yt_dlp/__init__.py

   1 f'You are using an unsupported version of Python. Only Python versions 3.7 and above are supported by yt-dlp'  # noqa: F541
   2
   3 __license__ = 'Public Domain'
   4
   5 import collections
   6 import getpass
   7 import itertools
   8 import optparse
   9 import os
  10 import re
  11 import sys
  12
  13 from .compat import compat_shlex_quote
  14 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  15 from .downloader import FileDownloader
  16 from .downloader.external import get_external_downloader
  17 from .extractor import list_extractor_classes
  18 from .extractor.adobepass import MSO_INFO
  19 from .extractor.common import InfoExtractor
  20 from .options import parseOpts
  21 from .postprocessor import (
  22     FFmpegExtractAudioPP,
  23     FFmpegMergerPP,
  24     FFmpegPostProcessor,
  25     FFmpegSubtitlesConvertorPP,
  26     FFmpegThumbnailsConvertorPP,
  27     FFmpegVideoConvertorPP,
  28     FFmpegVideoRemuxerPP,
  29     MetadataFromFieldPP,
  30     MetadataParserPP,
  31 )
  32 from .update import Updater
  33 from .utils import (
  34     NO_DEFAULT,
  35     POSTPROCESS_WHEN,
  36     DateRange,
  37     DownloadCancelled,
  38     DownloadError,
  39     GeoUtils,
  40     PlaylistEntries,
  41     SameFileError,
  42     decodeOption,
  43     download_range_func,
  44     expand_path,
  45     float_or_none,
  46     format_field,
  47     int_or_none,
  48     match_filter_func,
  49     parse_duration,
  50     preferredencoding,
  51     read_batch_urls,
  52     read_stdin,
  53     render_table,
  54     setproctitle,
  55     std_headers,
  56     traverse_obj,
  57     variadic,
  58     write_string,
  59 )
  60 from .YoutubeDL import YoutubeDL
  61
  62
  63 def _exit(status=0, *args):
  64     for msg in args:
  65         sys.stderr.write(msg)
  66     raise SystemExit(status)
  67
  68
  69 def get_urls(urls, batchfile, verbose):
  70     # Batch file verification
  71     batch_urls = []
  72     if batchfile is not None:
  73         try:
  74             batch_urls = read_batch_urls(
  75                 read_stdin('URLs') if batchfile == '-'
  76                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  77             if verbose:
  78                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  79         except OSError:
  80             _exit(f'ERROR: batch file {batchfile} could not be read')
  81     _enc = preferredencoding()
  82     return [
  83         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  84         for url in batch_urls + urls]
  85
  86
  87 def print_extractor_information(opts, urls):
  88     # Importing GenericIE is currently slow since it imports other extractors
  89     # TODO: Move this back to module level after generalization of embed detection
  90     from .extractor.generic import GenericIE
  91
  92     out = ''
  93     if opts.list_extractors:
  94         urls = dict.fromkeys(urls, False)
  95         for ie in list_extractor_classes(opts.age_limit):
  96             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
  97             if ie == GenericIE:
  98                 matched_urls = [url for url, matched in urls.items() if not matched]
  99             else:
 100                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
 101                 urls.update(dict.fromkeys(matched_urls, True))
 102             out += ''.join(f'  {url}\n' for url in matched_urls)
 103     elif opts.list_extractor_descriptions:
 104         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 105         out = '\n'.join(
 106             ie.description(markdown=False, search_examples=_SEARCHES)
 107             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
 108     elif opts.ap_list_mso:
 109         out = 'Supported TV Providers:\n%s\n' % render_table(
 110             ['mso', 'mso name'],
 111             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 112     else:
 113         return False
 114     write_string(out, out=sys.stdout)
 115     return True
 116
 117
 118 def set_compat_opts(opts):
 119     def _unused_compat_opt(name):
 120         if name not in opts.compat_opts:
 121             return False
 122         opts.compat_opts.discard(name)
 123         opts.compat_opts.update(['*%s' % name])
 124         return True
 125
 126     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 127         attr = getattr(opts, opt_name)
 128         if compat_name in opts.compat_opts:
 129             if attr is None:
 130                 setattr(opts, opt_name, not default)
 131                 return True
 132             else:
 133                 if remove_compat:
 134                     _unused_compat_opt(compat_name)
 135                 return False
 136         elif attr is None:
 137             setattr(opts, opt_name, default)
 138         return None
 139
 140     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 141     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 142     set_default_compat('no-clean-infojson', 'clean_infojson')
 143     if 'no-attach-info-json' in opts.compat_opts:
 144         if opts.embed_infojson:
 145             _unused_compat_opt('no-attach-info-json')
 146         else:
 147             opts.embed_infojson = False
 148     if 'format-sort' in opts.compat_opts:
 149         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 150     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 151     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 152     if _video_multistreams_set is False and _audio_multistreams_set is False:
 153         _unused_compat_opt('multistreams')
 154     if 'filename' in opts.compat_opts:
 155         if opts.outtmpl.get('default') is None:
 156             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 157         else:
 158             _unused_compat_opt('filename')
 159
 160
 161 def validate_options(opts):
 162     def validate(cndn, name, value=None, msg=None):
 163         if cndn:
 164             return True
 165         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 166
 167     def validate_in(name, value, items, msg=None):
 168         return validate(value is None or value in items, name, value, msg)
 169
 170     def validate_regex(name, value, regex):
 171         return validate(value is None or re.match(regex, value), name, value)
 172
 173     def validate_positive(name, value, strict=False):
 174         return validate(value is None or value > 0 or (not strict and value == 0),
 175                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 176
 177     def validate_minmax(min_val, max_val, min_name, max_name=None):
 178         if max_val is None or min_val is None or max_val >= min_val:
 179             return
 180         if not max_name:
 181             min_name, max_name = f'min {min_name}', f'max {min_name}'
 182         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 183
 184     # Usernames and passwords
 185     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 186              '.netrc', msg='using {name} conflicts with giving username/password')
 187     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 188     validate(opts.ap_password is None or opts.ap_username is not None,
 189              'TV Provider account username', msg='{name} missing')
 190     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 191                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 192
 193     # Numbers
 194     validate_positive('autonumber start', opts.autonumber_start)
 195     validate_positive('autonumber size', opts.autonumber_size, True)
 196     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 197     validate_positive('playlist start', opts.playliststart, True)
 198     if opts.playlistend != -1:
 199         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 200
 201     # Time ranges
 202     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 203     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 204     validate_positive('sleep interval', opts.sleep_interval)
 205     validate_positive('max sleep interval', opts.max_sleep_interval)
 206     if opts.sleep_interval is None:
 207         validate(
 208             opts.max_sleep_interval is None, 'min sleep interval',
 209             msg='{name} must be specified; use --min-sleep-interval')
 210     elif opts.max_sleep_interval is None:
 211         opts.max_sleep_interval = opts.sleep_interval
 212     else:
 213         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 214
 215     if opts.wait_for_video is not None:
 216         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 217         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 218                  'time range to wait for video', opts.wait_for_video)
 219         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 220         opts.wait_for_video = (min_wait, max_wait)
 221
 222     # Format sort
 223     for f in opts.format_sort:
 224         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 225
 226     # Postprocessor formats
 227     validate_in('merge output format', opts.merge_output_format, FFmpegMergerPP.SUPPORTED_EXTS)
 228     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 229     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 230     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 231     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 232     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 233     if opts.audioquality:
 234         opts.audioquality = opts.audioquality.strip('k').strip('K')
 235         # int_or_none prevents inf, nan
 236         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 237
 238     # Retries
 239     def parse_retries(name, value):
 240         if value is None:
 241             return None
 242         elif value in ('inf', 'infinite'):
 243             return float('inf')
 244         try:
 245             return int(value)
 246         except (TypeError, ValueError):
 247             validate(False, f'{name} retry count', value)
 248
 249     opts.retries = parse_retries('download', opts.retries)
 250     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 251     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 252     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 253
 254     # Retry sleep function
 255     def parse_sleep_func(expr):
 256         NUMBER_RE = r'\d+(?:\.\d+)?'
 257         op, start, limit, step, *_ = tuple(re.fullmatch(
 258             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 259             expr.strip()).groups()) + (None, None)
 260
 261         if op == 'exp':
 262             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 263         else:
 264             default_step = start if op or limit else 0
 265             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 266
 267     for key, expr in opts.retry_sleep.items():
 268         if not expr:
 269             del opts.retry_sleep[key]
 270             continue
 271         try:
 272             opts.retry_sleep[key] = parse_sleep_func(expr)
 273         except AttributeError:
 274             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 275
 276     # Bytes
 277     def parse_bytes(name, value):
 278         if value is None:
 279             return None
 280         numeric_limit = FileDownloader.parse_bytes(value)
 281         validate(numeric_limit is not None, 'rate limit', value)
 282         return numeric_limit
 283
 284     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 285     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 286     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 287     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 288     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 289     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 290
 291     # Output templates
 292     def validate_outtmpl(tmpl, msg):
 293         err = YoutubeDL.validate_outtmpl(tmpl)
 294         if err:
 295             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 296
 297     for k, tmpl in opts.outtmpl.items():
 298         validate_outtmpl(tmpl, f'{k} output template')
 299     for type_, tmpl_list in opts.forceprint.items():
 300         for tmpl in tmpl_list:
 301             validate_outtmpl(tmpl, f'{type_} print template')
 302     for type_, tmpl_list in opts.print_to_file.items():
 303         for tmpl, file in tmpl_list:
 304             validate_outtmpl(tmpl, f'{type_} print to file template')
 305             validate_outtmpl(file, f'{type_} print to file filename')
 306     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 307     for k, tmpl in opts.progress_template.items():
 308         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 309         validate_outtmpl(tmpl, f'{k} template')
 310
 311     outtmpl_default = opts.outtmpl.get('default')
 312     if outtmpl_default == '':
 313         opts.skip_download = None
 314         del opts.outtmpl['default']
 315     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 316         raise ValueError(
 317             'Cannot download a video and extract audio into the same file! '
 318             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 319
 320     def parse_chapters(name, value):
 321         chapters, ranges = [], []
 322         for regex in value or []:
 323             if regex.startswith('*'):
 324                 for range in regex[1:].split(','):
 325                     dur = tuple(map(parse_duration, range.strip().split('-')))
 326                     if len(dur) == 2 and all(t is not None for t in dur):
 327                         ranges.append(dur)
 328                     else:
 329                         raise ValueError(f'invalid {name} time range "{regex}". Must be of the form *start-end')
 330                 continue
 331             try:
 332                 chapters.append(re.compile(regex))
 333             except re.error as err:
 334                 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 335         return chapters, ranges
 336
 337     opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
 338     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
 339
 340     # Cookies from browser
 341     if opts.cookiesfrombrowser:
 342         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 343         if mobj is None:
 344             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 345         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 346         browser_name = browser_name.lower()
 347         if browser_name not in SUPPORTED_BROWSERS:
 348             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 349                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 350         if keyring is not None:
 351             keyring = keyring.upper()
 352             if keyring not in SUPPORTED_KEYRINGS:
 353                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 354                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 355         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 356
 357     # MetadataParser
 358     def metadataparser_actions(f):
 359         if isinstance(f, str):
 360             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 361             try:
 362                 actions = [MetadataFromFieldPP.to_action(f)]
 363             except Exception as err:
 364                 raise ValueError(f'{cmd} is invalid; {err}')
 365         else:
 366             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 367             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 368
 369         for action in actions:
 370             try:
 371                 MetadataParserPP.validate_action(*action)
 372             except Exception as err:
 373                 raise ValueError(f'{cmd} is invalid; {err}')
 374             yield action
 375
 376     parse_metadata = opts.parse_metadata or []
 377     if opts.metafromtitle is not None:
 378         parse_metadata.append('title:%s' % opts.metafromtitle)
 379     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 380
 381     # Other options
 382     if opts.playlist_items is not None:
 383         try:
 384             tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
 385         except Exception as err:
 386             raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
 387
 388     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 389     if geo_bypass_code is not None:
 390         try:
 391             GeoUtils.random_ipv4(geo_bypass_code)
 392         except Exception:
 393             raise ValueError('unsupported geo-bypass country or ip-block')
 394
 395     opts.match_filter = match_filter_func(opts.match_filter)
 396
 397     if opts.download_archive is not None:
 398         opts.download_archive = expand_path(opts.download_archive)
 399
 400     if opts.user_agent is not None:
 401         opts.headers.setdefault('User-Agent', opts.user_agent)
 402     if opts.referer is not None:
 403         opts.headers.setdefault('Referer', opts.referer)
 404
 405     if opts.no_sponsorblock:
 406         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 407
 408     default_downloader = None
 409     for proto, path in opts.external_downloader.items():
 410         if path == 'native':
 411             continue
 412         ed = get_external_downloader(path)
 413         if ed is None:
 414             raise ValueError(
 415                 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
 416         elif ed and proto == 'default':
 417             default_downloader = ed.get_basename()
 418
 419     warnings, deprecation_warnings = [], []
 420
 421     # Common mistake: -f best
 422     if opts.format == 'best':
 423         warnings.append('.\n         '.join((
 424             '"-f best" selects the best pre-merged format which is often not the best option',
 425             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 426             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 427
 428     # --(postprocessor/downloader)-args without name
 429     def report_args_compat(name, value, key1, key2=None, where=None):
 430         if key1 in value and key2 not in value:
 431             warnings.append(f'{name.title()} arguments given without specifying name. '
 432                             f'The arguments will be given to {where or f"all {name}s"}')
 433             return True
 434         return False
 435
 436     if report_args_compat('external downloader', opts.external_downloader_args,
 437                           'default', where=default_downloader) and default_downloader:
 438         # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
 439         opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
 440
 441     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 442         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 443         opts.postprocessor_args.setdefault('sponskrub', [])
 444
 445     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 446                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 447         if val2 is NO_DEFAULT:
 448             val2 = getattr(opts, opt2)
 449         if not val2:
 450             return
 451
 452         if val1 is NO_DEFAULT:
 453             val1 = getattr(opts, opt1)
 454         if val1:
 455             warnings.append(f'{arg1} is ignored since {arg2} was given')
 456         setattr(opts, opt1, default)
 457
 458     # Conflicting options
 459     report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
 460     report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
 461     report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
 462     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 463     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 464     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 465                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 466     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 467     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 468     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 469     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 470     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 471     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 472                     val1=opts.sponskrub and opts.sponskrub_cut)
 473
 474     # Conflicts with --allow-unplayable-formats
 475     report_conflict('--add-metadata', 'addmetadata')
 476     report_conflict('--embed-chapters', 'addchapters')
 477     report_conflict('--embed-info-json', 'embed_infojson')
 478     report_conflict('--embed-subs', 'embedsubtitles')
 479     report_conflict('--embed-thumbnail', 'embedthumbnail')
 480     report_conflict('--extract-audio', 'extractaudio')
 481     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 482     report_conflict('--recode-video', 'recodevideo')
 483     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 484     report_conflict('--remux-video', 'remuxvideo')
 485     report_conflict('--sponskrub', 'sponskrub')
 486     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 487     report_conflict('--xattrs', 'xattrs')
 488
 489     # Fully deprecated options
 490     def report_deprecation(val, old, new=None):
 491         if not val:
 492             return
 493         deprecation_warnings.append(
 494             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 495             else f'{old} is deprecated and may not work as expected')
 496
 497     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 498     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 499     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 500     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 501     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 502
 503     # Dependent options
 504     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 505
 506     if opts.exec_before_dl_cmd:
 507         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 508
 509     if opts.useid:  # --id is not deprecated in youtube-dl
 510         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 511
 512     if opts.overwrites:  # --force-overwrites implies --no-continue
 513         opts.continue_dl = False
 514
 515     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 516         # Add chapters when adding metadata or marking sponsors
 517         opts.addchapters = True
 518
 519     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 520         # Do not unnecessarily download audio
 521         opts.format = 'bestaudio/best'
 522
 523     if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
 524         # If JSON is not printed anywhere, but comments are requested, save it to file
 525         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 526             opts.writeinfojson = True
 527
 528     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 529         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 530         opts.writesubtitles = True
 531
 532     if opts.addmetadata and opts.embed_infojson is None:
 533         # If embedding metadata and infojson is present, embed it
 534         opts.embed_infojson = 'if_exists'
 535
 536     # Ask for passwords
 537     if opts.username is not None and opts.password is None:
 538         opts.password = getpass.getpass('Type account password and press [Return]: ')
 539     if opts.ap_username is not None and opts.ap_password is None:
 540         opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 541
 542     return warnings, deprecation_warnings
 543
 544
 545 def get_postprocessors(opts):
 546     yield from opts.add_postprocessors
 547
 548     if opts.parse_metadata:
 549         yield {
 550             'key': 'MetadataParser',
 551             'actions': opts.parse_metadata,
 552             'when': 'pre_process'
 553         }
 554     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 555     if sponsorblock_query:
 556         yield {
 557             'key': 'SponsorBlock',
 558             'categories': sponsorblock_query,
 559             'api': opts.sponsorblock_api,
 560             'when': 'after_filter'
 561         }
 562     if opts.convertsubtitles:
 563         yield {
 564             'key': 'FFmpegSubtitlesConvertor',
 565             'format': opts.convertsubtitles,
 566             'when': 'before_dl'
 567         }
 568     if opts.convertthumbnails:
 569         yield {
 570             'key': 'FFmpegThumbnailsConvertor',
 571             'format': opts.convertthumbnails,
 572             'when': 'before_dl'
 573         }
 574     if opts.extractaudio:
 575         yield {
 576             'key': 'FFmpegExtractAudio',
 577             'preferredcodec': opts.audioformat,
 578             'preferredquality': opts.audioquality,
 579             'nopostoverwrites': opts.nopostoverwrites,
 580         }
 581     if opts.remuxvideo:
 582         yield {
 583             'key': 'FFmpegVideoRemuxer',
 584             'preferedformat': opts.remuxvideo,
 585         }
 586     if opts.recodevideo:
 587         yield {
 588             'key': 'FFmpegVideoConvertor',
 589             'preferedformat': opts.recodevideo,
 590         }
 591     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 592     if opts.embedsubtitles:
 593         keep_subs = 'no-keep-subs' not in opts.compat_opts
 594         yield {
 595             'key': 'FFmpegEmbedSubtitle',
 596             # already_have_subtitle = True prevents the file from being deleted after embedding
 597             'already_have_subtitle': opts.writesubtitles and keep_subs
 598         }
 599         if not opts.writeautomaticsub and keep_subs:
 600             opts.writesubtitles = True
 601
 602     # ModifyChapters must run before FFmpegMetadataPP
 603     if opts.remove_chapters or sponsorblock_query:
 604         yield {
 605             'key': 'ModifyChapters',
 606             'remove_chapters_patterns': opts.remove_chapters,
 607             'remove_sponsor_segments': opts.sponsorblock_remove,
 608             'remove_ranges': opts.remove_ranges,
 609             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 610             'force_keyframes': opts.force_keyframes_at_cuts
 611         }
 612     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 613     # FFmpegExtractAudioPP as containers before conversion may not support
 614     # metadata (3gp, webm, etc.)
 615     # By default ffmpeg preserves metadata applicable for both
 616     # source and target containers. From this point the container won't change,
 617     # so metadata can be added here.
 618     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 619         yield {
 620             'key': 'FFmpegMetadata',
 621             'add_chapters': opts.addchapters,
 622             'add_metadata': opts.addmetadata,
 623             'add_infojson': opts.embed_infojson,
 624         }
 625     # Deprecated
 626     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 627     # but must be below EmbedSubtitle and FFmpegMetadata
 628     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 629     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 630     if opts.sponskrub is not False:
 631         yield {
 632             'key': 'SponSkrub',
 633             'path': opts.sponskrub_path,
 634             'args': opts.sponskrub_args,
 635             'cut': opts.sponskrub_cut,
 636             'force': opts.sponskrub_force,
 637             'ignoreerror': opts.sponskrub is None,
 638             '_from_cli': True,
 639         }
 640     if opts.embedthumbnail:
 641         yield {
 642             'key': 'EmbedThumbnail',
 643             # already_have_thumbnail = True prevents the file from being deleted after embedding
 644             'already_have_thumbnail': opts.writethumbnail
 645         }
 646         if not opts.writethumbnail:
 647             opts.writethumbnail = True
 648             opts.outtmpl['pl_thumbnail'] = ''
 649     if opts.split_chapters:
 650         yield {
 651             'key': 'FFmpegSplitChapters',
 652             'force_keyframes': opts.force_keyframes_at_cuts,
 653         }
 654     # XAttrMetadataPP should be run after post-processors that may change file contents
 655     if opts.xattrs:
 656         yield {'key': 'XAttrMetadata'}
 657     if opts.concat_playlist != 'never':
 658         yield {
 659             'key': 'FFmpegConcat',
 660             'only_multi_video': opts.concat_playlist != 'always',
 661             'when': 'playlist',
 662         }
 663     # Exec must be the last PP of each category
 664     for when, exec_cmd in opts.exec_cmd.items():
 665         yield {
 666             'key': 'Exec',
 667             'exec_cmd': exec_cmd,
 668             'when': when,
 669         }
 670
 671
 672 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
 673
 674
 675 def parse_options(argv=None):
 676     """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
 677     parser, opts, urls = parseOpts(argv)
 678     urls = get_urls(urls, opts.batchfile, opts.verbose)
 679
 680     set_compat_opts(opts)
 681     try:
 682         warnings, deprecation_warnings = validate_options(opts)
 683     except ValueError as err:
 684         parser.error(f'{err}\n')
 685
 686     postprocessors = list(get_postprocessors(opts))
 687
 688     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[2:])
 689     any_getting = any(getattr(opts, k) for k in (
 690         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 691         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 692     ))
 693
 694     playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
 695     write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
 696                                and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
 697     if not any((
 698         opts.extract_flat,
 699         opts.dump_single_json,
 700         opts.forceprint.get('playlist'),
 701         opts.print_to_file.get('playlist'),
 702         write_playlist_infojson,
 703     )):
 704         if not playlist_pps:
 705             opts.extract_flat = 'discard'
 706         elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
 707             opts.extract_flat = 'discard_in_playlist'
 708
 709     final_ext = (
 710         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 711         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 712         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 713         else None)
 714
 715     return ParsedOptions(parser, opts, urls, {
 716         'usenetrc': opts.usenetrc,
 717         'netrc_location': opts.netrc_location,
 718         'username': opts.username,
 719         'password': opts.password,
 720         'twofactor': opts.twofactor,
 721         'videopassword': opts.videopassword,
 722         'ap_mso': opts.ap_mso,
 723         'ap_username': opts.ap_username,
 724         'ap_password': opts.ap_password,
 725         'client_certificate': opts.client_certificate,
 726         'client_certificate_key': opts.client_certificate_key,
 727         'client_certificate_password': opts.client_certificate_password,
 728         'quiet': opts.quiet or any_getting or opts.print_json or bool(opts.forceprint),
 729         'no_warnings': opts.no_warnings,
 730         'forceurl': opts.geturl,
 731         'forcetitle': opts.gettitle,
 732         'forceid': opts.getid,
 733         'forcethumbnail': opts.getthumbnail,
 734         'forcedescription': opts.getdescription,
 735         'forceduration': opts.getduration,
 736         'forcefilename': opts.getfilename,
 737         'forceformat': opts.getformat,
 738         'forceprint': opts.forceprint,
 739         'print_to_file': opts.print_to_file,
 740         'forcejson': opts.dumpjson or opts.print_json,
 741         'dump_single_json': opts.dump_single_json,
 742         'force_write_download_archive': opts.force_write_download_archive,
 743         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 744         'skip_download': opts.skip_download,
 745         'format': opts.format,
 746         'allow_unplayable_formats': opts.allow_unplayable_formats,
 747         'ignore_no_formats_error': opts.ignore_no_formats_error,
 748         'format_sort': opts.format_sort,
 749         'format_sort_force': opts.format_sort_force,
 750         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 751         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 752         'check_formats': opts.check_formats,
 753         'listformats': opts.listformats,
 754         'listformats_table': opts.listformats_table,
 755         'outtmpl': opts.outtmpl,
 756         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 757         'paths': opts.paths,
 758         'autonumber_size': opts.autonumber_size,
 759         'autonumber_start': opts.autonumber_start,
 760         'restrictfilenames': opts.restrictfilenames,
 761         'windowsfilenames': opts.windowsfilenames,
 762         'ignoreerrors': opts.ignoreerrors,
 763         'force_generic_extractor': opts.force_generic_extractor,
 764         'ratelimit': opts.ratelimit,
 765         'throttledratelimit': opts.throttledratelimit,
 766         'overwrites': opts.overwrites,
 767         'retries': opts.retries,
 768         'file_access_retries': opts.file_access_retries,
 769         'fragment_retries': opts.fragment_retries,
 770         'extractor_retries': opts.extractor_retries,
 771         'retry_sleep_functions': opts.retry_sleep,
 772         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 773         'keep_fragments': opts.keep_fragments,
 774         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 775         'buffersize': opts.buffersize,
 776         'noresizebuffer': opts.noresizebuffer,
 777         'http_chunk_size': opts.http_chunk_size,
 778         'continuedl': opts.continue_dl,
 779         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 780         'progress_with_newline': opts.progress_with_newline,
 781         'progress_template': opts.progress_template,
 782         'playliststart': opts.playliststart,
 783         'playlistend': opts.playlistend,
 784         'playlistreverse': opts.playlist_reverse,
 785         'playlistrandom': opts.playlist_random,
 786         'lazy_playlist': opts.lazy_playlist,
 787         'noplaylist': opts.noplaylist,
 788         'logtostderr': opts.outtmpl.get('default') == '-',
 789         'consoletitle': opts.consoletitle,
 790         'nopart': opts.nopart,
 791         'updatetime': opts.updatetime,
 792         'writedescription': opts.writedescription,
 793         'writeannotations': opts.writeannotations,
 794         'writeinfojson': opts.writeinfojson,
 795         'allow_playlist_files': opts.allow_playlist_files,
 796         'clean_infojson': opts.clean_infojson,
 797         'getcomments': opts.getcomments,
 798         'writethumbnail': opts.writethumbnail is True,
 799         'write_all_thumbnails': opts.writethumbnail == 'all',
 800         'writelink': opts.writelink,
 801         'writeurllink': opts.writeurllink,
 802         'writewebloclink': opts.writewebloclink,
 803         'writedesktoplink': opts.writedesktoplink,
 804         'writesubtitles': opts.writesubtitles,
 805         'writeautomaticsub': opts.writeautomaticsub,
 806         'allsubtitles': opts.allsubtitles,
 807         'listsubtitles': opts.listsubtitles,
 808         'subtitlesformat': opts.subtitlesformat,
 809         'subtitleslangs': opts.subtitleslangs,
 810         'matchtitle': decodeOption(opts.matchtitle),
 811         'rejecttitle': decodeOption(opts.rejecttitle),
 812         'max_downloads': opts.max_downloads,
 813         'prefer_free_formats': opts.prefer_free_formats,
 814         'trim_file_name': opts.trim_file_name,
 815         'verbose': opts.verbose,
 816         'dump_intermediate_pages': opts.dump_intermediate_pages,
 817         'write_pages': opts.write_pages,
 818         'load_pages': opts.load_pages,
 819         'test': opts.test,
 820         'keepvideo': opts.keepvideo,
 821         'min_filesize': opts.min_filesize,
 822         'max_filesize': opts.max_filesize,
 823         'min_views': opts.min_views,
 824         'max_views': opts.max_views,
 825         'daterange': opts.date,
 826         'cachedir': opts.cachedir,
 827         'youtube_print_sig_code': opts.youtube_print_sig_code,
 828         'age_limit': opts.age_limit,
 829         'download_archive': opts.download_archive,
 830         'break_on_existing': opts.break_on_existing,
 831         'break_on_reject': opts.break_on_reject,
 832         'break_per_url': opts.break_per_url,
 833         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 834         'cookiefile': opts.cookiefile,
 835         'cookiesfrombrowser': opts.cookiesfrombrowser,
 836         'legacyserverconnect': opts.legacy_server_connect,
 837         'nocheckcertificate': opts.no_check_certificate,
 838         'prefer_insecure': opts.prefer_insecure,
 839         'http_headers': opts.headers,
 840         'proxy': opts.proxy,
 841         'socket_timeout': opts.socket_timeout,
 842         'bidi_workaround': opts.bidi_workaround,
 843         'debug_printtraffic': opts.debug_printtraffic,
 844         'prefer_ffmpeg': opts.prefer_ffmpeg,
 845         'include_ads': opts.include_ads,
 846         'default_search': opts.default_search,
 847         'dynamic_mpd': opts.dynamic_mpd,
 848         'extractor_args': opts.extractor_args,
 849         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 850         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 851         'encoding': opts.encoding,
 852         'extract_flat': opts.extract_flat,
 853         'live_from_start': opts.live_from_start,
 854         'wait_for_video': opts.wait_for_video,
 855         'mark_watched': opts.mark_watched,
 856         'merge_output_format': opts.merge_output_format,
 857         'final_ext': final_ext,
 858         'postprocessors': postprocessors,
 859         'fixup': opts.fixup,
 860         'source_address': opts.source_address,
 861         'call_home': opts.call_home,
 862         'sleep_interval_requests': opts.sleep_interval_requests,
 863         'sleep_interval': opts.sleep_interval,
 864         'max_sleep_interval': opts.max_sleep_interval,
 865         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 866         'external_downloader': opts.external_downloader,
 867         'download_ranges': opts.download_ranges,
 868         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 869         'list_thumbnails': opts.list_thumbnails,
 870         'playlist_items': opts.playlist_items,
 871         'xattr_set_filesize': opts.xattr_set_filesize,
 872         'match_filter': opts.match_filter,
 873         'no_color': opts.no_color,
 874         'ffmpeg_location': opts.ffmpeg_location,
 875         'hls_prefer_native': opts.hls_prefer_native,
 876         'hls_use_mpegts': opts.hls_use_mpegts,
 877         'hls_split_discontinuity': opts.hls_split_discontinuity,
 878         'external_downloader_args': opts.external_downloader_args,
 879         'postprocessor_args': opts.postprocessor_args,
 880         'cn_verification_proxy': opts.cn_verification_proxy,
 881         'geo_verification_proxy': opts.geo_verification_proxy,
 882         'geo_bypass': opts.geo_bypass,
 883         'geo_bypass_country': opts.geo_bypass_country,
 884         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 885         '_warnings': warnings,
 886         '_deprecation_warnings': deprecation_warnings,
 887         'compat_opts': opts.compat_opts,
 888     })
 889
 890
 891 def _real_main(argv=None):
 892     setproctitle('yt-dlp')
 893
 894     parser, opts, all_urls, ydl_opts = parse_options(argv)
 895
 896     # Dump user agent
 897     if opts.dump_user_agent:
 898         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 899         write_string(f'{ua}\n', out=sys.stdout)
 900         return
 901
 902     if print_extractor_information(opts, all_urls):
 903         return
 904
 905     # We may need ffmpeg_location without having access to the YoutubeDL instance
 906     # See https://github.com/yt-dlp/yt-dlp/issues/2191
 907     if opts.ffmpeg_location:
 908         FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
 909
 910     with YoutubeDL(ydl_opts) as ydl:
 911         pre_process = opts.update_self or opts.rm_cachedir
 912         actual_use = all_urls or opts.load_info_filename
 913
 914         if opts.rm_cachedir:
 915             ydl.cache.remove()
 916
 917         updater = Updater(ydl)
 918         if opts.update_self and updater.update() and actual_use:
 919             if updater.cmd:
 920                 return updater.restart()
 921             # This code is reachable only for zip variant in py < 3.10
 922             # It makes sense to exit here, but the old behavior is to continue
 923             ydl.report_warning('Restart yt-dlp to use the updated version')
 924             # return 100, 'ERROR: The program must exit for the update to complete'
 925
 926         if not actual_use:
 927             if pre_process:
 928                 return ydl._download_retcode
 929
 930             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 931             parser.error(
 932                 'You must provide at least one URL.\n'
 933                 'Type yt-dlp --help to see a list of all options.')
 934
 935         parser.destroy()
 936         try:
 937             if opts.load_info_filename is not None:
 938                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 939             else:
 940                 return ydl.download(all_urls)
 941         except DownloadCancelled:
 942             ydl.to_screen('Aborting remaining downloads')
 943             return 101
 944
 945
 946 def main(argv=None):
 947     try:
 948         _exit(*variadic(_real_main(argv)))
 949     except DownloadError:
 950         _exit(1)
 951     except SameFileError as e:
 952         _exit(f'ERROR: {e}')
 953     except KeyboardInterrupt:
 954         _exit('\nERROR: Interrupted by user')
 955     except BrokenPipeError as e:
 956         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 957         devnull = os.open(os.devnull, os.O_WRONLY)
 958         os.dup2(devnull, sys.stdout.fileno())
 959         _exit(f'\nERROR: {e}')
 960     except optparse.OptParseError as e:
 961         _exit(2, f'\n{e}')
 962
 963
 964 from .extractor import gen_extractors, list_extractors
 965
 966 __all__ = [
 967     'main',
 968     'YoutubeDL',
 969     'parse_options',
 970     'gen_extractors',
 971     'list_extractors',
 972 ]