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_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
 217     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 218     validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
 219     if opts.recodevideo is not None:
 220         opts.recodevideo = opts.recodevideo.replace(' ', '')
 221         validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 222     if opts.remuxvideo is not None:
 223         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 224         validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 225     if opts.audioquality:
 226         opts.audioquality = opts.audioquality.strip('k').strip('K')
 227         # int_or_none prevents inf, nan
 228         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 229
 230     # Retries
 231     def parse_retries(name, value):
 232         if value is None:
 233             return None
 234         elif value in ('inf', 'infinite'):
 235             return float('inf')
 236         try:
 237             return int(value)
 238         except (TypeError, ValueError):
 239             validate(False, f'{name} retry count', value)
 240
 241     opts.retries = parse_retries('download', opts.retries)
 242     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 243     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 244     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 245
 246     # Retry sleep function
 247     def parse_sleep_func(expr):
 248         NUMBER_RE = r'\d+(?:\.\d+)?'
 249         op, start, limit, step, *_ = tuple(re.fullmatch(
 250             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 251             expr.strip()).groups()) + (None, None)
 252
 253         if op == 'exp':
 254             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 255         else:
 256             default_step = start if op or limit else 0
 257             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 258
 259     for key, expr in opts.retry_sleep.items():
 260         if not expr:
 261             del opts.retry_sleep[key]
 262             continue
 263         try:
 264             opts.retry_sleep[key] = parse_sleep_func(expr)
 265         except AttributeError:
 266             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 267
 268     # Bytes
 269     def parse_bytes(name, value):
 270         if value is None:
 271             return None
 272         numeric_limit = FileDownloader.parse_bytes(value)
 273         validate(numeric_limit is not None, 'rate limit', value)
 274         return numeric_limit
 275
 276     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 277     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 278     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 279     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 280     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 281     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 282
 283     # Output templates
 284     def validate_outtmpl(tmpl, msg):
 285         err = YoutubeDL.validate_outtmpl(tmpl)
 286         if err:
 287             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 288
 289     for k, tmpl in opts.outtmpl.items():
 290         validate_outtmpl(tmpl, f'{k} output template')
 291     for type_, tmpl_list in opts.forceprint.items():
 292         for tmpl in tmpl_list:
 293             validate_outtmpl(tmpl, f'{type_} print template')
 294     for type_, tmpl_list in opts.print_to_file.items():
 295         for tmpl, file in tmpl_list:
 296             validate_outtmpl(tmpl, f'{type_} print to file template')
 297             validate_outtmpl(file, f'{type_} print to file filename')
 298     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 299     for k, tmpl in opts.progress_template.items():
 300         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 301         validate_outtmpl(tmpl, f'{k} template')
 302
 303     outtmpl_default = opts.outtmpl.get('default')
 304     if outtmpl_default == '':
 305         opts.skip_download = None
 306         del opts.outtmpl['default']
 307     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 308         raise ValueError(
 309             'Cannot download a video and extract audio into the same file! '
 310             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 311
 312     # Remove chapters
 313     remove_chapters_patterns, opts.remove_ranges = [], []
 314     for regex in opts.remove_chapters or []:
 315         if regex.startswith('*'):
 316             dur = list(map(parse_duration, regex[1:].split('-')))
 317             if len(dur) == 2 and all(t is not None for t in dur):
 318                 opts.remove_ranges.append(tuple(dur))
 319                 continue
 320             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 321         try:
 322             remove_chapters_patterns.append(re.compile(regex))
 323         except re.error as err:
 324             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 325     opts.remove_chapters = remove_chapters_patterns
 326
 327     # Cookies from browser
 328     if opts.cookiesfrombrowser:
 329         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 330         if mobj is None:
 331             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 332         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 333         browser_name = browser_name.lower()
 334         if browser_name not in SUPPORTED_BROWSERS:
 335             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 336                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 337         if keyring is not None:
 338             keyring = keyring.upper()
 339             if keyring not in SUPPORTED_KEYRINGS:
 340                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 341                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 342         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 343
 344     # MetadataParser
 345     def metadataparser_actions(f):
 346         if isinstance(f, str):
 347             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 348             try:
 349                 actions = [MetadataFromFieldPP.to_action(f)]
 350             except Exception as err:
 351                 raise ValueError(f'{cmd} is invalid; {err}')
 352         else:
 353             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 354             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 355
 356         for action in actions:
 357             try:
 358                 MetadataParserPP.validate_action(*action)
 359             except Exception as err:
 360                 raise ValueError(f'{cmd} is invalid; {err}')
 361             yield action
 362
 363     parse_metadata = opts.parse_metadata or []
 364     if opts.metafromtitle is not None:
 365         parse_metadata.append('title:%s' % opts.metafromtitle)
 366     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 367
 368     # Other options
 369     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 370     if geo_bypass_code is not None:
 371         try:
 372             GeoUtils.random_ipv4(geo_bypass_code)
 373         except Exception:
 374             raise ValueError('unsupported geo-bypass country or ip-block')
 375
 376     opts.match_filter = match_filter_func(opts.match_filter)
 377
 378     if opts.download_archive is not None:
 379         opts.download_archive = expand_path(opts.download_archive)
 380
 381     if opts.user_agent is not None:
 382         opts.headers.setdefault('User-Agent', opts.user_agent)
 383     if opts.referer is not None:
 384         opts.headers.setdefault('Referer', opts.referer)
 385
 386     if opts.no_sponsorblock:
 387         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 388
 389     warnings, deprecation_warnings = [], []
 390
 391     # Common mistake: -f best
 392     if opts.format == 'best':
 393         warnings.append('.\n         '.join((
 394             '"-f best" selects the best pre-merged format which is often not the best option',
 395             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 396             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 397
 398     # --(postprocessor/downloader)-args without name
 399     def report_args_compat(name, value, key1, key2=None):
 400         if key1 in value and key2 not in value:
 401             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 402             return True
 403         return False
 404
 405     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 406     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 407         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 408         opts.postprocessor_args.setdefault('sponskrub', [])
 409
 410     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 411                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 412         if val2 is NO_DEFAULT:
 413             val2 = getattr(opts, opt2)
 414         if not val2:
 415             return
 416
 417         if val1 is NO_DEFAULT:
 418             val1 = getattr(opts, opt1)
 419         if val1:
 420             warnings.append(f'{arg1} is ignored since {arg2} was given')
 421         setattr(opts, opt1, default)
 422
 423     # Conflicting options
 424     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 425     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 426     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 427                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 428     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 429     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 430     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 431     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 432     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 433     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 434                     val1=opts.sponskrub and opts.sponskrub_cut)
 435
 436     # Conflicts with --allow-unplayable-formats
 437     report_conflict('--add-metadata', 'addmetadata')
 438     report_conflict('--embed-chapters', 'addchapters')
 439     report_conflict('--embed-info-json', 'embed_infojson')
 440     report_conflict('--embed-subs', 'embedsubtitles')
 441     report_conflict('--embed-thumbnail', 'embedthumbnail')
 442     report_conflict('--extract-audio', 'extractaudio')
 443     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 444     report_conflict('--recode-video', 'recodevideo')
 445     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 446     report_conflict('--remux-video', 'remuxvideo')
 447     report_conflict('--sponskrub', 'sponskrub')
 448     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 449     report_conflict('--xattrs', 'xattrs')
 450
 451     # Fully deprecated options
 452     def report_deprecation(val, old, new=None):
 453         if not val:
 454             return
 455         deprecation_warnings.append(
 456             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 457             else f'{old} is deprecated and may not work as expected')
 458
 459     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 460     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 461     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 462     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 463     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 464
 465     # Dependent options
 466     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 467
 468     if opts.exec_before_dl_cmd:
 469         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 470
 471     if opts.useid:  # --id is not deprecated in youtube-dl
 472         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 473
 474     if opts.overwrites:  # --force-overwrites implies --no-continue
 475         opts.continue_dl = False
 476
 477     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 478         # Add chapters when adding metadata or marking sponsors
 479         opts.addchapters = True
 480
 481     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 482         # Do not unnecessarily download audio
 483         opts.format = 'bestaudio/best'
 484
 485     if opts.getcomments and opts.writeinfojson is None:
 486         # If JSON is not printed anywhere, but comments are requested, save it to file
 487         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 488             opts.writeinfojson = True
 489
 490     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 491         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 492         opts.writesubtitles = True
 493
 494     if opts.addmetadata and opts.embed_infojson is None:
 495         # If embedding metadata and infojson is present, embed it
 496         opts.embed_infojson = 'if_exists'
 497
 498     # Ask for passwords
 499     if opts.username is not None and opts.password is None:
 500         opts.password = compat_getpass('Type account password and press [Return]: ')
 501     if opts.ap_username is not None and opts.ap_password is None:
 502         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 503
 504     return warnings, deprecation_warnings
 505
 506
 507 def get_postprocessors(opts):
 508     yield from opts.add_postprocessors
 509
 510     if opts.parse_metadata:
 511         yield {
 512             'key': 'MetadataParser',
 513             'actions': opts.parse_metadata,
 514             'when': 'pre_process'
 515         }
 516     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 517     if sponsorblock_query:
 518         yield {
 519             'key': 'SponsorBlock',
 520             'categories': sponsorblock_query,
 521             'api': opts.sponsorblock_api,
 522             'when': 'after_filter'
 523         }
 524     if opts.convertsubtitles:
 525         yield {
 526             'key': 'FFmpegSubtitlesConvertor',
 527             'format': opts.convertsubtitles,
 528             'when': 'before_dl'
 529         }
 530     if opts.convertthumbnails:
 531         yield {
 532             'key': 'FFmpegThumbnailsConvertor',
 533             'format': opts.convertthumbnails,
 534             'when': 'before_dl'
 535         }
 536     if opts.extractaudio:
 537         yield {
 538             'key': 'FFmpegExtractAudio',
 539             'preferredcodec': opts.audioformat,
 540             'preferredquality': opts.audioquality,
 541             'nopostoverwrites': opts.nopostoverwrites,
 542         }
 543     if opts.remuxvideo:
 544         yield {
 545             'key': 'FFmpegVideoRemuxer',
 546             'preferedformat': opts.remuxvideo,
 547         }
 548     if opts.recodevideo:
 549         yield {
 550             'key': 'FFmpegVideoConvertor',
 551             'preferedformat': opts.recodevideo,
 552         }
 553     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 554     if opts.embedsubtitles:
 555         keep_subs = 'no-keep-subs' not in opts.compat_opts
 556         yield {
 557             'key': 'FFmpegEmbedSubtitle',
 558             # already_have_subtitle = True prevents the file from being deleted after embedding
 559             'already_have_subtitle': opts.writesubtitles and keep_subs
 560         }
 561         if not opts.writeautomaticsub and keep_subs:
 562             opts.writesubtitles = True
 563
 564     # ModifyChapters must run before FFmpegMetadataPP
 565     if opts.remove_chapters or sponsorblock_query:
 566         yield {
 567             'key': 'ModifyChapters',
 568             'remove_chapters_patterns': opts.remove_chapters,
 569             'remove_sponsor_segments': opts.sponsorblock_remove,
 570             'remove_ranges': opts.remove_ranges,
 571             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 572             'force_keyframes': opts.force_keyframes_at_cuts
 573         }
 574     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 575     # FFmpegExtractAudioPP as containers before conversion may not support
 576     # metadata (3gp, webm, etc.)
 577     # By default ffmpeg preserves metadata applicable for both
 578     # source and target containers. From this point the container won't change,
 579     # so metadata can be added here.
 580     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 581         yield {
 582             'key': 'FFmpegMetadata',
 583             'add_chapters': opts.addchapters,
 584             'add_metadata': opts.addmetadata,
 585             'add_infojson': opts.embed_infojson,
 586         }
 587     # Deprecated
 588     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 589     # but must be below EmbedSubtitle and FFmpegMetadata
 590     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 591     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 592     if opts.sponskrub is not False:
 593         yield {
 594             'key': 'SponSkrub',
 595             'path': opts.sponskrub_path,
 596             'args': opts.sponskrub_args,
 597             'cut': opts.sponskrub_cut,
 598             'force': opts.sponskrub_force,
 599             'ignoreerror': opts.sponskrub is None,
 600             '_from_cli': True,
 601         }
 602     if opts.embedthumbnail:
 603         yield {
 604             'key': 'EmbedThumbnail',
 605             # already_have_thumbnail = True prevents the file from being deleted after embedding
 606             'already_have_thumbnail': opts.writethumbnail
 607         }
 608         if not opts.writethumbnail:
 609             opts.writethumbnail = True
 610             opts.outtmpl['pl_thumbnail'] = ''
 611     if opts.split_chapters:
 612         yield {
 613             'key': 'FFmpegSplitChapters',
 614             'force_keyframes': opts.force_keyframes_at_cuts,
 615         }
 616     # XAttrMetadataPP should be run after post-processors that may change file contents
 617     if opts.xattrs:
 618         yield {'key': 'XAttrMetadata'}
 619     if opts.concat_playlist != 'never':
 620         yield {
 621             'key': 'FFmpegConcat',
 622             'only_multi_video': opts.concat_playlist != 'always',
 623             'when': 'playlist',
 624         }
 625     # Exec must be the last PP of each category
 626     for when, exec_cmd in opts.exec_cmd.items():
 627         yield {
 628             'key': 'Exec',
 629             'exec_cmd': exec_cmd,
 630             'when': when,
 631         }
 632
 633
 634 def parse_options(argv=None):
 635     """ @returns (parser, opts, urls, ydl_opts) """
 636     parser, opts, urls = parseOpts(argv)
 637     urls = get_urls(urls, opts.batchfile, opts.verbose)
 638
 639     set_compat_opts(opts)
 640     try:
 641         warnings, deprecation_warnings = validate_options(opts)
 642     except ValueError as err:
 643         parser.error(f'{err}\n')
 644
 645     postprocessors = list(get_postprocessors(opts))
 646
 647     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[2:])
 648     any_getting = any(getattr(opts, k) for k in (
 649         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 650         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 651     ))
 652
 653     final_ext = (
 654         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 655         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 656         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 657         else None)
 658
 659     return parser, opts, urls, {
 660         'usenetrc': opts.usenetrc,
 661         'netrc_location': opts.netrc_location,
 662         'username': opts.username,
 663         'password': opts.password,
 664         'twofactor': opts.twofactor,
 665         'videopassword': opts.videopassword,
 666         'ap_mso': opts.ap_mso,
 667         'ap_username': opts.ap_username,
 668         'ap_password': opts.ap_password,
 669         'client_certificate': opts.client_certificate,
 670         'client_certificate_key': opts.client_certificate_key,
 671         'client_certificate_password': opts.client_certificate_password,
 672         'quiet': opts.quiet or any_getting or opts.print_json or bool(opts.forceprint),
 673         'no_warnings': opts.no_warnings,
 674         'forceurl': opts.geturl,
 675         'forcetitle': opts.gettitle,
 676         'forceid': opts.getid,
 677         'forcethumbnail': opts.getthumbnail,
 678         'forcedescription': opts.getdescription,
 679         'forceduration': opts.getduration,
 680         'forcefilename': opts.getfilename,
 681         'forceformat': opts.getformat,
 682         'forceprint': opts.forceprint,
 683         'print_to_file': opts.print_to_file,
 684         'forcejson': opts.dumpjson or opts.print_json,
 685         'dump_single_json': opts.dump_single_json,
 686         'force_write_download_archive': opts.force_write_download_archive,
 687         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 688         'skip_download': opts.skip_download,
 689         'format': opts.format,
 690         'allow_unplayable_formats': opts.allow_unplayable_formats,
 691         'ignore_no_formats_error': opts.ignore_no_formats_error,
 692         'format_sort': opts.format_sort,
 693         'format_sort_force': opts.format_sort_force,
 694         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 695         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 696         'check_formats': opts.check_formats,
 697         'listformats': opts.listformats,
 698         'listformats_table': opts.listformats_table,
 699         'outtmpl': opts.outtmpl,
 700         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 701         'paths': opts.paths,
 702         'autonumber_size': opts.autonumber_size,
 703         'autonumber_start': opts.autonumber_start,
 704         'restrictfilenames': opts.restrictfilenames,
 705         'windowsfilenames': opts.windowsfilenames,
 706         'ignoreerrors': opts.ignoreerrors,
 707         'force_generic_extractor': opts.force_generic_extractor,
 708         'ratelimit': opts.ratelimit,
 709         'throttledratelimit': opts.throttledratelimit,
 710         'overwrites': opts.overwrites,
 711         'retries': opts.retries,
 712         'file_access_retries': opts.file_access_retries,
 713         'fragment_retries': opts.fragment_retries,
 714         'extractor_retries': opts.extractor_retries,
 715         'retry_sleep_functions': opts.retry_sleep,
 716         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 717         'keep_fragments': opts.keep_fragments,
 718         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 719         'buffersize': opts.buffersize,
 720         'noresizebuffer': opts.noresizebuffer,
 721         'http_chunk_size': opts.http_chunk_size,
 722         'continuedl': opts.continue_dl,
 723         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 724         'progress_with_newline': opts.progress_with_newline,
 725         'progress_template': opts.progress_template,
 726         'playliststart': opts.playliststart,
 727         'playlistend': opts.playlistend,
 728         'playlistreverse': opts.playlist_reverse,
 729         'playlistrandom': opts.playlist_random,
 730         'noplaylist': opts.noplaylist,
 731         'logtostderr': opts.outtmpl.get('default') == '-',
 732         'consoletitle': opts.consoletitle,
 733         'nopart': opts.nopart,
 734         'updatetime': opts.updatetime,
 735         'writedescription': opts.writedescription,
 736         'writeannotations': opts.writeannotations,
 737         'writeinfojson': opts.writeinfojson,
 738         'allow_playlist_files': opts.allow_playlist_files,
 739         'clean_infojson': opts.clean_infojson,
 740         'getcomments': opts.getcomments,
 741         'writethumbnail': opts.writethumbnail is True,
 742         'write_all_thumbnails': opts.writethumbnail == 'all',
 743         'writelink': opts.writelink,
 744         'writeurllink': opts.writeurllink,
 745         'writewebloclink': opts.writewebloclink,
 746         'writedesktoplink': opts.writedesktoplink,
 747         'writesubtitles': opts.writesubtitles,
 748         'writeautomaticsub': opts.writeautomaticsub,
 749         'allsubtitles': opts.allsubtitles,
 750         'listsubtitles': opts.listsubtitles,
 751         'subtitlesformat': opts.subtitlesformat,
 752         'subtitleslangs': opts.subtitleslangs,
 753         'matchtitle': decodeOption(opts.matchtitle),
 754         'rejecttitle': decodeOption(opts.rejecttitle),
 755         'max_downloads': opts.max_downloads,
 756         'prefer_free_formats': opts.prefer_free_formats,
 757         'trim_file_name': opts.trim_file_name,
 758         'verbose': opts.verbose,
 759         'dump_intermediate_pages': opts.dump_intermediate_pages,
 760         'write_pages': opts.write_pages,
 761         'test': opts.test,
 762         'keepvideo': opts.keepvideo,
 763         'min_filesize': opts.min_filesize,
 764         'max_filesize': opts.max_filesize,
 765         'min_views': opts.min_views,
 766         'max_views': opts.max_views,
 767         'daterange': opts.date,
 768         'cachedir': opts.cachedir,
 769         'youtube_print_sig_code': opts.youtube_print_sig_code,
 770         'age_limit': opts.age_limit,
 771         'download_archive': opts.download_archive,
 772         'break_on_existing': opts.break_on_existing,
 773         'break_on_reject': opts.break_on_reject,
 774         'break_per_url': opts.break_per_url,
 775         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 776         'cookiefile': opts.cookiefile,
 777         'cookiesfrombrowser': opts.cookiesfrombrowser,
 778         'legacyserverconnect': opts.legacy_server_connect,
 779         'nocheckcertificate': opts.no_check_certificate,
 780         'prefer_insecure': opts.prefer_insecure,
 781         'http_headers': opts.headers,
 782         'proxy': opts.proxy,
 783         'socket_timeout': opts.socket_timeout,
 784         'bidi_workaround': opts.bidi_workaround,
 785         'debug_printtraffic': opts.debug_printtraffic,
 786         'prefer_ffmpeg': opts.prefer_ffmpeg,
 787         'include_ads': opts.include_ads,
 788         'default_search': opts.default_search,
 789         'dynamic_mpd': opts.dynamic_mpd,
 790         'extractor_args': opts.extractor_args,
 791         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 792         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 793         'encoding': opts.encoding,
 794         'extract_flat': opts.extract_flat,
 795         'live_from_start': opts.live_from_start,
 796         'wait_for_video': opts.wait_for_video,
 797         'mark_watched': opts.mark_watched,
 798         'merge_output_format': opts.merge_output_format,
 799         'final_ext': final_ext,
 800         'postprocessors': postprocessors,
 801         'fixup': opts.fixup,
 802         'source_address': opts.source_address,
 803         'call_home': opts.call_home,
 804         'sleep_interval_requests': opts.sleep_interval_requests,
 805         'sleep_interval': opts.sleep_interval,
 806         'max_sleep_interval': opts.max_sleep_interval,
 807         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 808         'external_downloader': opts.external_downloader,
 809         'list_thumbnails': opts.list_thumbnails,
 810         'playlist_items': opts.playlist_items,
 811         'xattr_set_filesize': opts.xattr_set_filesize,
 812         'match_filter': opts.match_filter,
 813         'no_color': opts.no_color,
 814         'ffmpeg_location': opts.ffmpeg_location,
 815         'hls_prefer_native': opts.hls_prefer_native,
 816         'hls_use_mpegts': opts.hls_use_mpegts,
 817         'hls_split_discontinuity': opts.hls_split_discontinuity,
 818         'external_downloader_args': opts.external_downloader_args,
 819         'postprocessor_args': opts.postprocessor_args,
 820         'cn_verification_proxy': opts.cn_verification_proxy,
 821         'geo_verification_proxy': opts.geo_verification_proxy,
 822         'geo_bypass': opts.geo_bypass,
 823         'geo_bypass_country': opts.geo_bypass_country,
 824         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 825         '_warnings': warnings,
 826         '_deprecation_warnings': deprecation_warnings,
 827         'compat_opts': opts.compat_opts,
 828     }
 829
 830
 831 def _real_main(argv=None):
 832     setproctitle('yt-dlp')
 833
 834     parser, opts, all_urls, ydl_opts = parse_options(argv)
 835
 836     # Dump user agent
 837     if opts.dump_user_agent:
 838         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 839         write_string(f'{ua}\n', out=sys.stdout)
 840         return
 841
 842     if print_extractor_information(opts, all_urls):
 843         return
 844
 845     with YoutubeDL(ydl_opts) as ydl:
 846         actual_use = all_urls or opts.load_info_filename
 847
 848         if opts.rm_cachedir:
 849             ydl.cache.remove()
 850
 851         if opts.update_self and run_update(ydl) and actual_use:
 852             # If updater returns True, exit. Required for windows
 853             return 100, 'ERROR: The program must exit for the update to complete'
 854
 855         if not actual_use:
 856             if opts.update_self or opts.rm_cachedir:
 857                 return ydl._download_retcode
 858
 859             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 860             parser.error(
 861                 'You must provide at least one URL.\n'
 862                 'Type yt-dlp --help to see a list of all options.')
 863
 864         parser.destroy()
 865         try:
 866             if opts.load_info_filename is not None:
 867                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 868             else:
 869                 return ydl.download(all_urls)
 870         except DownloadCancelled:
 871             ydl.to_screen('Aborting remaining downloads')
 872             return 101
 873
 874
 875 def main(argv=None):
 876     try:
 877         _exit(*variadic(_real_main(argv)))
 878     except DownloadError:
 879         _exit(1)
 880     except SameFileError as e:
 881         _exit(f'ERROR: {e}')
 882     except KeyboardInterrupt:
 883         _exit('\nERROR: Interrupted by user')
 884     except BrokenPipeError as e:
 885         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 886         devnull = os.open(os.devnull, os.O_WRONLY)
 887         os.dup2(devnull, sys.stdout.fileno())
 888         _exit(f'\nERROR: {e}')
 889     except optparse.OptParseError as e:
 890         _exit(2, f'\n{e}')
 891
 892
 893 from .extractor import gen_extractors, list_extractors
 894
 895 __all__ = [
 896     'main',
 897     'YoutubeDL',
 898     'parse_options',
 899     'gen_extractors',
 900     'list_extractors',
 901 ]