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