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