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