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