206     # Time ranges
 207     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 208     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 209     validate_positive('sleep interval', opts.sleep_interval)
 210     validate_positive('max sleep interval', opts.max_sleep_interval)
 211     if opts.sleep_interval is None:
 212         validate(
 213             opts.max_sleep_interval is None, 'min sleep interval',
 214             msg='{name} must be specified; use --min-sleep-interval')
 215     elif opts.max_sleep_interval is None:
 216         opts.max_sleep_interval = opts.sleep_interval
 217     else:
 218         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 219
 220     if opts.wait_for_video is not None:
 221         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 222         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 223                  'time range to wait for video', opts.wait_for_video)
 224         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 225         opts.wait_for_video = (min_wait, max_wait)
 226
 227     # Format sort
 228     for f in opts.format_sort:
 229         validate_regex('format sorting', f, FormatSorter.regex)
 230
 231     # Postprocessor formats
 232     validate_regex('merge output format', opts.merge_output_format,
 233                    r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
 234     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 235     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 236     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 237     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 238     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 239     if opts.audioquality:
 240         opts.audioquality = opts.audioquality.strip('k').strip('K')
 241         # int_or_none prevents inf, nan
 242         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 243
 244     # Retries
 245     def parse_retries(name, value):
 246         if value is None:
 247             return None
 248         elif value in ('inf', 'infinite'):
 249             return float('inf')
 250         try:
 251             return int(value)
 252         except (TypeError, ValueError):
 253             validate(False, f'{name} retry count', value)
 254
 255     opts.retries = parse_retries('download', opts.retries)
 256     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 257     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 258     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 259
 260     # Retry sleep function
 261     def parse_sleep_func(expr):
 262         NUMBER_RE = r'\d+(?:\.\d+)?'
 263         op, start, limit, step, *_ = tuple(re.fullmatch(
 264             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 265             expr.strip()).groups()) + (None, None)
 266
 267         if op == 'exp':
 268             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 269         else:
 270             default_step = start if op or limit else 0
 271             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 272
 273     for key, expr in opts.retry_sleep.items():
 274         if not expr:
 275             del opts.retry_sleep[key]
 276             continue
 277         try:
 278             opts.retry_sleep[key] = parse_sleep_func(expr)
 279         except AttributeError:
 280             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 281
 282     # Bytes
 283     def validate_bytes(name, value):
 284         if value is None:
 285             return None
 286         numeric_limit = parse_bytes(value)
 287         validate(numeric_limit is not None, 'rate limit', value)
 288         return numeric_limit
 289
 290     opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
 291     opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
 292     opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
 293     opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
 294     opts.buffersize = validate_bytes('buffer size', opts.buffersize)
 295     opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
 296
 297     # Output templates
 298     def validate_outtmpl(tmpl, msg):
 299         err = YoutubeDL.validate_outtmpl(tmpl)
 300         if err:
 301             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 302
 303     for k, tmpl in opts.outtmpl.items():
 304         validate_outtmpl(tmpl, f'{k} output template')
 305     for type_, tmpl_list in opts.forceprint.items():
 306         for tmpl in tmpl_list:
 307             validate_outtmpl(tmpl, f'{type_} print template')
 308     for type_, tmpl_list in opts.print_to_file.items():
 309         for tmpl, file in tmpl_list:
 310             validate_outtmpl(tmpl, f'{type_} print to file template')
 311             validate_outtmpl(file, f'{type_} print to file filename')
 312     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 313     for k, tmpl in opts.progress_template.items():
 314         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 315         validate_outtmpl(tmpl, f'{k} template')
 316
 317     outtmpl_default = opts.outtmpl.get('default')
 318     if outtmpl_default == '':
 319         opts.skip_download = None
 320         del opts.outtmpl['default']
 321
 322     def parse_chapters(name, value, advanced=False):
 323         parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
 324         TIMESTAMP_RE = r'''(?x)(?:
 325             (?P<start_sign>-?)(?P<start>[^-]+)
 326         )?\s*-\s*(?:
 327             (?P<end_sign>-?)(?P<end>[^-]+)
 328         )?'''
 329
 330         chapters, ranges, from_url = [], [], False
 331         for regex in value or []:
 332             if advanced and regex == '*from-url':
 333                 from_url = True
 334                 continue
 335             elif not regex.startswith('*'):
 336                 try:
 337                     chapters.append(re.compile(regex))
 338                 except re.error as err:
 339                     raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 340                 continue
 341
 342             for range_ in map(str.strip, regex[1:].split(',')):
 343                 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
 344                 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
 345                 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
 346
 347                 err = None
 348                 if None in (dur or [None]):
 349                     err = 'Must be of the form "*start-end"'
 350                 elif not advanced and any(signs):
 351                     err = 'Negative timestamps are not allowed'
 352                 else:
 353                     dur[0] *= -1 if signs[0] else 1
 354                     dur[1] *= -1 if signs[1] else 1
 355                     if dur[1] == float('-inf'):
 356                         err = '"-inf" is not a valid end'
 357                 if err:
 358                     raise ValueError(f'invalid {name} time range "{regex}". {err}')
 359                 ranges.append(dur)
 360
 361         return chapters, ranges, from_url
 362
 363     opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
 364     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
 365
 366     # Cookies from browser
 367     if opts.cookiesfrombrowser:
 368         container = None
 369         mobj = re.fullmatch(r'''(?x)
 370             (?P<name>[^+:]+)
 371             (?:\s*\+\s*(?P<keyring>[^:]+))?
 372             (?:\s*:\s*(?!:)(?P<profile>.+?))?
 373             (?:\s*::\s*(?P<container>.+))?
 374         ''', opts.cookiesfrombrowser)
 375         if mobj is None:
 376             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 377         browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
 378         browser_name = browser_name.lower()
 379         if browser_name not in SUPPORTED_BROWSERS:
 380             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 381                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 382         if keyring is not None:
 383             keyring = keyring.upper()
 384             if keyring not in SUPPORTED_KEYRINGS:
 385                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 386                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 387         opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
 388
 389     # MetadataParser
 390     def metadataparser_actions(f):
 391         if isinstance(f, str):
 392             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 393             try:
 394                 actions = [MetadataFromFieldPP.to_action(f)]
 395             except Exception as err:
 396                 raise ValueError(f'{cmd} is invalid; {err}')
 397         else:
 398             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 399             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 400
 401         for action in actions:
 402             try:
 403                 MetadataParserPP.validate_action(*action)
 404             except Exception as err:
 405                 raise ValueError(f'{cmd} is invalid; {err}')
 406             yield action
 407
 408     if opts.metafromtitle is not None:
 409         opts.parse_metadata.setdefault('pre_process', []).append('title:%s' % opts.metafromtitle)
 410     opts.parse_metadata = {
 411         k: list(itertools.chain(*map(metadataparser_actions, v)))
 412         for k, v in opts.parse_metadata.items()
 413     }
 414
 415     # Other options
 416     if opts.playlist_items is not None:
 417         try:
 418             tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
 419         except Exception as err:
 420             raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
 421
 422     opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
 423     if opts.geo_bypass.lower() not in ('default', 'never'):
 424         try:
 425             GeoUtils.random_ipv4(opts.geo_bypass)
 426         except Exception:
 427             raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
 428         if len(opts.geo_bypass) == 2:
 429             opts.geo_bypass_country = opts.geo_bypass
 430         else:
 431             opts.geo_bypass_ip_block = opts.geo_bypass
 432     opts.geo_bypass = opts.geo_bypass.lower() != 'never'
 433
 434     opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
 435
 436     if opts.download_archive is not None:
 437         opts.download_archive = expand_path(opts.download_archive)
 438
 439     if opts.ffmpeg_location is not None:
 440         opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
 441
 442     if opts.user_agent is not None:
 443         opts.headers.setdefault('User-Agent', opts.user_agent)
 444     if opts.referer is not None:
 445         opts.headers.setdefault('Referer', opts.referer)
 446
 447     if opts.no_sponsorblock:
 448         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 449
 450     default_downloader = None
 451     for proto, path in opts.external_downloader.items():
 452         if path == 'native':
 453             continue
 454         ed = get_external_downloader(path)
 455         if ed is None:
 456             raise ValueError(
 457                 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
 458         elif ed and proto == 'default':
 459             default_downloader = ed.get_basename()
 460
 461     for policy in opts.color.values():
 462         if policy not in ('always', 'auto', 'no_color', 'never'):
 463             raise ValueError(f'"{policy}" is not a valid color policy')
 464
 465     warnings, deprecation_warnings = [], []
 466
 467     # Common mistake: -f best
 468     if opts.format == 'best':
 469         warnings.append('.\n         '.join((
 470             '"-f best" selects the best pre-merged format which is often not the best option',
 471             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 472             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 473
 474     # --(postprocessor/downloader)-args without name
 475     def report_args_compat(name, value, key1, key2=None, where=None):
 476         if key1 in value and key2 not in value:
 477             warnings.append(f'{name.title()} arguments given without specifying name. '
 478                             f'The arguments will be given to {where or f"all {name}s"}')
 479             return True
 480         return False
 481
 482     if report_args_compat('external downloader', opts.external_downloader_args,
 483                           'default', where=default_downloader) and default_downloader:
 484         # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
 485         opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
 486
 487     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 488         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 489         opts.postprocessor_args.setdefault('sponskrub', [])
 490
 491     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 492                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 493         if val2 is NO_DEFAULT:
 494             val2 = getattr(opts, opt2)
 495         if not val2:
 496             return
 497
 498         if val1 is NO_DEFAULT:
 499             val1 = getattr(opts, opt1)
 500         if val1:
 501             warnings.append(f'{arg1} is ignored since {arg2} was given')
 502         setattr(opts, opt1, default)
 503
 504     # Conflicting options
 505     report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
 506     report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
 507     report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
 508     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 509     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 510     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 511                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 512     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 513     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 514     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 515     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 516     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 517     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 518                     val1=opts.sponskrub and opts.sponskrub_cut)
 519
 520     # Conflicts with --allow-unplayable-formats
 521     report_conflict('--embed-metadata', 'addmetadata')
 522     report_conflict('--embed-chapters', 'addchapters')
 523     report_conflict('--embed-info-json', 'embed_infojson')
 524     report_conflict('--embed-subs', 'embedsubtitles')
 525     report_conflict('--embed-thumbnail', 'embedthumbnail')
 526     report_conflict('--extract-audio', 'extractaudio')
 527     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 528     report_conflict('--recode-video', 'recodevideo')
 529     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 530     report_conflict('--remux-video', 'remuxvideo')
 531     report_conflict('--sponskrub', 'sponskrub')
 532     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 533     report_conflict('--xattrs', 'xattrs')
 534
 535     # Fully deprecated options
 536     def report_deprecation(val, old, new=None):
 537         if not val:
 538             return
 539         deprecation_warnings.append(
 540             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 541             else f'{old} is deprecated and may not work as expected')
 542
 543     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 544     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 545     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 546     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 547     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 548
 549     # Dependent options
 550     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 551
 552     if opts.exec_before_dl_cmd:
 553         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 554
 555     if opts.useid:  # --id is not deprecated in youtube-dl
 556         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 557
 558     if opts.overwrites:  # --force-overwrites implies --no-continue
 559         opts.continue_dl = False
 560
 561     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 562         # Add chapters when adding metadata or marking sponsors
 563         opts.addchapters = True
 564
 565     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 566         # Do not unnecessarily download audio
 567         opts.format = 'bestaudio/best'
 568
 569     if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
 570         # If JSON is not printed anywhere, but comments are requested, save it to file
 571         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 572             opts.writeinfojson = True
 573
 574     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 575         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 576         opts.writesubtitles = True
 577
 578     if opts.addmetadata and opts.embed_infojson is None:
 579         # If embedding metadata and infojson is present, embed it
 580         opts.embed_infojson = 'if_exists'
 581
 582     # Ask for passwords
 583     if opts.username is not None and opts.password is None:
 584         opts.password = getpass.getpass('Type account password and press [Return]: ')
 585     if opts.ap_username is not None and opts.ap_password is None:
 586         opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 587
 588     return warnings, deprecation_warnings
 589
 590
 591 def get_postprocessors(opts):
 592     yield from opts.add_postprocessors
 593
 594     for when, actions in opts.parse_metadata.items():
 595         yield {
 596             'key': 'MetadataParser',
 597             'actions': actions,
 598             'when': when
 599         }
 600     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 601     if sponsorblock_query:
 602         yield {
 603             'key': 'SponsorBlock',
 604             'categories': sponsorblock_query,
 605             'api': opts.sponsorblock_api,
 606             'when': 'after_filter'
 607         }
 608     if opts.convertsubtitles:
 609         yield {
 610             'key': 'FFmpegSubtitlesConvertor',
 611             'format': opts.convertsubtitles,
 612             'when': 'before_dl'
 613         }
 614     if opts.convertthumbnails:
 615         yield {
 616             'key': 'FFmpegThumbnailsConvertor',
 617             'format': opts.convertthumbnails,
 618             'when': 'before_dl'
 619         }
 620     if opts.extractaudio:
 621         yield {
 622             'key': 'FFmpegExtractAudio',
 623             'preferredcodec': opts.audioformat,
 624             'preferredquality': opts.audioquality,
 625             'nopostoverwrites': opts.nopostoverwrites,
 626         }
 627     if opts.remuxvideo:
 628         yield {
 629             'key': 'FFmpegVideoRemuxer',
 630             'preferedformat': opts.remuxvideo,
 631         }
 632     if opts.recodevideo:
 633         yield {
 634             'key': 'FFmpegVideoConvertor',
 635             'preferedformat': opts.recodevideo,
 636         }
 637     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 638     if opts.embedsubtitles:
 639         keep_subs = 'no-keep-subs' not in opts.compat_opts
 640         yield {
 641             'key': 'FFmpegEmbedSubtitle',
 642             # already_have_subtitle = True prevents the file from being deleted after embedding
 643             'already_have_subtitle': opts.writesubtitles and keep_subs
 644         }
 645         if not opts.writeautomaticsub and keep_subs:
 646             opts.writesubtitles = True
 647
 648     # ModifyChapters must run before FFmpegMetadataPP
 649     if opts.remove_chapters or sponsorblock_query:
 650         yield {
 651             'key': 'ModifyChapters',
 652             'remove_chapters_patterns': opts.remove_chapters,
 653             'remove_sponsor_segments': opts.sponsorblock_remove,
 654             'remove_ranges': opts.remove_ranges,
 655             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 656             'force_keyframes': opts.force_keyframes_at_cuts
 657         }
 658     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 659     # FFmpegExtractAudioPP as containers before conversion may not support
 660     # metadata (3gp, webm, etc.)
 661     # By default ffmpeg preserves metadata applicable for both
 662     # source and target containers. From this point the container won't change,
 663     # so metadata can be added here.
 664     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 665         yield {
 666             'key': 'FFmpegMetadata',
 667             'add_chapters': opts.addchapters,
 668             'add_metadata': opts.addmetadata,
 669             'add_infojson': opts.embed_infojson,
 670         }
 671     # Deprecated
 672     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 673     # but must be below EmbedSubtitle and FFmpegMetadata
 674     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 675     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 676     if opts.sponskrub is not False:
 677         yield {
 678             'key': 'SponSkrub',
 679             'path': opts.sponskrub_path,
 680             'args': opts.sponskrub_args,
 681             'cut': opts.sponskrub_cut,
 682             'force': opts.sponskrub_force,
 683             'ignoreerror': opts.sponskrub is None,
 684             '_from_cli': True,
 685         }
 686     if opts.embedthumbnail:
 687         yield {
 688             'key': 'EmbedThumbnail',
 689             # already_have_thumbnail = True prevents the file from being deleted after embedding
 690             'already_have_thumbnail': opts.writethumbnail
 691         }
 692         if not opts.writethumbnail:
 693             opts.writethumbnail = True
 694             opts.outtmpl['pl_thumbnail'] = ''
 695     if opts.split_chapters:
 696         yield {
 697             'key': 'FFmpegSplitChapters',
 698             'force_keyframes': opts.force_keyframes_at_cuts,
 699         }
 700     # XAttrMetadataPP should be run after post-processors that may change file contents
 701     if opts.xattrs:
 702         yield {'key': 'XAttrMetadata'}
 703     if opts.concat_playlist != 'never':
 704         yield {
 705             'key': 'FFmpegConcat',
 706             'only_multi_video': opts.concat_playlist != 'always',
 707             'when': 'playlist',
 708         }
 709     # Exec must be the last PP of each category
 710     for when, exec_cmd in opts.exec_cmd.items():
 711         yield {
 712             'key': 'Exec',
 713             'exec_cmd': exec_cmd,
 714             'when': when,
 715         }
 716
 717
 718 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
 719
 720
 721 def parse_options(argv=None):
 722     """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
 723     parser, opts, urls = parseOpts(argv)
 724     urls = get_urls(urls, opts.batchfile, opts.verbose)
 725
 726     set_compat_opts(opts)
 727     try:
 728         warnings, deprecation_warnings = validate_options(opts)
 729     except ValueError as err:
 730         parser.error(f'{err}\n')
 731
 732     postprocessors = list(get_postprocessors(opts))
 733
 734     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
 735     any_getting = any(getattr(opts, k) for k in (
 736         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 737         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 738     ))
 739     if opts.quiet is None:
 740         opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
 741
 742     playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
 743     write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
 744                                and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
 745     if not any((
 746         opts.extract_flat,
 747         opts.dump_single_json,
 748         opts.forceprint.get('playlist'),
 749         opts.print_to_file.get('playlist'),
 750         write_playlist_infojson,
 751     )):
 752         if not playlist_pps:
 753             opts.extract_flat = 'discard'
 754         elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
 755             opts.extract_flat = 'discard_in_playlist'
 756
 757     final_ext = (
 758         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 759         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 760         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 761         else None)
 762
 763     return ParsedOptions(parser, opts, urls, {
 764         'usenetrc': opts.usenetrc,
 765         'netrc_location': opts.netrc_location,
 766         'netrc_cmd': opts.netrc_cmd,
 767         'username': opts.username,
 768         'password': opts.password,
 769         'twofactor': opts.twofactor,
 770         'videopassword': opts.videopassword,
 771         'ap_mso': opts.ap_mso,
 772         'ap_username': opts.ap_username,
 773         'ap_password': opts.ap_password,
 774         'client_certificate': opts.client_certificate,
 775         'client_certificate_key': opts.client_certificate_key,
 776         'client_certificate_password': opts.client_certificate_password,
 777         'quiet': opts.quiet,
 778         'no_warnings': opts.no_warnings,
 779         'forceurl': opts.geturl,
 780         'forcetitle': opts.gettitle,
 781         'forceid': opts.getid,
 782         'forcethumbnail': opts.getthumbnail,
 783         'forcedescription': opts.getdescription,
 784         'forceduration': opts.getduration,
 785         'forcefilename': opts.getfilename,
 786         'forceformat': opts.getformat,
 787         'forceprint': opts.forceprint,
 788         'print_to_file': opts.print_to_file,
 789         'forcejson': opts.dumpjson or opts.print_json,
 790         'dump_single_json': opts.dump_single_json,
 791         'force_write_download_archive': opts.force_write_download_archive,
 792         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 793         'skip_download': opts.skip_download,
 794         'format': opts.format,
 795         'allow_unplayable_formats': opts.allow_unplayable_formats,
 796         'ignore_no_formats_error': opts.ignore_no_formats_error,
 797         'format_sort': opts.format_sort,
 798         'format_sort_force': opts.format_sort_force,
 799         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 800         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 801         'check_formats': opts.check_formats,
 802         'listformats': opts.listformats,
 803         'listformats_table': opts.listformats_table,
 804         'outtmpl': opts.outtmpl,
 805         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 806         'paths': opts.paths,
 807         'autonumber_size': opts.autonumber_size,
 808         'autonumber_start': opts.autonumber_start,
 809         'restrictfilenames': opts.restrictfilenames,
 810         'windowsfilenames': opts.windowsfilenames,
 811         'ignoreerrors': opts.ignoreerrors,
 812         'force_generic_extractor': opts.force_generic_extractor,
 813         'allowed_extractors': opts.allowed_extractors or ['default'],
 814         'ratelimit': opts.ratelimit,
 815         'throttledratelimit': opts.throttledratelimit,
 816         'overwrites': opts.overwrites,
 817         'retries': opts.retries,
 818         'file_access_retries': opts.file_access_retries,
 819         'fragment_retries': opts.fragment_retries,
 820         'extractor_retries': opts.extractor_retries,
 821         'retry_sleep_functions': opts.retry_sleep,
 822         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 823         'keep_fragments': opts.keep_fragments,
 824         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 825         'buffersize': opts.buffersize,
 826         'noresizebuffer': opts.noresizebuffer,
 827         'http_chunk_size': opts.http_chunk_size,
 828         'continuedl': opts.continue_dl,
 829         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 830         'progress_with_newline': opts.progress_with_newline,
 831         'progress_template': opts.progress_template,
 832         'playliststart': opts.playliststart,
 833         'playlistend': opts.playlistend,
 834         'playlistreverse': opts.playlist_reverse,
 835         'playlistrandom': opts.playlist_random,
 836         'lazy_playlist': opts.lazy_playlist,
 837         'noplaylist': opts.noplaylist,
 838         'logtostderr': opts.outtmpl.get('default') == '-',
 839         'consoletitle': opts.consoletitle,
 840         'nopart': opts.nopart,
 841         'updatetime': opts.updatetime,
 842         'writedescription': opts.writedescription,
 843         'writeannotations': opts.writeannotations,
 844         'writeinfojson': opts.writeinfojson,
 845         'allow_playlist_files': opts.allow_playlist_files,
 846         'clean_infojson': opts.clean_infojson,
 847         'getcomments': opts.getcomments,
 848         'writethumbnail': opts.writethumbnail is True,
 849         'write_all_thumbnails': opts.writethumbnail == 'all',
 850         'writelink': opts.writelink,
 851         'writeurllink': opts.writeurllink,
 852         'writewebloclink': opts.writewebloclink,
 853         'writedesktoplink': opts.writedesktoplink,
 854         'writesubtitles': opts.writesubtitles,
 855         'writeautomaticsub': opts.writeautomaticsub,
 856         'allsubtitles': opts.allsubtitles,
 857         'listsubtitles': opts.listsubtitles,
 858         'subtitlesformat': opts.subtitlesformat,
 859         'subtitleslangs': opts.subtitleslangs,
 860         'matchtitle': decodeOption(opts.matchtitle),
 861         'rejecttitle': decodeOption(opts.rejecttitle),
 862         'max_downloads': opts.max_downloads,
 863         'prefer_free_formats': opts.prefer_free_formats,
 864         'trim_file_name': opts.trim_file_name,
 865         'verbose': opts.verbose,
 866         'dump_intermediate_pages': opts.dump_intermediate_pages,
 867         'write_pages': opts.write_pages,
 868         'load_pages': opts.load_pages,
 869         'test': opts.test,
 870         'keepvideo': opts.keepvideo,
 871         'min_filesize': opts.min_filesize,
 872         'max_filesize': opts.max_filesize,
 873         'min_views': opts.min_views,
 874         'max_views': opts.max_views,
 875         'daterange': opts.date,
 876         'cachedir': opts.cachedir,
 877         'youtube_print_sig_code': opts.youtube_print_sig_code,
 878         'age_limit': opts.age_limit,
 879         'download_archive': opts.download_archive,
 880         'break_on_existing': opts.break_on_existing,
 881         'break_on_reject': opts.break_on_reject,
 882         'break_per_url': opts.break_per_url,
 883         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 884         'cookiefile': opts.cookiefile,
 885         'cookiesfrombrowser': opts.cookiesfrombrowser,
 886         'legacyserverconnect': opts.legacy_server_connect,
 887         'nocheckcertificate': opts.no_check_certificate,
 888         'prefer_insecure': opts.prefer_insecure,
 889         'enable_file_urls': opts.enable_file_urls,
 890         'http_headers': opts.headers,
 891         'proxy': opts.proxy,
 892         'socket_timeout': opts.socket_timeout,
 893         'bidi_workaround': opts.bidi_workaround,
 894         'debug_printtraffic': opts.debug_printtraffic,
 895         'prefer_ffmpeg': opts.prefer_ffmpeg,
 896         'include_ads': opts.include_ads,
 897         'default_search': opts.default_search,
 898         'dynamic_mpd': opts.dynamic_mpd,
 899         'extractor_args': opts.extractor_args,
 900         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 901         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 902         'encoding': opts.encoding,
 903         'extract_flat': opts.extract_flat,
 904         'live_from_start': opts.live_from_start,
 905         'wait_for_video': opts.wait_for_video,
 906         'mark_watched': opts.mark_watched,
 907         'merge_output_format': opts.merge_output_format,
 908         'final_ext': final_ext,
 909         'postprocessors': postprocessors,
 910         'fixup': opts.fixup,
 911         'source_address': opts.source_address,
 912         'call_home': opts.call_home,
 913         'sleep_interval_requests': opts.sleep_interval_requests,
 914         'sleep_interval': opts.sleep_interval,
 915         'max_sleep_interval': opts.max_sleep_interval,
 916         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 917         'external_downloader': opts.external_downloader,
 918         'download_ranges': opts.download_ranges,
 919         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 920         'list_thumbnails': opts.list_thumbnails,
 921         'playlist_items': opts.playlist_items,
 922         'xattr_set_filesize': opts.xattr_set_filesize,
 923         'match_filter': opts.match_filter,
 924         'color': opts.color,
 925         'ffmpeg_location': opts.ffmpeg_location,
 926         'hls_prefer_native': opts.hls_prefer_native,
 927         'hls_use_mpegts': opts.hls_use_mpegts,
 928         'hls_split_discontinuity': opts.hls_split_discontinuity,
 929         'external_downloader_args': opts.external_downloader_args,
 930         'postprocessor_args': opts.postprocessor_args,
 931         'cn_verification_proxy': opts.cn_verification_proxy,
 932         'geo_verification_proxy': opts.geo_verification_proxy,
 933         'geo_bypass': opts.geo_bypass,
 934         'geo_bypass_country': opts.geo_bypass_country,
 935         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 936         '_warnings': warnings,
 937         '_deprecation_warnings': deprecation_warnings,
 938         'compat_opts': opts.compat_opts,
 939     })
 940
 941
 942 def _real_main(argv=None):
 943     setproctitle('yt-dlp')
 944
 945     parser, opts, all_urls, ydl_opts = parse_options(argv)
 946
 947     # Dump user agent
 948     if opts.dump_user_agent:
 949         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 950         write_string(f'{ua}\n', out=sys.stdout)
 951         return
 952
 953     if print_extractor_information(opts, all_urls):
 954         return
 955
 956     # We may need ffmpeg_location without having access to the YoutubeDL instance
 957     # See https://github.com/yt-dlp/yt-dlp/issues/2191
 958     if opts.ffmpeg_location:
 959         FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
 960
 961     with YoutubeDL(ydl_opts) as ydl:
 962         pre_process = opts.update_self or opts.rm_cachedir
 963         actual_use = all_urls or opts.load_info_filename
 964
 965         if opts.rm_cachedir:
 966             ydl.cache.remove()
 967
 968         try:
 969             updater = Updater(ydl, opts.update_self)
 970             if opts.update_self and updater.update() and actual_use:
 971                 if updater.cmd:
 972                     return updater.restart()
 973                 # This code is reachable only for zip variant in py < 3.10
 974                 # It makes sense to exit here, but the old behavior is to continue
 975                 ydl.report_warning('Restart yt-dlp to use the updated version')
 976                 # return 100, 'ERROR: The program must exit for the update to complete'
 977         except Exception:
 978             traceback.print_exc()
 979             ydl._download_retcode = 100
 980
 981         if not actual_use:
 982             if pre_process:
 983                 return ydl._download_retcode
 984
 985             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 986             parser.error(
 987                 'You must provide at least one URL.\n'
 988                 'Type yt-dlp --help to see a list of all options.')
 989
 990         parser.destroy()
 991         try:
 992             if opts.load_info_filename is not None:
 993                 if all_urls:
 994                     ydl.report_warning('URLs are ignored due to --load-info-json')
 995                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 996             else:
 997                 return ydl.download(all_urls)
 998         except DownloadCancelled:
 999             ydl.to_screen('Aborting remaining downloads')
1000             return 101
1001
1002
1003 def main(argv=None):
1004     global _IN_CLI
1005     _IN_CLI = True
1006     try:
1007         _exit(*variadic(_real_main(argv)))
1008     except DownloadError:
1009         _exit(1)
1010     except SameFileError as e:
1011         _exit(f'ERROR: {e}')
1012     except KeyboardInterrupt:
1013         _exit('\nERROR: Interrupted by user')
1014     except BrokenPipeError as e:
1015         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1016         devnull = os.open(os.devnull, os.O_WRONLY)
1017         os.dup2(devnull, sys.stdout.fileno())
1018         _exit(f'\nERROR: {e}')
1019     except optparse.OptParseError as e:
1020         _exit(2, f'\n{e}')
1021
1022
1023 from .extractor import gen_extractors, list_extractors
1024
1025 __all__ = [
1026     'main',
1027     'YoutubeDL',
1028     'parse_options',
1029     'gen_extractors',
1030     'list_extractors',
1031 ]