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     """
  77     @param verbose      -1: quiet, 0: normal, 1: verbose
  78     """
  79     batch_urls = []
  80     if batchfile is not None:
  81         try:
  82             batch_urls = read_batch_urls(
  83                 read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
  84                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  85             if verbose == 1:
  86                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  87         except OSError:
  88             _exit(f'ERROR: batch file {batchfile} could not be read')
  89     _enc = preferredencoding()
  90     return [
  91         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  92         for url in batch_urls + urls]
  93
  94
  95 def print_extractor_information(opts, urls):
  96     out = ''
  97     if opts.list_extractors:
  98         # Importing GenericIE is currently slow since it imports YoutubeIE
  99         from .extractor.generic import GenericIE
 100
 101         urls = dict.fromkeys(urls, False)
 102         for ie in list_extractor_classes(opts.age_limit):
 103             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
 104             if ie == GenericIE:
 105                 matched_urls = [url for url, matched in urls.items() if not matched]
 106             else:
 107                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
 108                 urls.update(dict.fromkeys(matched_urls, True))
 109             out += ''.join(f'  {url}\n' for url in matched_urls)
 110     elif opts.list_extractor_descriptions:
 111         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 112         out = '\n'.join(
 113             ie.description(markdown=False, search_examples=_SEARCHES)
 114             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
 115     elif opts.ap_list_mso:
 116         out = 'Supported TV Providers:\n%s\n' % render_table(
 117             ['mso', 'mso name'],
 118             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 119     else:
 120         return False
 121     write_string(out, out=sys.stdout)
 122     return True
 123
 124
 125 def set_compat_opts(opts):
 126     def _unused_compat_opt(name):
 127         if name not in opts.compat_opts:
 128             return False
 129         opts.compat_opts.discard(name)
 130         opts.compat_opts.update(['*%s' % name])
 131         return True
 132
 133     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 134         attr = getattr(opts, opt_name)
 135         if compat_name in opts.compat_opts:
 136             if attr is None:
 137                 setattr(opts, opt_name, not default)
 138                 return True
 139             else:
 140                 if remove_compat:
 141                     _unused_compat_opt(compat_name)
 142                 return False
 143         elif attr is None:
 144             setattr(opts, opt_name, default)
 145         return None
 146
 147     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 148     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 149     set_default_compat('no-clean-infojson', 'clean_infojson')
 150     if 'no-attach-info-json' in opts.compat_opts:
 151         if opts.embed_infojson:
 152             _unused_compat_opt('no-attach-info-json')
 153         else:
 154             opts.embed_infojson = False
 155     if 'format-sort' in opts.compat_opts:
 156         opts.format_sort.extend(FormatSorter.ytdl_default)
 157     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 158     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 159     if _video_multistreams_set is False and _audio_multistreams_set is False:
 160         _unused_compat_opt('multistreams')
 161     if 'filename' in opts.compat_opts:
 162         if opts.outtmpl.get('default') is None:
 163             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 164         else:
 165             _unused_compat_opt('filename')
 166
 167
 168 def validate_options(opts):
 169     def validate(cndn, name, value=None, msg=None):
 170         if cndn:
 171             return True
 172         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 173
 174     def validate_in(name, value, items, msg=None):
 175         return validate(value is None or value in items, name, value, msg)
 176
 177     def validate_regex(name, value, regex):
 178         return validate(value is None or re.match(regex, value), name, value)
 179
 180     def validate_positive(name, value, strict=False):
 181         return validate(value is None or value > 0 or (not strict and value == 0),
 182                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 183
 184     def validate_minmax(min_val, max_val, min_name, max_name=None):
 185         if max_val is None or min_val is None or max_val >= min_val:
 186             return
 187         if not max_name:
 188             min_name, max_name = f'min {min_name}', f'max {min_name}'
 189         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 190
 191     # Usernames and passwords
 192     validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
 193              msg='{name}, netrc command and username/password are mutually exclusive options')
 194     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 195     validate(opts.ap_password is None or opts.ap_username is not None,
 196              'TV Provider account username', msg='{name} missing')
 197     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 198                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 199
 200     # Numbers
 201     validate_positive('autonumber start', opts.autonumber_start)
 202     validate_positive('autonumber size', opts.autonumber_size, True)
 203     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 204     validate_positive('playlist start', opts.playliststart, True)
 205     if opts.playlistend != -1:
 206         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 207
 208     # Time ranges
 209     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 210     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 211     validate_positive('sleep interval', opts.sleep_interval)
 212     validate_positive('max sleep interval', opts.max_sleep_interval)
 213     if opts.sleep_interval is None:
 214         validate(
 215             opts.max_sleep_interval is None, 'min sleep interval',
 216             msg='{name} must be specified; use --min-sleep-interval')
 217     elif opts.max_sleep_interval is None:
 218         opts.max_sleep_interval = opts.sleep_interval
 219     else:
 220         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 221
 222     if opts.wait_for_video is not None:
 223         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 224         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 225                  'time range to wait for video', opts.wait_for_video)
 226         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 227         opts.wait_for_video = (min_wait, max_wait)
 228
 229     # Format sort
 230     for f in opts.format_sort:
 231         validate_regex('format sorting', f, FormatSorter.regex)
 232
 233     # Postprocessor formats
 234     validate_regex('merge output format', opts.merge_output_format,
 235                    r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
 236     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 237     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 238     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 239     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 240     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 241     if opts.audioquality:
 242         opts.audioquality = opts.audioquality.strip('k').strip('K')
 243         # int_or_none prevents inf, nan
 244         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 245
 246     # Retries
 247     def parse_retries(name, value):
 248         if value is None:
 249             return None
 250         elif value in ('inf', 'infinite'):
 251             return float('inf')
 252         try:
 253             return int(value)
 254         except (TypeError, ValueError):
 255             validate(False, f'{name} retry count', value)
 256
 257     opts.retries = parse_retries('download', opts.retries)
 258     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 259     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 260     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 261
 262     # Retry sleep function
 263     def parse_sleep_func(expr):
 264         NUMBER_RE = r'\d+(?:\.\d+)?'
 265         op, start, limit, step, *_ = tuple(re.fullmatch(
 266             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 267             expr.strip()).groups()) + (None, None)
 268
 269         if op == 'exp':
 270             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 271         else:
 272             default_step = start if op or limit else 0
 273             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 274
 275     for key, expr in opts.retry_sleep.items():
 276         if not expr:
 277             del opts.retry_sleep[key]
 278             continue
 279         try:
 280             opts.retry_sleep[key] = parse_sleep_func(expr)
 281         except AttributeError:
 282             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 283
 284     # Bytes
 285     def validate_bytes(name, value):
 286         if value is None:
 287             return None
 288         numeric_limit = parse_bytes(value)
 289         validate(numeric_limit is not None, 'rate limit', value)
 290         return numeric_limit
 291
 292     opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
 293     opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
 294     opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
 295     opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
 296     opts.buffersize = validate_bytes('buffer size', opts.buffersize)
 297     opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
 298
 299     # Output templates
 300     def validate_outtmpl(tmpl, msg):
 301         err = YoutubeDL.validate_outtmpl(tmpl)
 302         if err:
 303             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 304
 305     for k, tmpl in opts.outtmpl.items():
 306         validate_outtmpl(tmpl, f'{k} output template')
 307     for type_, tmpl_list in opts.forceprint.items():
 308         for tmpl in tmpl_list:
 309             validate_outtmpl(tmpl, f'{type_} print template')
 310     for type_, tmpl_list in opts.print_to_file.items():
 311         for tmpl, file in tmpl_list:
 312             validate_outtmpl(tmpl, f'{type_} print to file template')
 313             validate_outtmpl(file, f'{type_} print to file filename')
 314     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 315     for k, tmpl in opts.progress_template.items():
 316         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 317         validate_outtmpl(tmpl, f'{k} template')
 318
 319     outtmpl_default = opts.outtmpl.get('default')
 320     if outtmpl_default == '':
 321         opts.skip_download = None
 322         del opts.outtmpl['default']
 323
 324     def parse_chapters(name, value, advanced=False):
 325         parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
 326         TIMESTAMP_RE = r'''(?x)(?:
 327             (?P<start_sign>-?)(?P<start>[^-]+)
 328         )?\s*-\s*(?:
 329             (?P<end_sign>-?)(?P<end>[^-]+)
 330         )?'''
 331
 332         chapters, ranges, from_url = [], [], False
 333         for regex in value or []:
 334             if advanced and regex == '*from-url':
 335                 from_url = True
 336                 continue
 337             elif not regex.startswith('*'):
 338                 try:
 339                     chapters.append(re.compile(regex))
 340                 except re.error as err:
 341                     raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 342                 continue
 343
 344             for range_ in map(str.strip, regex[1:].split(',')):
 345                 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
 346                 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
 347                 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
 348
 349                 err = None
 350                 if None in (dur or [None]):
 351                     err = 'Must be of the form "*start-end"'
 352                 elif not advanced and any(signs):
 353                     err = 'Negative timestamps are not allowed'
 354                 else:
 355                     dur[0] *= -1 if signs[0] else 1
 356                     dur[1] *= -1 if signs[1] else 1
 357                     if dur[1] == float('-inf'):
 358                         err = '"-inf" is not a valid end'
 359                 if err:
 360                     raise ValueError(f'invalid {name} time range "{regex}". {err}')
 361                 ranges.append(dur)
 362
 363         return chapters, ranges, from_url
 364
 365     opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
 366     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
 367
 368     # Cookies from browser
 369     if opts.cookiesfrombrowser:
 370         container = None
 371         mobj = re.fullmatch(r'''(?x)
 372             (?P<name>[^+:]+)
 373             (?:\s*\+\s*(?P<keyring>[^:]+))?
 374             (?:\s*:\s*(?!:)(?P<profile>.+?))?
 375             (?:\s*::\s*(?P<container>.+))?
 376         ''', opts.cookiesfrombrowser)
 377         if mobj is None:
 378             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 379         browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
 380         browser_name = browser_name.lower()
 381         if browser_name not in SUPPORTED_BROWSERS:
 382             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 383                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 384         if keyring is not None:
 385             keyring = keyring.upper()
 386             if keyring not in SUPPORTED_KEYRINGS:
 387                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 388                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 389         opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
 390
 391     # MetadataParser
 392     def metadataparser_actions(f):
 393         if isinstance(f, str):
 394             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 395             try:
 396                 actions = [MetadataFromFieldPP.to_action(f)]
 397             except Exception as err:
 398                 raise ValueError(f'{cmd} is invalid; {err}')
 399         else:
 400             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 401             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 402
 403         for action in actions:
 404             try:
 405                 MetadataParserPP.validate_action(*action)
 406             except Exception as err:
 407                 raise ValueError(f'{cmd} is invalid; {err}')
 408             yield action
 409
 410     if opts.metafromtitle is not None:
 411         opts.parse_metadata.setdefault('pre_process', []).append('title:%s' % opts.metafromtitle)
 412     opts.parse_metadata = {
 413         k: list(itertools.chain(*map(metadataparser_actions, v)))
 414         for k, v in opts.parse_metadata.items()
 415     }
 416
 417     # Other options
 418     if opts.playlist_items is not None:
 419         try:
 420             tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
 421         except Exception as err:
 422             raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
 423
 424     opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
 425     if opts.geo_bypass.lower() not in ('default', 'never'):
 426         try:
 427             GeoUtils.random_ipv4(opts.geo_bypass)
 428         except Exception:
 429             raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
 430         if len(opts.geo_bypass) == 2:
 431             opts.geo_bypass_country = opts.geo_bypass
 432         else:
 433             opts.geo_bypass_ip_block = opts.geo_bypass
 434     opts.geo_bypass = opts.geo_bypass.lower() != 'never'
 435
 436     opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
 437
 438     if opts.download_archive is not None:
 439         opts.download_archive = expand_path(opts.download_archive)
 440
 441     if opts.ffmpeg_location is not None:
 442         opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
 443
 444     if opts.user_agent is not None:
 445         opts.headers.setdefault('User-Agent', opts.user_agent)
 446     if opts.referer is not None:
 447         opts.headers.setdefault('Referer', opts.referer)
 448
 449     if opts.no_sponsorblock:
 450         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 451
 452     default_downloader = None
 453     for proto, path in opts.external_downloader.items():
 454         if path == 'native':
 455             continue
 456         ed = get_external_downloader(path)
 457         if ed is None:
 458             raise ValueError(
 459                 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
 460         elif ed and proto == 'default':
 461             default_downloader = ed.get_basename()
 462
 463     for policy in opts.color.values():
 464         if policy not in ('always', 'auto', 'no_color', 'never'):
 465             raise ValueError(f'"{policy}" is not a valid color policy')
 466
 467     warnings, deprecation_warnings = [], []
 468
 469     # Common mistake: -f best
 470     if opts.format == 'best':
 471         warnings.append('.\n         '.join((
 472             '"-f best" selects the best pre-merged format which is often not the best option',
 473             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 474             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 475
 476     # --(postprocessor/downloader)-args without name
 477     def report_args_compat(name, value, key1, key2=None, where=None):
 478         if key1 in value and key2 not in value:
 479             warnings.append(f'{name.title()} arguments given without specifying name. '
 480                             f'The arguments will be given to {where or f"all {name}s"}')
 481             return True
 482         return False
 483
 484     if report_args_compat('external downloader', opts.external_downloader_args,
 485                           'default', where=default_downloader) and default_downloader:
 486         # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
 487         opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
 488
 489     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 490         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 491         opts.postprocessor_args.setdefault('sponskrub', [])
 492
 493     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 494                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 495         if val2 is NO_DEFAULT:
 496             val2 = getattr(opts, opt2)
 497         if not val2:
 498             return
 499
 500         if val1 is NO_DEFAULT:
 501             val1 = getattr(opts, opt1)
 502         if val1:
 503             warnings.append(f'{arg1} is ignored since {arg2} was given')
 504         setattr(opts, opt1, default)
 505
 506     # Conflicting options
 507     report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
 508     report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
 509     report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
 510     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 511     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 512     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 513                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 514     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 515     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 516     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 517     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 518     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 519     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 520                     val1=opts.sponskrub and opts.sponskrub_cut)
 521
 522     # Conflicts with --allow-unplayable-formats
 523     report_conflict('--embed-metadata', 'addmetadata')
 524     report_conflict('--embed-chapters', 'addchapters')
 525     report_conflict('--embed-info-json', 'embed_infojson')
 526     report_conflict('--embed-subs', 'embedsubtitles')
 527     report_conflict('--embed-thumbnail', 'embedthumbnail')
 528     report_conflict('--extract-audio', 'extractaudio')
 529     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 530     report_conflict('--recode-video', 'recodevideo')
 531     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 532     report_conflict('--remux-video', 'remuxvideo')
 533     report_conflict('--sponskrub', 'sponskrub')
 534     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 535     report_conflict('--xattrs', 'xattrs')
 536
 537     # Fully deprecated options
 538     def report_deprecation(val, old, new=None):
 539         if not val:
 540             return
 541         deprecation_warnings.append(
 542             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 543             else f'{old} is deprecated and may not work as expected')
 544
 545     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 546     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 547     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 548     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 549     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 550
 551     # Dependent options
 552     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 553
 554     if opts.exec_before_dl_cmd:
 555         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 556
 557     if opts.useid:  # --id is not deprecated in youtube-dl
 558         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 559
 560     if opts.overwrites:  # --force-overwrites implies --no-continue
 561         opts.continue_dl = False
 562
 563     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 564         # Add chapters when adding metadata or marking sponsors
 565         opts.addchapters = True
 566
 567     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 568         # Do not unnecessarily download audio
 569         opts.format = 'bestaudio/best'
 570
 571     if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
 572         # If JSON is not printed anywhere, but comments are requested, save it to file
 573         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 574             opts.writeinfojson = True
 575
 576     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 577         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 578         opts.writesubtitles = True
 579
 580     if opts.addmetadata and opts.embed_infojson is None:
 581         # If embedding metadata and infojson is present, embed it
 582         opts.embed_infojson = 'if_exists'
 583
 584     # Ask for passwords
 585     if opts.username is not None and opts.password is None:
 586         opts.password = getpass.getpass('Type account password and press [Return]: ')
 587     if opts.ap_username is not None and opts.ap_password is None:
 588         opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 589
 590     return warnings, deprecation_warnings
 591
 592
 593 def get_postprocessors(opts):
 594     yield from opts.add_postprocessors
 595
 596     for when, actions in opts.parse_metadata.items():
 597         yield {
 598             'key': 'MetadataParser',
 599             'actions': actions,
 600             'when': when
 601         }
 602     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 603     if sponsorblock_query:
 604         yield {
 605             'key': 'SponsorBlock',
 606             'categories': sponsorblock_query,
 607             'api': opts.sponsorblock_api,
 608             'when': 'after_filter'
 609         }
 610     if opts.convertsubtitles:
 611         yield {
 612             'key': 'FFmpegSubtitlesConvertor',
 613             'format': opts.convertsubtitles,
 614             'when': 'before_dl'
 615         }
 616     if opts.convertthumbnails:
 617         yield {
 618             'key': 'FFmpegThumbnailsConvertor',
 619             'format': opts.convertthumbnails,
 620             'when': 'before_dl'
 621         }
 622     if opts.extractaudio:
 623         yield {
 624             'key': 'FFmpegExtractAudio',
 625             'preferredcodec': opts.audioformat,
 626             'preferredquality': opts.audioquality,
 627             'nopostoverwrites': opts.nopostoverwrites,
 628         }
 629     if opts.remuxvideo:
 630         yield {
 631             'key': 'FFmpegVideoRemuxer',
 632             'preferedformat': opts.remuxvideo,
 633         }
 634     if opts.recodevideo:
 635         yield {
 636             'key': 'FFmpegVideoConvertor',
 637             'preferedformat': opts.recodevideo,
 638         }
 639     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 640     if opts.embedsubtitles:
 641         keep_subs = 'no-keep-subs' not in opts.compat_opts
 642         yield {
 643             'key': 'FFmpegEmbedSubtitle',
 644             # already_have_subtitle = True prevents the file from being deleted after embedding
 645             'already_have_subtitle': opts.writesubtitles and keep_subs
 646         }
 647         if not opts.writeautomaticsub and keep_subs:
 648             opts.writesubtitles = True
 649
 650     # ModifyChapters must run before FFmpegMetadataPP
 651     if opts.remove_chapters or sponsorblock_query:
 652         yield {
 653             'key': 'ModifyChapters',
 654             'remove_chapters_patterns': opts.remove_chapters,
 655             'remove_sponsor_segments': opts.sponsorblock_remove,
 656             'remove_ranges': opts.remove_ranges,
 657             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 658             'force_keyframes': opts.force_keyframes_at_cuts
 659         }
 660     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 661     # FFmpegExtractAudioPP as containers before conversion may not support
 662     # metadata (3gp, webm, etc.)
 663     # By default ffmpeg preserves metadata applicable for both
 664     # source and target containers. From this point the container won't change,
 665     # so metadata can be added here.
 666     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 667         yield {
 668             'key': 'FFmpegMetadata',
 669             'add_chapters': opts.addchapters,
 670             'add_metadata': opts.addmetadata,
 671             'add_infojson': opts.embed_infojson,
 672         }
 673     # Deprecated
 674     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 675     # but must be below EmbedSubtitle and FFmpegMetadata
 676     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 677     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 678     if opts.sponskrub is not False:
 679         yield {
 680             'key': 'SponSkrub',
 681             'path': opts.sponskrub_path,
 682             'args': opts.sponskrub_args,
 683             'cut': opts.sponskrub_cut,
 684             'force': opts.sponskrub_force,
 685             'ignoreerror': opts.sponskrub is None,
 686             '_from_cli': True,
 687         }
 688     if opts.embedthumbnail:
 689         yield {
 690             'key': 'EmbedThumbnail',
 691             # already_have_thumbnail = True prevents the file from being deleted after embedding
 692             'already_have_thumbnail': opts.writethumbnail
 693         }
 694         if not opts.writethumbnail:
 695             opts.writethumbnail = True
 696             opts.outtmpl['pl_thumbnail'] = ''
 697     if opts.split_chapters:
 698         yield {
 699             'key': 'FFmpegSplitChapters',
 700             'force_keyframes': opts.force_keyframes_at_cuts,
 701         }
 702     # XAttrMetadataPP should be run after post-processors that may change file contents
 703     if opts.xattrs:
 704         yield {'key': 'XAttrMetadata'}
 705     if opts.concat_playlist != 'never':
 706         yield {
 707             'key': 'FFmpegConcat',
 708             'only_multi_video': opts.concat_playlist != 'always',
 709             'when': 'playlist',
 710         }
 711     # Exec must be the last PP of each category
 712     for when, exec_cmd in opts.exec_cmd.items():
 713         yield {
 714             'key': 'Exec',
 715             'exec_cmd': exec_cmd,
 716             'when': when,
 717         }
 718
 719
 720 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
 721
 722
 723 def parse_options(argv=None):
 724     """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
 725     parser, opts, urls = parseOpts(argv)
 726     urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
 727
 728     set_compat_opts(opts)
 729     try:
 730         warnings, deprecation_warnings = validate_options(opts)
 731     except ValueError as err:
 732         parser.error(f'{err}\n')
 733
 734     postprocessors = list(get_postprocessors(opts))
 735
 736     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
 737     any_getting = any(getattr(opts, k) for k in (
 738         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 739         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 740     ))
 741     if opts.quiet is None:
 742         opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
 743
 744     playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
 745     write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
 746                                and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
 747     if not any((
 748         opts.extract_flat,
 749         opts.dump_single_json,
 750         opts.forceprint.get('playlist'),
 751         opts.print_to_file.get('playlist'),
 752         write_playlist_infojson,
 753     )):
 754         if not playlist_pps:
 755             opts.extract_flat = 'discard'
 756         elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
 757             opts.extract_flat = 'discard_in_playlist'
 758
 759     final_ext = (
 760         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 761         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 762         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 763         else None)
 764
 765     return ParsedOptions(parser, opts, urls, {
 766         'usenetrc': opts.usenetrc,
 767         'netrc_location': opts.netrc_location,
 768         'netrc_cmd': opts.netrc_cmd,
 769         'username': opts.username,
 770         'password': opts.password,
 771         'twofactor': opts.twofactor,
 772         'videopassword': opts.videopassword,
 773         'ap_mso': opts.ap_mso,
 774         'ap_username': opts.ap_username,
 775         'ap_password': opts.ap_password,
 776         'client_certificate': opts.client_certificate,
 777         'client_certificate_key': opts.client_certificate_key,
 778         'client_certificate_password': opts.client_certificate_password,
 779         'quiet': opts.quiet,
 780         'no_warnings': opts.no_warnings,
 781         'forceurl': opts.geturl,
 782         'forcetitle': opts.gettitle,
 783         'forceid': opts.getid,
 784         'forcethumbnail': opts.getthumbnail,
 785         'forcedescription': opts.getdescription,
 786         'forceduration': opts.getduration,
 787         'forcefilename': opts.getfilename,
 788         'forceformat': opts.getformat,
 789         'forceprint': opts.forceprint,
 790         'print_to_file': opts.print_to_file,
 791         'forcejson': opts.dumpjson or opts.print_json,
 792         'dump_single_json': opts.dump_single_json,
 793         'force_write_download_archive': opts.force_write_download_archive,
 794         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 795         'skip_download': opts.skip_download,
 796         'format': opts.format,
 797         'allow_unplayable_formats': opts.allow_unplayable_formats,
 798         'ignore_no_formats_error': opts.ignore_no_formats_error,
 799         'format_sort': opts.format_sort,
 800         'format_sort_force': opts.format_sort_force,
 801         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 802         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 803         'check_formats': opts.check_formats,
 804         'listformats': opts.listformats,
 805         'listformats_table': opts.listformats_table,
 806         'outtmpl': opts.outtmpl,
 807         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 808         'paths': opts.paths,
 809         'autonumber_size': opts.autonumber_size,
 810         'autonumber_start': opts.autonumber_start,
 811         'restrictfilenames': opts.restrictfilenames,
 812         'windowsfilenames': opts.windowsfilenames,
 813         'ignoreerrors': opts.ignoreerrors,
 814         'force_generic_extractor': opts.force_generic_extractor,
 815         'allowed_extractors': opts.allowed_extractors or ['default'],
 816         'ratelimit': opts.ratelimit,
 817         'throttledratelimit': opts.throttledratelimit,
 818         'overwrites': opts.overwrites,
 819         'retries': opts.retries,
 820         'file_access_retries': opts.file_access_retries,
 821         'fragment_retries': opts.fragment_retries,
 822         'extractor_retries': opts.extractor_retries,
 823         'retry_sleep_functions': opts.retry_sleep,
 824         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 825         'keep_fragments': opts.keep_fragments,
 826         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 827         'buffersize': opts.buffersize,
 828         'noresizebuffer': opts.noresizebuffer,
 829         'http_chunk_size': opts.http_chunk_size,
 830         'continuedl': opts.continue_dl,
 831         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 832         'progress_with_newline': opts.progress_with_newline,
 833         'progress_template': opts.progress_template,
 834         'playliststart': opts.playliststart,
 835         'playlistend': opts.playlistend,
 836         'playlistreverse': opts.playlist_reverse,
 837         'playlistrandom': opts.playlist_random,
 838         'lazy_playlist': opts.lazy_playlist,
 839         'noplaylist': opts.noplaylist,
 840         'logtostderr': opts.outtmpl.get('default') == '-',
 841         'consoletitle': opts.consoletitle,
 842         'nopart': opts.nopart,
 843         'updatetime': opts.updatetime,
 844         'writedescription': opts.writedescription,
 845         'writeannotations': opts.writeannotations,
 846         'writeinfojson': opts.writeinfojson,
 847         'allow_playlist_files': opts.allow_playlist_files,
 848         'clean_infojson': opts.clean_infojson,
 849         'getcomments': opts.getcomments,
 850         'writethumbnail': opts.writethumbnail is True,
 851         'write_all_thumbnails': opts.writethumbnail == 'all',
 852         'writelink': opts.writelink,
 853         'writeurllink': opts.writeurllink,
 854         'writewebloclink': opts.writewebloclink,
 855         'writedesktoplink': opts.writedesktoplink,
 856         'writesubtitles': opts.writesubtitles,
 857         'writeautomaticsub': opts.writeautomaticsub,
 858         'allsubtitles': opts.allsubtitles,
 859         'listsubtitles': opts.listsubtitles,
 860         'subtitlesformat': opts.subtitlesformat,
 861         'subtitleslangs': opts.subtitleslangs,
 862         'matchtitle': decodeOption(opts.matchtitle),
 863         'rejecttitle': decodeOption(opts.rejecttitle),
 864         'max_downloads': opts.max_downloads,
 865         'prefer_free_formats': opts.prefer_free_formats,
 866         'trim_file_name': opts.trim_file_name,
 867         'verbose': opts.verbose,
 868         'dump_intermediate_pages': opts.dump_intermediate_pages,
 869         'write_pages': opts.write_pages,
 870         'load_pages': opts.load_pages,
 871         'test': opts.test,
 872         'keepvideo': opts.keepvideo,
 873         'min_filesize': opts.min_filesize,
 874         'max_filesize': opts.max_filesize,
 875         'min_views': opts.min_views,
 876         'max_views': opts.max_views,
 877         'daterange': opts.date,
 878         'cachedir': opts.cachedir,
 879         'youtube_print_sig_code': opts.youtube_print_sig_code,
 880         'age_limit': opts.age_limit,
 881         'download_archive': opts.download_archive,
 882         'break_on_existing': opts.break_on_existing,
 883         'break_on_reject': opts.break_on_reject,
 884         'break_per_url': opts.break_per_url,
 885         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 886         'cookiefile': opts.cookiefile,
 887         'cookiesfrombrowser': opts.cookiesfrombrowser,
 888         'legacyserverconnect': opts.legacy_server_connect,
 889         'nocheckcertificate': opts.no_check_certificate,
 890         'prefer_insecure': opts.prefer_insecure,
 891         'enable_file_urls': opts.enable_file_urls,
 892         'http_headers': opts.headers,
 893         'proxy': opts.proxy,
 894         'socket_timeout': opts.socket_timeout,
 895         'bidi_workaround': opts.bidi_workaround,
 896         'debug_printtraffic': opts.debug_printtraffic,
 897         'prefer_ffmpeg': opts.prefer_ffmpeg,
 898         'include_ads': opts.include_ads,
 899         'default_search': opts.default_search,
 900         'dynamic_mpd': opts.dynamic_mpd,
 901         'extractor_args': opts.extractor_args,
 902         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 903         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 904         'encoding': opts.encoding,
 905         'extract_flat': opts.extract_flat,
 906         'live_from_start': opts.live_from_start,
 907         'wait_for_video': opts.wait_for_video,
 908         'mark_watched': opts.mark_watched,
 909         'merge_output_format': opts.merge_output_format,
 910         'final_ext': final_ext,
 911         'postprocessors': postprocessors,
 912         'fixup': opts.fixup,
 913         'source_address': opts.source_address,
 914         'call_home': opts.call_home,
 915         'sleep_interval_requests': opts.sleep_interval_requests,
 916         'sleep_interval': opts.sleep_interval,
 917         'max_sleep_interval': opts.max_sleep_interval,
 918         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 919         'external_downloader': opts.external_downloader,
 920         'download_ranges': opts.download_ranges,
 921         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 922         'list_thumbnails': opts.list_thumbnails,
 923         'playlist_items': opts.playlist_items,
 924         'xattr_set_filesize': opts.xattr_set_filesize,
 925         'match_filter': opts.match_filter,
 926         'color': opts.color,
 927         'ffmpeg_location': opts.ffmpeg_location,
 928         'hls_prefer_native': opts.hls_prefer_native,
 929         'hls_use_mpegts': opts.hls_use_mpegts,
 930         'hls_split_discontinuity': opts.hls_split_discontinuity,
 931         'external_downloader_args': opts.external_downloader_args,
 932         'postprocessor_args': opts.postprocessor_args,
 933         'cn_verification_proxy': opts.cn_verification_proxy,
 934         'geo_verification_proxy': opts.geo_verification_proxy,
 935         'geo_bypass': opts.geo_bypass,
 936         'geo_bypass_country': opts.geo_bypass_country,
 937         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 938         '_warnings': warnings,
 939         '_deprecation_warnings': deprecation_warnings,
 940         'compat_opts': opts.compat_opts,
 941     })
 942
 943
 944 def _real_main(argv=None):
 945     setproctitle('yt-dlp')
 946
 947     parser, opts, all_urls, ydl_opts = parse_options(argv)
 948
 949     # Dump user agent
 950     if opts.dump_user_agent:
 951         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 952         write_string(f'{ua}\n', out=sys.stdout)
 953         return
 954
 955     if print_extractor_information(opts, all_urls):
 956         return
 957
 958     # We may need ffmpeg_location without having access to the YoutubeDL instance
 959     # See https://github.com/yt-dlp/yt-dlp/issues/2191
 960     if opts.ffmpeg_location:
 961         FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
 962
 963     with YoutubeDL(ydl_opts) as ydl:
 964         pre_process = opts.update_self or opts.rm_cachedir
 965         actual_use = all_urls or opts.load_info_filename
 966
 967         if opts.rm_cachedir:
 968             ydl.cache.remove()
 969
 970         try:
 971             updater = Updater(ydl, opts.update_self)
 972             if opts.update_self and updater.update() and actual_use:
 973                 if updater.cmd:
 974                     return updater.restart()
 975                 # This code is reachable only for zip variant in py < 3.10
 976                 # It makes sense to exit here, but the old behavior is to continue
 977                 ydl.report_warning('Restart yt-dlp to use the updated version')
 978                 # return 100, 'ERROR: The program must exit for the update to complete'
 979         except Exception:
 980             traceback.print_exc()
 981             ydl._download_retcode = 100
 982
 983         if not actual_use:
 984             if pre_process:
 985                 return ydl._download_retcode
 986
 987             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 988             parser.error(
 989                 'You must provide at least one URL.\n'
 990                 'Type yt-dlp --help to see a list of all options.')
 991
 992         parser.destroy()
 993         try:
 994             if opts.load_info_filename is not None:
 995                 if all_urls:
 996                     ydl.report_warning('URLs are ignored due to --load-info-json')
 997                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 998             else:
 999                 return ydl.download(all_urls)
1000         except DownloadCancelled:
1001             ydl.to_screen('Aborting remaining downloads')
1002             return 101
1003
1004
1005 def main(argv=None):
1006     global _IN_CLI
1007     _IN_CLI = True
1008     try:
1009         _exit(*variadic(_real_main(argv)))
1010     except DownloadError:
1011         _exit(1)
1012     except SameFileError as e:
1013         _exit(f'ERROR: {e}')
1014     except KeyboardInterrupt:
1015         _exit('\nERROR: Interrupted by user')
1016     except BrokenPipeError as e:
1017         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1018         devnull = os.open(os.devnull, os.O_WRONLY)
1019         os.dup2(devnull, sys.stdout.fileno())
1020         _exit(f'\nERROR: {e}')
1021     except optparse.OptParseError as e:
1022         _exit(2, f'\n{e}')
1023
1024
1025 from .extractor import gen_extractors, list_extractors
1026
1027 __all__ = [
1028     'main',
1029     'YoutubeDL',
1030     'parse_options',
1031     'gen_extractors',
1032     'list_extractors',
1033 ]