yt_dlp/__init__.py

   1 #!/usr/bin/env python3
   2 f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp'  # noqa: F541
   3
   4 __license__ = 'Public Domain'
   5
   6 import itertools
   7 import optparse
   8 import os
   9 import re
  10 import sys
  11
  12 from .compat import compat_getpass, compat_shlex_quote
  13 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  14 from .downloader import FileDownloader
  15 from .extractor import GenericIE, list_extractor_classes
  16 from .extractor.adobepass import MSO_INFO
  17 from .extractor.common import InfoExtractor
  18 from .options import parseOpts
  19 from .postprocessor import (
  20     FFmpegExtractAudioPP,
  21     FFmpegSubtitlesConvertorPP,
  22     FFmpegThumbnailsConvertorPP,
  23     FFmpegVideoConvertorPP,
  24     FFmpegVideoRemuxerPP,
  25     MetadataFromFieldPP,
  26     MetadataParserPP,
  27 )
  28 from .update import run_update
  29 from .utils import (
  30     NO_DEFAULT,
  31     POSTPROCESS_WHEN,
  32     DateRange,
  33     DownloadCancelled,
  34     DownloadError,
  35     GeoUtils,
  36     SameFileError,
  37     decodeOption,
  38     expand_path,
  39     float_or_none,
  40     int_or_none,
  41     match_filter_func,
  42     parse_duration,
  43     preferredencoding,
  44     read_batch_urls,
  45     read_stdin,
  46     render_table,
  47     setproctitle,
  48     std_headers,
  49     traverse_obj,
  50     variadic,
  51     write_string,
  52 )
  53 from .YoutubeDL import YoutubeDL
  54
  55
  56 def _exit(status=0, *args):
  57     for msg in args:
  58         sys.stderr.write(msg)
  59     raise SystemExit(status)
  60
  61
  62 def get_urls(urls, batchfile, verbose):
  63     # Batch file verification
  64     batch_urls = []
  65     if batchfile is not None:
  66         try:
  67             batch_urls = read_batch_urls(
  68                 read_stdin('URLs') if batchfile == '-'
  69                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  70             if verbose:
  71                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  72         except OSError:
  73             _exit(f'ERROR: batch file {batchfile} could not be read')
  74     _enc = preferredencoding()
  75     return [
  76         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  77         for url in batch_urls + urls]
  78
  79
  80 def print_extractor_information(opts, urls):
  81     out = ''
  82     if opts.list_extractors:
  83         urls = dict.fromkeys(urls, False)
  84         for ie in list_extractor_classes(opts.age_limit):
  85             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
  86             if ie == GenericIE:
  87                 matched_urls = [url for url, matched in urls.items() if not matched]
  88             else:
  89                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
  90                 urls.update(dict.fromkeys(matched_urls, True))
  91             out += ''.join(f'  {url}\n' for url in matched_urls)
  92     elif opts.list_extractor_descriptions:
  93         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
  94         out = '\n'.join(
  95             ie.description(markdown=False, search_examples=_SEARCHES)
  96             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
  97     elif opts.ap_list_mso:
  98         out = 'Supported TV Providers:\n%s\n' % render_table(
  99             ['mso', 'mso name'],
 100             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 101     else:
 102         return False
 103     write_string(out, out=sys.stdout)
 104     return True
 105
 106
 107 def set_compat_opts(opts):
 108     def _unused_compat_opt(name):
 109         if name not in opts.compat_opts:
 110             return False
 111         opts.compat_opts.discard(name)
 112         opts.compat_opts.update(['*%s' % name])
 113         return True
 114
 115     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 116         attr = getattr(opts, opt_name)
 117         if compat_name in opts.compat_opts:
 118             if attr is None:
 119                 setattr(opts, opt_name, not default)
 120                 return True
 121             else:
 122                 if remove_compat:
 123                     _unused_compat_opt(compat_name)
 124                 return False
 125         elif attr is None:
 126             setattr(opts, opt_name, default)
 127         return None
 128
 129     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 130     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 131     set_default_compat('no-clean-infojson', 'clean_infojson')
 132     if 'no-attach-info-json' in opts.compat_opts:
 133         if opts.embed_infojson:
 134             _unused_compat_opt('no-attach-info-json')
 135         else:
 136             opts.embed_infojson = False
 137     if 'format-sort' in opts.compat_opts:
 138         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 139     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 140     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 141     if _video_multistreams_set is False and _audio_multistreams_set is False:
 142         _unused_compat_opt('multistreams')
 143     if 'filename' in opts.compat_opts:
 144         if opts.outtmpl.get('default') is None:
 145             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 146         else:
 147             _unused_compat_opt('filename')
 148
 149
 150 def validate_options(opts):
 151     def validate(cndn, name, value=None, msg=None):
 152         if cndn:
 153             return True
 154         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 155
 156     def validate_in(name, value, items, msg=None):
 157         return validate(value is None or value in items, name, value, msg)
 158
 159     def validate_regex(name, value, regex):
 160         return validate(value is None or re.match(regex, value), name, value)
 161
 162     def validate_positive(name, value, strict=False):
 163         return validate(value is None or value > 0 or (not strict and value == 0),
 164                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 165
 166     def validate_minmax(min_val, max_val, min_name, max_name=None):
 167         if max_val is None or min_val is None or max_val >= min_val:
 168             return
 169         if not max_name:
 170             min_name, max_name = f'min {min_name}', f'max {min_name}'
 171         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 172
 173     # Usernames and passwords
 174     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 175              '.netrc', msg='using {name} conflicts with giving username/password')
 176     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 177     validate(opts.ap_password is None or opts.ap_username is not None,
 178              'TV Provider account username', msg='{name} missing')
 179     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 180                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 181
 182     # Numbers
 183     validate_positive('autonumber start', opts.autonumber_start)
 184     validate_positive('autonumber size', opts.autonumber_size, True)
 185     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 186     validate_positive('playlist start', opts.playliststart, True)
 187     if opts.playlistend != -1:
 188         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 189
 190     # Time ranges
 191     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 192     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 193     validate_positive('sleep interval', opts.sleep_interval)
 194     validate_positive('max sleep interval', opts.max_sleep_interval)
 195     if opts.sleep_interval is None:
 196         validate(
 197             opts.max_sleep_interval is None, 'min sleep interval',
 198             msg='{name} must be specified; use --min-sleep-interval')
 199     elif opts.max_sleep_interval is None:
 200         opts.max_sleep_interval = opts.sleep_interval
 201     else:
 202         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 203
 204     if opts.wait_for_video is not None:
 205         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 206         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 207                  'time range to wait for video', opts.wait_for_video)
 208         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 209         opts.wait_for_video = (min_wait, max_wait)
 210
 211     # Format sort
 212     for f in opts.format_sort:
 213         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 214
 215     # Postprocessor formats
 216     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 217     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 218     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 219     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 220     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 221     if opts.audioquality:
 222         opts.audioquality = opts.audioquality.strip('k').strip('K')
 223         # int_or_none prevents inf, nan
 224         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 225
 226     # Retries
 227     def parse_retries(name, value):
 228         if value is None:
 229             return None
 230         elif value in ('inf', 'infinite'):
 231             return float('inf')
 232         try:
 233             return int(value)
 234         except (TypeError, ValueError):
 235             validate(False, f'{name} retry count', value)
 236
 237     opts.retries = parse_retries('download', opts.retries)
 238     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 239     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 240     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 241
 242     # Retry sleep function
 243     def parse_sleep_func(expr):
 244         NUMBER_RE = r'\d+(?:\.\d+)?'
 245         op, start, limit, step, *_ = tuple(re.fullmatch(
 246             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 247             expr.strip()).groups()) + (None, None)
 248
 249         if op == 'exp':
 250             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 251         else:
 252             default_step = start if op or limit else 0
 253             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 254
 255     for key, expr in opts.retry_sleep.items():
 256         if not expr:
 257             del opts.retry_sleep[key]
 258             continue
 259         try:
 260             opts.retry_sleep[key] = parse_sleep_func(expr)
 261         except AttributeError:
 262             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 263
 264     # Bytes
 265     def parse_bytes(name, value):
 266         if value is None:
 267             return None
 268         numeric_limit = FileDownloader.parse_bytes(value)
 269         validate(numeric_limit is not None, 'rate limit', value)
 270         return numeric_limit
 271
 272     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 273     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 274     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 275     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 276     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 277     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 278
 279     # Output templates
 280     def validate_outtmpl(tmpl, msg):
 281         err = YoutubeDL.validate_outtmpl(tmpl)
 282         if err:
 283             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 284
 285     for k, tmpl in opts.outtmpl.items():
 286         validate_outtmpl(tmpl, f'{k} output template')
 287     for type_, tmpl_list in opts.forceprint.items():
 288         for tmpl in tmpl_list:
 289             validate_outtmpl(tmpl, f'{type_} print template')
 290     for type_, tmpl_list in opts.print_to_file.items():
 291         for tmpl, file in tmpl_list:
 292             validate_outtmpl(tmpl, f'{type_} print to file template')
 293             validate_outtmpl(file, f'{type_} print to file filename')
 294     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 295     for k, tmpl in opts.progress_template.items():
 296         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 297         validate_outtmpl(tmpl, f'{k} template')
 298
 299     outtmpl_default = opts.outtmpl.get('default')
 300     if outtmpl_default == '':
 301         opts.skip_download = None
 302         del opts.outtmpl['default']
 303     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 304         raise ValueError(
 305             'Cannot download a video and extract audio into the same file! '
 306             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 307
 308     # Remove chapters
 309     remove_chapters_patterns, opts.remove_ranges = [], []
 310     for regex in opts.remove_chapters or []:
 311         if regex.startswith('*'):
 312             dur = list(map(parse_duration, regex[1:].split('-')))
 313             if len(dur) == 2 and all(t is not None for t in dur):
 314                 opts.remove_ranges.append(tuple(dur))
 315                 continue
 316             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 317         try:
 318             remove_chapters_patterns.append(re.compile(regex))
 319         except re.error as err:
 320             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 321     opts.remove_chapters = remove_chapters_patterns
 322
 323     # Cookies from browser
 324     if opts.cookiesfrombrowser:
 325         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 326         if mobj is None:
 327             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 328         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 329         browser_name = browser_name.lower()
 330         if browser_name not in SUPPORTED_BROWSERS:
 331             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 332                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 333         if keyring is not None:
 334             keyring = keyring.upper()
 335             if keyring not in SUPPORTED_KEYRINGS:
 336                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 337                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 338         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 339
 340     # MetadataParser
 341     def metadataparser_actions(f):
 342         if isinstance(f, str):
 343             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 344             try:
 345                 actions = [MetadataFromFieldPP.to_action(f)]
 346             except Exception as err:
 347                 raise ValueError(f'{cmd} is invalid; {err}')
 348         else:
 349             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 350             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 351
 352         for action in actions:
 353             try:
 354                 MetadataParserPP.validate_action(*action)
 355             except Exception as err:
 356                 raise ValueError(f'{cmd} is invalid; {err}')
 357             yield action
 358
 359     parse_metadata = opts.parse_metadata or []
 360     if opts.metafromtitle is not None:
 361         parse_metadata.append('title:%s' % opts.metafromtitle)
 362     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 363
 364     # Other options
 365     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 366     if geo_bypass_code is not None:
 367         try:
 368             GeoUtils.random_ipv4(geo_bypass_code)
 369         except Exception:
 370             raise ValueError('unsupported geo-bypass country or ip-block')
 371
 372     opts.match_filter = match_filter_func(opts.match_filter)
 373
 374     if opts.download_archive is not None:
 375         opts.download_archive = expand_path(opts.download_archive)
 376
 377     if opts.user_agent is not None:
 378         opts.headers.setdefault('User-Agent', opts.user_agent)
 379     if opts.referer is not None:
 380         opts.headers.setdefault('Referer', opts.referer)
 381
 382     if opts.no_sponsorblock:
 383         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 384
 385     warnings, deprecation_warnings = [], []
 386
 387     # Common mistake: -f best
 388     if opts.format == 'best':
 389         warnings.append('.\n         '.join((
 390             '"-f best" selects the best pre-merged format which is often not the best option',
 391             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 392             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 393
 394     # --(postprocessor/downloader)-args without name
 395     def report_args_compat(name, value, key1, key2=None):
 396         if key1 in value and key2 not in value:
 397             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 398             return True
 399         return False
 400
 401     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 402     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 403         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 404         opts.postprocessor_args.setdefault('sponskrub', [])
 405
 406     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 407                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 408         if val2 is NO_DEFAULT:
 409             val2 = getattr(opts, opt2)
 410         if not val2:
 411             return
 412
 413         if val1 is NO_DEFAULT:
 414             val1 = getattr(opts, opt1)
 415         if val1:
 416             warnings.append(f'{arg1} is ignored since {arg2} was given')
 417         setattr(opts, opt1, default)
 418
 419     # Conflicting options
 420     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 421     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 422     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 423                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 424     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 425     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 426     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 427     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 428     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 429     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 430                     val1=opts.sponskrub and opts.sponskrub_cut)
 431
 432     # Conflicts with --allow-unplayable-formats
 433     report_conflict('--add-metadata', 'addmetadata')
 434     report_conflict('--embed-chapters', 'addchapters')
 435     report_conflict('--embed-info-json', 'embed_infojson')
 436     report_conflict('--embed-subs', 'embedsubtitles')
 437     report_conflict('--embed-thumbnail', 'embedthumbnail')
 438     report_conflict('--extract-audio', 'extractaudio')
 439     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 440     report_conflict('--recode-video', 'recodevideo')
 441     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 442     report_conflict('--remux-video', 'remuxvideo')
 443     report_conflict('--sponskrub', 'sponskrub')
 444     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 445     report_conflict('--xattrs', 'xattrs')
 446
 447     # Fully deprecated options
 448     def report_deprecation(val, old, new=None):
 449         if not val:
 450             return
 451         deprecation_warnings.append(
 452             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 453             else f'{old} is deprecated and may not work as expected')
 454
 455     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 456     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 457     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 458     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 459     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 460
 461     # Dependent options
 462     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 463
 464     if opts.exec_before_dl_cmd:
 465         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 466
 467     if opts.useid:  # --id is not deprecated in youtube-dl
 468         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 469
 470     if opts.overwrites:  # --force-overwrites implies --no-continue
 471         opts.continue_dl = False
 472
 473     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 474         # Add chapters when adding metadata or marking sponsors
 475         opts.addchapters = True
 476
 477     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 478         # Do not unnecessarily download audio
 479         opts.format = 'bestaudio/best'
 480
 481     if opts.getcomments and opts.writeinfojson is None:
 482         # If JSON is not printed anywhere, but comments are requested, save it to file
 483         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 484             opts.writeinfojson = True
 485
 486     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 487         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 488         opts.writesubtitles = True
 489
 490     if opts.addmetadata and opts.embed_infojson is None:
 491         # If embedding metadata and infojson is present, embed it
 492         opts.embed_infojson = 'if_exists'
 493
 494     # Ask for passwords
 495     if opts.username is not None and opts.password is None:
 496         opts.password = compat_getpass('Type account password and press [Return]: ')
 497     if opts.ap_username is not None and opts.ap_password is None:
 498         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 499
 500     return warnings, deprecation_warnings
 501
 502
 503 def get_postprocessors(opts):
 504     yield from opts.add_postprocessors
 505
 506     if opts.parse_metadata:
 507         yield {
 508             'key': 'MetadataParser',
 509             'actions': opts.parse_metadata,
 510             'when': 'pre_process'
 511         }
 512     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 513     if sponsorblock_query:
 514         yield {
 515             'key': 'SponsorBlock',
 516             'categories': sponsorblock_query,
 517             'api': opts.sponsorblock_api,
 518             'when': 'after_filter'
 519         }
 520     if opts.convertsubtitles:
 521         yield {
 522             'key': 'FFmpegSubtitlesConvertor',
 523             'format': opts.convertsubtitles,
 524             'when': 'before_dl'
 525         }
 526     if opts.convertthumbnails:
 527         yield {
 528             'key': 'FFmpegThumbnailsConvertor',
 529             'format': opts.convertthumbnails,
 530             'when': 'before_dl'
 531         }
 532     if opts.extractaudio:
 533         yield {
 534             'key': 'FFmpegExtractAudio',
 535             'preferredcodec': opts.audioformat,
 536             'preferredquality': opts.audioquality,
 537             'nopostoverwrites': opts.nopostoverwrites,
 538         }
 539     if opts.remuxvideo:
 540         yield {
 541             'key': 'FFmpegVideoRemuxer',
 542             'preferedformat': opts.remuxvideo,
 543         }
 544     if opts.recodevideo:
 545         yield {
 546             'key': 'FFmpegVideoConvertor',
 547             'preferedformat': opts.recodevideo,
 548         }
 549     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 550     if opts.embedsubtitles:
 551         keep_subs = 'no-keep-subs' not in opts.compat_opts
 552         yield {
 553             'key': 'FFmpegEmbedSubtitle',
 554             # already_have_subtitle = True prevents the file from being deleted after embedding
 555             'already_have_subtitle': opts.writesubtitles and keep_subs
 556         }
 557         if not opts.writeautomaticsub and keep_subs:
 558             opts.writesubtitles = True
 559
 560     # ModifyChapters must run before FFmpegMetadataPP
 561     if opts.remove_chapters or sponsorblock_query:
 562         yield {
 563             'key': 'ModifyChapters',
 564             'remove_chapters_patterns': opts.remove_chapters,
 565             'remove_sponsor_segments': opts.sponsorblock_remove,
 566             'remove_ranges': opts.remove_ranges,
 567             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 568             'force_keyframes': opts.force_keyframes_at_cuts
 569         }
 570     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 571     # FFmpegExtractAudioPP as containers before conversion may not support
 572     # metadata (3gp, webm, etc.)
 573     # By default ffmpeg preserves metadata applicable for both
 574     # source and target containers. From this point the container won't change,
 575     # so metadata can be added here.
 576     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 577         yield {
 578             'key': 'FFmpegMetadata',
 579             'add_chapters': opts.addchapters,
 580             'add_metadata': opts.addmetadata,
 581             'add_infojson': opts.embed_infojson,
 582         }
 583     # Deprecated
 584     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 585     # but must be below EmbedSubtitle and FFmpegMetadata
 586     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 587     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 588     if opts.sponskrub is not False:
 589         yield {
 590             'key': 'SponSkrub',
 591             'path': opts.sponskrub_path,
 592             'args': opts.sponskrub_args,
 593             'cut': opts.sponskrub_cut,
 594             'force': opts.sponskrub_force,
 595             'ignoreerror': opts.sponskrub is None,
 596             '_from_cli': True,
 597         }
 598     if opts.embedthumbnail:
 599         yield {
 600             'key': 'EmbedThumbnail',
 601             # already_have_thumbnail = True prevents the file from being deleted after embedding
 602             'already_have_thumbnail': opts.writethumbnail
 603         }
 604         if not opts.writethumbnail:
 605             opts.writethumbnail = True
 606             opts.outtmpl['pl_thumbnail'] = ''
 607     if opts.split_chapters:
 608         yield {
 609             'key': 'FFmpegSplitChapters',
 610             'force_keyframes': opts.force_keyframes_at_cuts,
 611         }
 612     # XAttrMetadataPP should be run after post-processors that may change file contents
 613     if opts.xattrs:
 614         yield {'key': 'XAttrMetadata'}
 615     if opts.concat_playlist != 'never':
 616         yield {
 617             'key': 'FFmpegConcat',
 618             'only_multi_video': opts.concat_playlist != 'always',
 619             'when': 'playlist',
 620         }
 621     # Exec must be the last PP of each category
 622     for when, exec_cmd in opts.exec_cmd.items():
 623         yield {
 624             'key': 'Exec',
 625             'exec_cmd': exec_cmd,
 626             'when': when,
 627         }
 628
 629
 630 def parse_options(argv=None):
 631     """ @returns (parser, opts, urls, ydl_opts) """
 632     parser, opts, urls = parseOpts(argv)
 633     urls = get_urls(urls, opts.batchfile, opts.verbose)
 634
 635     set_compat_opts(opts)
 636     try:
 637         warnings, deprecation_warnings = validate_options(opts)
 638     except ValueError as err:
 639         parser.error(f'{err}\n')
 640
 641     postprocessors = list(get_postprocessors(opts))
 642
 643     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[2:])
 644     any_getting = any(getattr(opts, k) for k in (
 645         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 646         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 647     ))
 648
 649     final_ext = (
 650         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 651         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 652         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 653         else None)
 654
 655     return parser, opts, urls, {
 656         'usenetrc': opts.usenetrc,
 657         'netrc_location': opts.netrc_location,
 658         'username': opts.username,
 659         'password': opts.password,
 660         'twofactor': opts.twofactor,
 661         'videopassword': opts.videopassword,
 662         'ap_mso': opts.ap_mso,
 663         'ap_username': opts.ap_username,
 664         'ap_password': opts.ap_password,
 665         'client_certificate': opts.client_certificate,
 666         'client_certificate_key': opts.client_certificate_key,
 667         'client_certificate_password': opts.client_certificate_password,
 668         'quiet': opts.quiet or any_getting or opts.print_json or bool(opts.forceprint),
 669         'no_warnings': opts.no_warnings,
 670         'forceurl': opts.geturl,
 671         'forcetitle': opts.gettitle,
 672         'forceid': opts.getid,
 673         'forcethumbnail': opts.getthumbnail,
 674         'forcedescription': opts.getdescription,
 675         'forceduration': opts.getduration,
 676         'forcefilename': opts.getfilename,
 677         'forceformat': opts.getformat,
 678         'forceprint': opts.forceprint,
 679         'print_to_file': opts.print_to_file,
 680         'forcejson': opts.dumpjson or opts.print_json,
 681         'dump_single_json': opts.dump_single_json,
 682         'force_write_download_archive': opts.force_write_download_archive,
 683         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 684         'skip_download': opts.skip_download,
 685         'format': opts.format,
 686         'allow_unplayable_formats': opts.allow_unplayable_formats,
 687         'ignore_no_formats_error': opts.ignore_no_formats_error,
 688         'format_sort': opts.format_sort,
 689         'format_sort_force': opts.format_sort_force,
 690         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 691         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 692         'check_formats': opts.check_formats,
 693         'listformats': opts.listformats,
 694         'listformats_table': opts.listformats_table,
 695         'outtmpl': opts.outtmpl,
 696         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 697         'paths': opts.paths,
 698         'autonumber_size': opts.autonumber_size,
 699         'autonumber_start': opts.autonumber_start,
 700         'restrictfilenames': opts.restrictfilenames,
 701         'windowsfilenames': opts.windowsfilenames,
 702         'ignoreerrors': opts.ignoreerrors,
 703         'force_generic_extractor': opts.force_generic_extractor,
 704         'ratelimit': opts.ratelimit,
 705         'throttledratelimit': opts.throttledratelimit,
 706         'overwrites': opts.overwrites,
 707         'retries': opts.retries,
 708         'file_access_retries': opts.file_access_retries,
 709         'fragment_retries': opts.fragment_retries,
 710         'extractor_retries': opts.extractor_retries,
 711         'retry_sleep_functions': opts.retry_sleep,
 712         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 713         'keep_fragments': opts.keep_fragments,
 714         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 715         'buffersize': opts.buffersize,
 716         'noresizebuffer': opts.noresizebuffer,
 717         'http_chunk_size': opts.http_chunk_size,
 718         'continuedl': opts.continue_dl,
 719         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 720         'progress_with_newline': opts.progress_with_newline,
 721         'progress_template': opts.progress_template,
 722         'playliststart': opts.playliststart,
 723         'playlistend': opts.playlistend,
 724         'playlistreverse': opts.playlist_reverse,
 725         'playlistrandom': opts.playlist_random,
 726         'noplaylist': opts.noplaylist,
 727         'logtostderr': opts.outtmpl.get('default') == '-',
 728         'consoletitle': opts.consoletitle,
 729         'nopart': opts.nopart,
 730         'updatetime': opts.updatetime,
 731         'writedescription': opts.writedescription,
 732         'writeannotations': opts.writeannotations,
 733         'writeinfojson': opts.writeinfojson,
 734         'allow_playlist_files': opts.allow_playlist_files,
 735         'clean_infojson': opts.clean_infojson,
 736         'getcomments': opts.getcomments,
 737         'writethumbnail': opts.writethumbnail is True,
 738         'write_all_thumbnails': opts.writethumbnail == 'all',
 739         'writelink': opts.writelink,
 740         'writeurllink': opts.writeurllink,
 741         'writewebloclink': opts.writewebloclink,
 742         'writedesktoplink': opts.writedesktoplink,
 743         'writesubtitles': opts.writesubtitles,
 744         'writeautomaticsub': opts.writeautomaticsub,
 745         'allsubtitles': opts.allsubtitles,
 746         'listsubtitles': opts.listsubtitles,
 747         'subtitlesformat': opts.subtitlesformat,
 748         'subtitleslangs': opts.subtitleslangs,
 749         'matchtitle': decodeOption(opts.matchtitle),
 750         'rejecttitle': decodeOption(opts.rejecttitle),
 751         'max_downloads': opts.max_downloads,
 752         'prefer_free_formats': opts.prefer_free_formats,
 753         'trim_file_name': opts.trim_file_name,
 754         'verbose': opts.verbose,
 755         'dump_intermediate_pages': opts.dump_intermediate_pages,
 756         'write_pages': opts.write_pages,
 757         'load_pages': opts.load_pages,
 758         'test': opts.test,
 759         'keepvideo': opts.keepvideo,
 760         'min_filesize': opts.min_filesize,
 761         'max_filesize': opts.max_filesize,
 762         'min_views': opts.min_views,
 763         'max_views': opts.max_views,
 764         'daterange': opts.date,
 765         'cachedir': opts.cachedir,
 766         'youtube_print_sig_code': opts.youtube_print_sig_code,
 767         'age_limit': opts.age_limit,
 768         'download_archive': opts.download_archive,
 769         'break_on_existing': opts.break_on_existing,
 770         'break_on_reject': opts.break_on_reject,
 771         'break_per_url': opts.break_per_url,
 772         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 773         'cookiefile': opts.cookiefile,
 774         'cookiesfrombrowser': opts.cookiesfrombrowser,
 775         'legacyserverconnect': opts.legacy_server_connect,
 776         'nocheckcertificate': opts.no_check_certificate,
 777         'prefer_insecure': opts.prefer_insecure,
 778         'http_headers': opts.headers,
 779         'proxy': opts.proxy,
 780         'socket_timeout': opts.socket_timeout,
 781         'bidi_workaround': opts.bidi_workaround,
 782         'debug_printtraffic': opts.debug_printtraffic,
 783         'prefer_ffmpeg': opts.prefer_ffmpeg,
 784         'include_ads': opts.include_ads,
 785         'default_search': opts.default_search,
 786         'dynamic_mpd': opts.dynamic_mpd,
 787         'extractor_args': opts.extractor_args,
 788         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 789         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 790         'encoding': opts.encoding,
 791         'extract_flat': opts.extract_flat,
 792         'live_from_start': opts.live_from_start,
 793         'wait_for_video': opts.wait_for_video,
 794         'mark_watched': opts.mark_watched,
 795         'merge_output_format': opts.merge_output_format,
 796         'final_ext': final_ext,
 797         'postprocessors': postprocessors,
 798         'fixup': opts.fixup,
 799         'source_address': opts.source_address,
 800         'call_home': opts.call_home,
 801         'sleep_interval_requests': opts.sleep_interval_requests,
 802         'sleep_interval': opts.sleep_interval,
 803         'max_sleep_interval': opts.max_sleep_interval,
 804         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 805         'external_downloader': opts.external_downloader,
 806         'list_thumbnails': opts.list_thumbnails,
 807         'playlist_items': opts.playlist_items,
 808         'xattr_set_filesize': opts.xattr_set_filesize,
 809         'match_filter': opts.match_filter,
 810         'no_color': opts.no_color,
 811         'ffmpeg_location': opts.ffmpeg_location,
 812         'hls_prefer_native': opts.hls_prefer_native,
 813         'hls_use_mpegts': opts.hls_use_mpegts,
 814         'hls_split_discontinuity': opts.hls_split_discontinuity,
 815         'external_downloader_args': opts.external_downloader_args,
 816         'postprocessor_args': opts.postprocessor_args,
 817         'cn_verification_proxy': opts.cn_verification_proxy,
 818         'geo_verification_proxy': opts.geo_verification_proxy,
 819         'geo_bypass': opts.geo_bypass,
 820         'geo_bypass_country': opts.geo_bypass_country,
 821         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 822         '_warnings': warnings,
 823         '_deprecation_warnings': deprecation_warnings,
 824         'compat_opts': opts.compat_opts,
 825     }
 826
 827
 828 def _real_main(argv=None):
 829     setproctitle('yt-dlp')
 830
 831     parser, opts, all_urls, ydl_opts = parse_options(argv)
 832
 833     # Dump user agent
 834     if opts.dump_user_agent:
 835         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 836         write_string(f'{ua}\n', out=sys.stdout)
 837         return
 838
 839     if print_extractor_information(opts, all_urls):
 840         return
 841
 842     with YoutubeDL(ydl_opts) as ydl:
 843         actual_use = all_urls or opts.load_info_filename
 844
 845         if opts.rm_cachedir:
 846             ydl.cache.remove()
 847
 848         if opts.update_self and run_update(ydl) and actual_use:
 849             # If updater returns True, exit. Required for windows
 850             return 100, 'ERROR: The program must exit for the update to complete'
 851
 852         if not actual_use:
 853             if opts.update_self or opts.rm_cachedir:
 854                 return ydl._download_retcode
 855
 856             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 857             parser.error(
 858                 'You must provide at least one URL.\n'
 859                 'Type yt-dlp --help to see a list of all options.')
 860
 861         parser.destroy()
 862         try:
 863             if opts.load_info_filename is not None:
 864                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 865             else:
 866                 return ydl.download(all_urls)
 867         except DownloadCancelled:
 868             ydl.to_screen('Aborting remaining downloads')
 869             return 101
 870
 871
 872 def main(argv=None):
 873     try:
 874         _exit(*variadic(_real_main(argv)))
 875     except DownloadError:
 876         _exit(1)
 877     except SameFileError as e:
 878         _exit(f'ERROR: {e}')
 879     except KeyboardInterrupt:
 880         _exit('\nERROR: Interrupted by user')
 881     except BrokenPipeError as e:
 882         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 883         devnull = os.open(os.devnull, os.O_WRONLY)
 884         os.dup2(devnull, sys.stdout.fileno())
 885         _exit(f'\nERROR: {e}')
 886     except optparse.OptParseError as e:
 887         _exit(2, f'\n{e}')
 888
 889
 890 from .extractor import gen_extractors, list_extractors
 891
 892 __all__ = [
 893     'main',
 894     'YoutubeDL',
 895     'parse_options',
 896     'gen_extractors',
 897     'list_extractors',
 898 ]