yt_dlp/__init__.py

   1 try:
   2     import contextvars  # noqa: F401
   3 except Exception:
   4     raise Exception(
   5         f'You are using an unsupported version of Python. Only Python versions 3.7 and above are supported by yt-dlp')  # noqa: F541
   6
   7 __license__ = 'Public Domain'
   8
   9 import collections
  10 import getpass
  11 import itertools
  12 import optparse
  13 import os
  14 import re
  15 import sys
  16
  17 from .compat import compat_shlex_quote
  18 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  19 from .downloader import FileDownloader
  20 from .downloader.external import get_external_downloader
  21 from .extractor import list_extractor_classes
  22 from .extractor.adobepass import MSO_INFO
  23 from .extractor.common import InfoExtractor
  24 from .options import parseOpts
  25 from .postprocessor import (
  26     FFmpegExtractAudioPP,
  27     FFmpegMergerPP,
  28     FFmpegPostProcessor,
  29     FFmpegSubtitlesConvertorPP,
  30     FFmpegThumbnailsConvertorPP,
  31     FFmpegVideoConvertorPP,
  32     FFmpegVideoRemuxerPP,
  33     MetadataFromFieldPP,
  34     MetadataParserPP,
  35 )
  36 from .update import Updater
  37 from .utils import (
  38     NO_DEFAULT,
  39     POSTPROCESS_WHEN,
  40     DateRange,
  41     DownloadCancelled,
  42     DownloadError,
  43     GeoUtils,
  44     PlaylistEntries,
  45     SameFileError,
  46     decodeOption,
  47     download_range_func,
  48     expand_path,
  49     float_or_none,
  50     format_field,
  51     int_or_none,
  52     match_filter_func,
  53     parse_duration,
  54     preferredencoding,
  55     read_batch_urls,
  56     read_stdin,
  57     render_table,
  58     setproctitle,
  59     std_headers,
  60     traverse_obj,
  61     variadic,
  62     write_string,
  63 )
  64 from .YoutubeDL import YoutubeDL
  65
  66
  67 def _exit(status=0, *args):
  68     for msg in args:
  69         sys.stderr.write(msg)
  70     raise SystemExit(status)
  71
  72
  73 def get_urls(urls, batchfile, verbose):
  74     # Batch file verification
  75     batch_urls = []
  76     if batchfile is not None:
  77         try:
  78             batch_urls = read_batch_urls(
  79                 read_stdin('URLs') if batchfile == '-'
  80                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  81             if verbose:
  82                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  83         except OSError:
  84             _exit(f'ERROR: batch file {batchfile} could not be read')
  85     _enc = preferredencoding()
  86     return [
  87         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  88         for url in batch_urls + urls]
  89
  90
  91 def print_extractor_information(opts, urls):
  92     # Importing GenericIE is currently slow since it imports other extractors
  93     # TODO: Move this back to module level after generalization of embed detection
  94     from .extractor.generic import GenericIE
  95
  96     out = ''
  97     if opts.list_extractors:
  98         urls = dict.fromkeys(urls, False)
  99         for ie in list_extractor_classes(opts.age_limit):
 100             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
 101             if ie == GenericIE:
 102                 matched_urls = [url for url, matched in urls.items() if not matched]
 103             else:
 104                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
 105                 urls.update(dict.fromkeys(matched_urls, True))
 106             out += ''.join(f'  {url}\n' for url in matched_urls)
 107     elif opts.list_extractor_descriptions:
 108         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 109         out = '\n'.join(
 110             ie.description(markdown=False, search_examples=_SEARCHES)
 111             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
 112     elif opts.ap_list_mso:
 113         out = 'Supported TV Providers:\n%s\n' % render_table(
 114             ['mso', 'mso name'],
 115             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 116     else:
 117         return False
 118     write_string(out, out=sys.stdout)
 119     return True
 120
 121
 122 def set_compat_opts(opts):
 123     def _unused_compat_opt(name):
 124         if name not in opts.compat_opts:
 125             return False
 126         opts.compat_opts.discard(name)
 127         opts.compat_opts.update(['*%s' % name])
 128         return True
 129
 130     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 131         attr = getattr(opts, opt_name)
 132         if compat_name in opts.compat_opts:
 133             if attr is None:
 134                 setattr(opts, opt_name, not default)
 135                 return True
 136             else:
 137                 if remove_compat:
 138                     _unused_compat_opt(compat_name)
 139                 return False
 140         elif attr is None:
 141             setattr(opts, opt_name, default)
 142         return None
 143
 144     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 145     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 146     set_default_compat('no-clean-infojson', 'clean_infojson')
 147     if 'no-attach-info-json' in opts.compat_opts:
 148         if opts.embed_infojson:
 149             _unused_compat_opt('no-attach-info-json')
 150         else:
 151             opts.embed_infojson = False
 152     if 'format-sort' in opts.compat_opts:
 153         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 154     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 155     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 156     if _video_multistreams_set is False and _audio_multistreams_set is False:
 157         _unused_compat_opt('multistreams')
 158     if 'filename' in opts.compat_opts:
 159         if opts.outtmpl.get('default') is None:
 160             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 161         else:
 162             _unused_compat_opt('filename')
 163
 164
 165 def validate_options(opts):
 166     def validate(cndn, name, value=None, msg=None):
 167         if cndn:
 168             return True
 169         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 170
 171     def validate_in(name, value, items, msg=None):
 172         return validate(value is None or value in items, name, value, msg)
 173
 174     def validate_regex(name, value, regex):
 175         return validate(value is None or re.match(regex, value), name, value)
 176
 177     def validate_positive(name, value, strict=False):
 178         return validate(value is None or value > 0 or (not strict and value == 0),
 179                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 180
 181     def validate_minmax(min_val, max_val, min_name, max_name=None):
 182         if max_val is None or min_val is None or max_val >= min_val:
 183             return
 184         if not max_name:
 185             min_name, max_name = f'min {min_name}', f'max {min_name}'
 186         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 187
 188     # Usernames and passwords
 189     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 190              '.netrc', msg='using {name} conflicts with giving username/password')
 191     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 192     validate(opts.ap_password is None or opts.ap_username is not None,
 193              'TV Provider account username', msg='{name} missing')
 194     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 195                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 196
 197     # Numbers
 198     validate_positive('autonumber start', opts.autonumber_start)
 199     validate_positive('autonumber size', opts.autonumber_size, True)
 200     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 201     validate_positive('playlist start', opts.playliststart, True)
 202     if opts.playlistend != -1:
 203         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 204
 205     # Time ranges
 206     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 207     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 208     validate_positive('sleep interval', opts.sleep_interval)
 209     validate_positive('max sleep interval', opts.max_sleep_interval)
 210     if opts.sleep_interval is None:
 211         validate(
 212             opts.max_sleep_interval is None, 'min sleep interval',
 213             msg='{name} must be specified; use --min-sleep-interval')
 214     elif opts.max_sleep_interval is None:
 215         opts.max_sleep_interval = opts.sleep_interval
 216     else:
 217         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 218
 219     if opts.wait_for_video is not None:
 220         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 221         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 222                  'time range to wait for video', opts.wait_for_video)
 223         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 224         opts.wait_for_video = (min_wait, max_wait)
 225
 226     # Format sort
 227     for f in opts.format_sort:
 228         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 229
 230     # Postprocessor formats
 231     validate_in('merge output format', opts.merge_output_format, FFmpegMergerPP.SUPPORTED_EXTS)
 232     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 233     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 234     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 235     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 236     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 237     if opts.audioquality:
 238         opts.audioquality = opts.audioquality.strip('k').strip('K')
 239         # int_or_none prevents inf, nan
 240         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 241
 242     # Retries
 243     def parse_retries(name, value):
 244         if value is None:
 245             return None
 246         elif value in ('inf', 'infinite'):
 247             return float('inf')
 248         try:
 249             return int(value)
 250         except (TypeError, ValueError):
 251             validate(False, f'{name} retry count', value)
 252
 253     opts.retries = parse_retries('download', opts.retries)
 254     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 255     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 256     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 257
 258     # Retry sleep function
 259     def parse_sleep_func(expr):
 260         NUMBER_RE = r'\d+(?:\.\d+)?'
 261         op, start, limit, step, *_ = tuple(re.fullmatch(
 262             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 263             expr.strip()).groups()) + (None, None)
 264
 265         if op == 'exp':
 266             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 267         else:
 268             default_step = start if op or limit else 0
 269             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 270
 271     for key, expr in opts.retry_sleep.items():
 272         if not expr:
 273             del opts.retry_sleep[key]
 274             continue
 275         try:
 276             opts.retry_sleep[key] = parse_sleep_func(expr)
 277         except AttributeError:
 278             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 279
 280     # Bytes
 281     def parse_bytes(name, value):
 282         if value is None:
 283             return None
 284         numeric_limit = FileDownloader.parse_bytes(value)
 285         validate(numeric_limit is not None, 'rate limit', value)
 286         return numeric_limit
 287
 288     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 289     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 290     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 291     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 292     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 293     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 294
 295     # Output templates
 296     def validate_outtmpl(tmpl, msg):
 297         err = YoutubeDL.validate_outtmpl(tmpl)
 298         if err:
 299             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 300
 301     for k, tmpl in opts.outtmpl.items():
 302         validate_outtmpl(tmpl, f'{k} output template')
 303     for type_, tmpl_list in opts.forceprint.items():
 304         for tmpl in tmpl_list:
 305             validate_outtmpl(tmpl, f'{type_} print template')
 306     for type_, tmpl_list in opts.print_to_file.items():
 307         for tmpl, file in tmpl_list:
 308             validate_outtmpl(tmpl, f'{type_} print to file template')
 309             validate_outtmpl(file, f'{type_} print to file filename')
 310     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 311     for k, tmpl in opts.progress_template.items():
 312         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 313         validate_outtmpl(tmpl, f'{k} template')
 314
 315     outtmpl_default = opts.outtmpl.get('default')
 316     if outtmpl_default == '':
 317         opts.skip_download = None
 318         del opts.outtmpl['default']
 319     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 320         raise ValueError(
 321             'Cannot download a video and extract audio into the same file! '
 322             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 323
 324     def parse_chapters(name, value):
 325         chapters, ranges = [], []
 326         for regex in value or []:
 327             if regex.startswith('*'):
 328                 for range in regex[1:].split(','):
 329                     dur = tuple(map(parse_duration, range.strip().split('-')))
 330                     if len(dur) == 2 and all(t is not None for t in dur):
 331                         ranges.append(dur)
 332                     else:
 333                         raise ValueError(f'invalid {name} time range "{regex}". Must be of the form *start-end')
 334                 continue
 335             try:
 336                 chapters.append(re.compile(regex))
 337             except re.error as err:
 338                 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 339         return chapters, ranges
 340
 341     opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
 342     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
 343
 344     # Cookies from browser
 345     if opts.cookiesfrombrowser:
 346         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 347         if mobj is None:
 348             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 349         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 350         browser_name = browser_name.lower()
 351         if browser_name not in SUPPORTED_BROWSERS:
 352             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 353                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 354         if keyring is not None:
 355             keyring = keyring.upper()
 356             if keyring not in SUPPORTED_KEYRINGS:
 357                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 358                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 359         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 360
 361     # MetadataParser
 362     def metadataparser_actions(f):
 363         if isinstance(f, str):
 364             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 365             try:
 366                 actions = [MetadataFromFieldPP.to_action(f)]
 367             except Exception as err:
 368                 raise ValueError(f'{cmd} is invalid; {err}')
 369         else:
 370             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 371             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 372
 373         for action in actions:
 374             try:
 375                 MetadataParserPP.validate_action(*action)
 376             except Exception as err:
 377                 raise ValueError(f'{cmd} is invalid; {err}')
 378             yield action
 379
 380     parse_metadata = opts.parse_metadata or []
 381     if opts.metafromtitle is not None:
 382         parse_metadata.append('title:%s' % opts.metafromtitle)
 383     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 384
 385     # Other options
 386     if opts.playlist_items is not None:
 387         try:
 388             tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
 389         except Exception as err:
 390             raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
 391
 392     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 393     if geo_bypass_code is not None:
 394         try:
 395             GeoUtils.random_ipv4(geo_bypass_code)
 396         except Exception:
 397             raise ValueError('unsupported geo-bypass country or ip-block')
 398
 399     opts.match_filter = match_filter_func(opts.match_filter)
 400
 401     if opts.download_archive is not None:
 402         opts.download_archive = expand_path(opts.download_archive)
 403
 404     if opts.user_agent is not None:
 405         opts.headers.setdefault('User-Agent', opts.user_agent)
 406     if opts.referer is not None:
 407         opts.headers.setdefault('Referer', opts.referer)
 408
 409     if opts.no_sponsorblock:
 410         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 411
 412     default_downloader = None
 413     for proto, path in opts.external_downloader.items():
 414         if path == 'native':
 415             continue
 416         ed = get_external_downloader(path)
 417         if ed is None:
 418             raise ValueError(
 419                 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
 420         elif ed and proto == 'default':
 421             default_downloader = ed.get_basename()
 422
 423     warnings, deprecation_warnings = [], []
 424
 425     # Common mistake: -f best
 426     if opts.format == 'best':
 427         warnings.append('.\n         '.join((
 428             '"-f best" selects the best pre-merged format which is often not the best option',
 429             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 430             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 431
 432     # --(postprocessor/downloader)-args without name
 433     def report_args_compat(name, value, key1, key2=None, where=None):
 434         if key1 in value and key2 not in value:
 435             warnings.append(f'{name.title()} arguments given without specifying name. '
 436                             f'The arguments will be given to {where or f"all {name}s"}')
 437             return True
 438         return False
 439
 440     if report_args_compat('external downloader', opts.external_downloader_args,
 441                           'default', where=default_downloader) and default_downloader:
 442         # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
 443         opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
 444
 445     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 446         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 447         opts.postprocessor_args.setdefault('sponskrub', [])
 448
 449     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 450                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 451         if val2 is NO_DEFAULT:
 452             val2 = getattr(opts, opt2)
 453         if not val2:
 454             return
 455
 456         if val1 is NO_DEFAULT:
 457             val1 = getattr(opts, opt1)
 458         if val1:
 459             warnings.append(f'{arg1} is ignored since {arg2} was given')
 460         setattr(opts, opt1, default)
 461
 462     # Conflicting options
 463     report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
 464     report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
 465     report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
 466     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 467     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 468     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 469                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 470     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 471     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 472     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 473     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 474     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 475     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 476                     val1=opts.sponskrub and opts.sponskrub_cut)
 477
 478     # Conflicts with --allow-unplayable-formats
 479     report_conflict('--add-metadata', 'addmetadata')
 480     report_conflict('--embed-chapters', 'addchapters')
 481     report_conflict('--embed-info-json', 'embed_infojson')
 482     report_conflict('--embed-subs', 'embedsubtitles')
 483     report_conflict('--embed-thumbnail', 'embedthumbnail')
 484     report_conflict('--extract-audio', 'extractaudio')
 485     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 486     report_conflict('--recode-video', 'recodevideo')
 487     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 488     report_conflict('--remux-video', 'remuxvideo')
 489     report_conflict('--sponskrub', 'sponskrub')
 490     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 491     report_conflict('--xattrs', 'xattrs')
 492
 493     # Fully deprecated options
 494     def report_deprecation(val, old, new=None):
 495         if not val:
 496             return
 497         deprecation_warnings.append(
 498             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 499             else f'{old} is deprecated and may not work as expected')
 500
 501     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 502     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 503     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 504     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 505     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 506
 507     # Dependent options
 508     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 509
 510     if opts.exec_before_dl_cmd:
 511         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 512
 513     if opts.useid:  # --id is not deprecated in youtube-dl
 514         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 515
 516     if opts.overwrites:  # --force-overwrites implies --no-continue
 517         opts.continue_dl = False
 518
 519     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 520         # Add chapters when adding metadata or marking sponsors
 521         opts.addchapters = True
 522
 523     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 524         # Do not unnecessarily download audio
 525         opts.format = 'bestaudio/best'
 526
 527     if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
 528         # If JSON is not printed anywhere, but comments are requested, save it to file
 529         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 530             opts.writeinfojson = True
 531
 532     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 533         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 534         opts.writesubtitles = True
 535
 536     if opts.addmetadata and opts.embed_infojson is None:
 537         # If embedding metadata and infojson is present, embed it
 538         opts.embed_infojson = 'if_exists'
 539
 540     # Ask for passwords
 541     if opts.username is not None and opts.password is None:
 542         opts.password = getpass.getpass('Type account password and press [Return]: ')
 543     if opts.ap_username is not None and opts.ap_password is None:
 544         opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
 545
 546     return warnings, deprecation_warnings
 547
 548
 549 def get_postprocessors(opts):
 550     yield from opts.add_postprocessors
 551
 552     if opts.parse_metadata:
 553         yield {
 554             'key': 'MetadataParser',
 555             'actions': opts.parse_metadata,
 556             'when': 'pre_process'
 557         }
 558     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 559     if sponsorblock_query:
 560         yield {
 561             'key': 'SponsorBlock',
 562             'categories': sponsorblock_query,
 563             'api': opts.sponsorblock_api,
 564             'when': 'after_filter'
 565         }
 566     if opts.convertsubtitles:
 567         yield {
 568             'key': 'FFmpegSubtitlesConvertor',
 569             'format': opts.convertsubtitles,
 570             'when': 'before_dl'
 571         }
 572     if opts.convertthumbnails:
 573         yield {
 574             'key': 'FFmpegThumbnailsConvertor',
 575             'format': opts.convertthumbnails,
 576             'when': 'before_dl'
 577         }
 578     if opts.extractaudio:
 579         yield {
 580             'key': 'FFmpegExtractAudio',
 581             'preferredcodec': opts.audioformat,
 582             'preferredquality': opts.audioquality,
 583             'nopostoverwrites': opts.nopostoverwrites,
 584         }
 585     if opts.remuxvideo:
 586         yield {
 587             'key': 'FFmpegVideoRemuxer',
 588             'preferedformat': opts.remuxvideo,
 589         }
 590     if opts.recodevideo:
 591         yield {
 592             'key': 'FFmpegVideoConvertor',
 593             'preferedformat': opts.recodevideo,
 594         }
 595     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 596     if opts.embedsubtitles:
 597         keep_subs = 'no-keep-subs' not in opts.compat_opts
 598         yield {
 599             'key': 'FFmpegEmbedSubtitle',
 600             # already_have_subtitle = True prevents the file from being deleted after embedding
 601             'already_have_subtitle': opts.writesubtitles and keep_subs
 602         }
 603         if not opts.writeautomaticsub and keep_subs:
 604             opts.writesubtitles = True
 605
 606     # ModifyChapters must run before FFmpegMetadataPP
 607     if opts.remove_chapters or sponsorblock_query:
 608         yield {
 609             'key': 'ModifyChapters',
 610             'remove_chapters_patterns': opts.remove_chapters,
 611             'remove_sponsor_segments': opts.sponsorblock_remove,
 612             'remove_ranges': opts.remove_ranges,
 613             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 614             'force_keyframes': opts.force_keyframes_at_cuts
 615         }
 616     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 617     # FFmpegExtractAudioPP as containers before conversion may not support
 618     # metadata (3gp, webm, etc.)
 619     # By default ffmpeg preserves metadata applicable for both
 620     # source and target containers. From this point the container won't change,
 621     # so metadata can be added here.
 622     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 623         yield {
 624             'key': 'FFmpegMetadata',
 625             'add_chapters': opts.addchapters,
 626             'add_metadata': opts.addmetadata,
 627             'add_infojson': opts.embed_infojson,
 628         }
 629     # Deprecated
 630     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 631     # but must be below EmbedSubtitle and FFmpegMetadata
 632     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 633     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 634     if opts.sponskrub is not False:
 635         yield {
 636             'key': 'SponSkrub',
 637             'path': opts.sponskrub_path,
 638             'args': opts.sponskrub_args,
 639             'cut': opts.sponskrub_cut,
 640             'force': opts.sponskrub_force,
 641             'ignoreerror': opts.sponskrub is None,
 642             '_from_cli': True,
 643         }
 644     if opts.embedthumbnail:
 645         yield {
 646             'key': 'EmbedThumbnail',
 647             # already_have_thumbnail = True prevents the file from being deleted after embedding
 648             'already_have_thumbnail': opts.writethumbnail
 649         }
 650         if not opts.writethumbnail:
 651             opts.writethumbnail = True
 652             opts.outtmpl['pl_thumbnail'] = ''
 653     if opts.split_chapters:
 654         yield {
 655             'key': 'FFmpegSplitChapters',
 656             'force_keyframes': opts.force_keyframes_at_cuts,
 657         }
 658     # XAttrMetadataPP should be run after post-processors that may change file contents
 659     if opts.xattrs:
 660         yield {'key': 'XAttrMetadata'}
 661     if opts.concat_playlist != 'never':
 662         yield {
 663             'key': 'FFmpegConcat',
 664             'only_multi_video': opts.concat_playlist != 'always',
 665             'when': 'playlist',
 666         }
 667     # Exec must be the last PP of each category
 668     for when, exec_cmd in opts.exec_cmd.items():
 669         yield {
 670             'key': 'Exec',
 671             'exec_cmd': exec_cmd,
 672             'when': when,
 673         }
 674
 675
 676 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
 677
 678
 679 def parse_options(argv=None):
 680     """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
 681     parser, opts, urls = parseOpts(argv)
 682     urls = get_urls(urls, opts.batchfile, opts.verbose)
 683
 684     set_compat_opts(opts)
 685     try:
 686         warnings, deprecation_warnings = validate_options(opts)
 687     except ValueError as err:
 688         parser.error(f'{err}\n')
 689
 690     postprocessors = list(get_postprocessors(opts))
 691
 692     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[2:])
 693     any_getting = any(getattr(opts, k) for k in (
 694         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 695         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 696     ))
 697
 698     playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
 699     write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
 700                                and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
 701     if not any((
 702         opts.extract_flat,
 703         opts.dump_single_json,
 704         opts.forceprint.get('playlist'),
 705         opts.print_to_file.get('playlist'),
 706         write_playlist_infojson,
 707     )):
 708         if not playlist_pps:
 709             opts.extract_flat = 'discard'
 710         elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
 711             opts.extract_flat = 'discard_in_playlist'
 712
 713     final_ext = (
 714         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 715         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 716         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 717         else None)
 718
 719     return ParsedOptions(parser, opts, urls, {
 720         'usenetrc': opts.usenetrc,
 721         'netrc_location': opts.netrc_location,
 722         'username': opts.username,
 723         'password': opts.password,
 724         'twofactor': opts.twofactor,
 725         'videopassword': opts.videopassword,
 726         'ap_mso': opts.ap_mso,
 727         'ap_username': opts.ap_username,
 728         'ap_password': opts.ap_password,
 729         'client_certificate': opts.client_certificate,
 730         'client_certificate_key': opts.client_certificate_key,
 731         'client_certificate_password': opts.client_certificate_password,
 732         'quiet': opts.quiet or any_getting or opts.print_json or bool(opts.forceprint),
 733         'no_warnings': opts.no_warnings,
 734         'forceurl': opts.geturl,
 735         'forcetitle': opts.gettitle,
 736         'forceid': opts.getid,
 737         'forcethumbnail': opts.getthumbnail,
 738         'forcedescription': opts.getdescription,
 739         'forceduration': opts.getduration,
 740         'forcefilename': opts.getfilename,
 741         'forceformat': opts.getformat,
 742         'forceprint': opts.forceprint,
 743         'print_to_file': opts.print_to_file,
 744         'forcejson': opts.dumpjson or opts.print_json,
 745         'dump_single_json': opts.dump_single_json,
 746         'force_write_download_archive': opts.force_write_download_archive,
 747         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 748         'skip_download': opts.skip_download,
 749         'format': opts.format,
 750         'allow_unplayable_formats': opts.allow_unplayable_formats,
 751         'ignore_no_formats_error': opts.ignore_no_formats_error,
 752         'format_sort': opts.format_sort,
 753         'format_sort_force': opts.format_sort_force,
 754         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 755         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 756         'check_formats': opts.check_formats,
 757         'listformats': opts.listformats,
 758         'listformats_table': opts.listformats_table,
 759         'outtmpl': opts.outtmpl,
 760         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 761         'paths': opts.paths,
 762         'autonumber_size': opts.autonumber_size,
 763         'autonumber_start': opts.autonumber_start,
 764         'restrictfilenames': opts.restrictfilenames,
 765         'windowsfilenames': opts.windowsfilenames,
 766         'ignoreerrors': opts.ignoreerrors,
 767         'force_generic_extractor': opts.force_generic_extractor,
 768         'ratelimit': opts.ratelimit,
 769         'throttledratelimit': opts.throttledratelimit,
 770         'overwrites': opts.overwrites,
 771         'retries': opts.retries,
 772         'file_access_retries': opts.file_access_retries,
 773         'fragment_retries': opts.fragment_retries,
 774         'extractor_retries': opts.extractor_retries,
 775         'retry_sleep_functions': opts.retry_sleep,
 776         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 777         'keep_fragments': opts.keep_fragments,
 778         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 779         'buffersize': opts.buffersize,
 780         'noresizebuffer': opts.noresizebuffer,
 781         'http_chunk_size': opts.http_chunk_size,
 782         'continuedl': opts.continue_dl,
 783         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 784         'progress_with_newline': opts.progress_with_newline,
 785         'progress_template': opts.progress_template,
 786         'playliststart': opts.playliststart,
 787         'playlistend': opts.playlistend,
 788         'playlistreverse': opts.playlist_reverse,
 789         'playlistrandom': opts.playlist_random,
 790         'lazy_playlist': opts.lazy_playlist,
 791         'noplaylist': opts.noplaylist,
 792         'logtostderr': opts.outtmpl.get('default') == '-',
 793         'consoletitle': opts.consoletitle,
 794         'nopart': opts.nopart,
 795         'updatetime': opts.updatetime,
 796         'writedescription': opts.writedescription,
 797         'writeannotations': opts.writeannotations,
 798         'writeinfojson': opts.writeinfojson,
 799         'allow_playlist_files': opts.allow_playlist_files,
 800         'clean_infojson': opts.clean_infojson,
 801         'getcomments': opts.getcomments,
 802         'writethumbnail': opts.writethumbnail is True,
 803         'write_all_thumbnails': opts.writethumbnail == 'all',
 804         'writelink': opts.writelink,
 805         'writeurllink': opts.writeurllink,
 806         'writewebloclink': opts.writewebloclink,
 807         'writedesktoplink': opts.writedesktoplink,
 808         'writesubtitles': opts.writesubtitles,
 809         'writeautomaticsub': opts.writeautomaticsub,
 810         'allsubtitles': opts.allsubtitles,
 811         'listsubtitles': opts.listsubtitles,
 812         'subtitlesformat': opts.subtitlesformat,
 813         'subtitleslangs': opts.subtitleslangs,
 814         'matchtitle': decodeOption(opts.matchtitle),
 815         'rejecttitle': decodeOption(opts.rejecttitle),
 816         'max_downloads': opts.max_downloads,
 817         'prefer_free_formats': opts.prefer_free_formats,
 818         'trim_file_name': opts.trim_file_name,
 819         'verbose': opts.verbose,
 820         'dump_intermediate_pages': opts.dump_intermediate_pages,
 821         'write_pages': opts.write_pages,
 822         'load_pages': opts.load_pages,
 823         'test': opts.test,
 824         'keepvideo': opts.keepvideo,
 825         'min_filesize': opts.min_filesize,
 826         'max_filesize': opts.max_filesize,
 827         'min_views': opts.min_views,
 828         'max_views': opts.max_views,
 829         'daterange': opts.date,
 830         'cachedir': opts.cachedir,
 831         'youtube_print_sig_code': opts.youtube_print_sig_code,
 832         'age_limit': opts.age_limit,
 833         'download_archive': opts.download_archive,
 834         'break_on_existing': opts.break_on_existing,
 835         'break_on_reject': opts.break_on_reject,
 836         'break_per_url': opts.break_per_url,
 837         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 838         'cookiefile': opts.cookiefile,
 839         'cookiesfrombrowser': opts.cookiesfrombrowser,
 840         'legacyserverconnect': opts.legacy_server_connect,
 841         'nocheckcertificate': opts.no_check_certificate,
 842         'prefer_insecure': opts.prefer_insecure,
 843         'http_headers': opts.headers,
 844         'proxy': opts.proxy,
 845         'socket_timeout': opts.socket_timeout,
 846         'bidi_workaround': opts.bidi_workaround,
 847         'debug_printtraffic': opts.debug_printtraffic,
 848         'prefer_ffmpeg': opts.prefer_ffmpeg,
 849         'include_ads': opts.include_ads,
 850         'default_search': opts.default_search,
 851         'dynamic_mpd': opts.dynamic_mpd,
 852         'extractor_args': opts.extractor_args,
 853         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 854         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 855         'encoding': opts.encoding,
 856         'extract_flat': opts.extract_flat,
 857         'live_from_start': opts.live_from_start,
 858         'wait_for_video': opts.wait_for_video,
 859         'mark_watched': opts.mark_watched,
 860         'merge_output_format': opts.merge_output_format,
 861         'final_ext': final_ext,
 862         'postprocessors': postprocessors,
 863         'fixup': opts.fixup,
 864         'source_address': opts.source_address,
 865         'call_home': opts.call_home,
 866         'sleep_interval_requests': opts.sleep_interval_requests,
 867         'sleep_interval': opts.sleep_interval,
 868         'max_sleep_interval': opts.max_sleep_interval,
 869         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 870         'external_downloader': opts.external_downloader,
 871         'download_ranges': opts.download_ranges,
 872         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 873         'list_thumbnails': opts.list_thumbnails,
 874         'playlist_items': opts.playlist_items,
 875         'xattr_set_filesize': opts.xattr_set_filesize,
 876         'match_filter': opts.match_filter,
 877         'no_color': opts.no_color,
 878         'ffmpeg_location': opts.ffmpeg_location,
 879         'hls_prefer_native': opts.hls_prefer_native,
 880         'hls_use_mpegts': opts.hls_use_mpegts,
 881         'hls_split_discontinuity': opts.hls_split_discontinuity,
 882         'external_downloader_args': opts.external_downloader_args,
 883         'postprocessor_args': opts.postprocessor_args,
 884         'cn_verification_proxy': opts.cn_verification_proxy,
 885         'geo_verification_proxy': opts.geo_verification_proxy,
 886         'geo_bypass': opts.geo_bypass,
 887         'geo_bypass_country': opts.geo_bypass_country,
 888         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 889         '_warnings': warnings,
 890         '_deprecation_warnings': deprecation_warnings,
 891         'compat_opts': opts.compat_opts,
 892     })
 893
 894
 895 def _real_main(argv=None):
 896     setproctitle('yt-dlp')
 897
 898     parser, opts, all_urls, ydl_opts = parse_options(argv)
 899
 900     # Dump user agent
 901     if opts.dump_user_agent:
 902         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 903         write_string(f'{ua}\n', out=sys.stdout)
 904         return
 905
 906     if print_extractor_information(opts, all_urls):
 907         return
 908
 909     # We may need ffmpeg_location without having access to the YoutubeDL instance
 910     # See https://github.com/yt-dlp/yt-dlp/issues/2191
 911     if opts.ffmpeg_location:
 912         FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
 913
 914     with YoutubeDL(ydl_opts) as ydl:
 915         pre_process = opts.update_self or opts.rm_cachedir
 916         actual_use = all_urls or opts.load_info_filename
 917
 918         if opts.rm_cachedir:
 919             ydl.cache.remove()
 920
 921         updater = Updater(ydl)
 922         if opts.update_self and updater.update() and actual_use:
 923             if updater.cmd:
 924                 return updater.restart()
 925             # This code is reachable only for zip variant in py < 3.10
 926             # It makes sense to exit here, but the old behavior is to continue
 927             ydl.report_warning('Restart yt-dlp to use the updated version')
 928             # return 100, 'ERROR: The program must exit for the update to complete'
 929
 930         if not actual_use:
 931             if pre_process:
 932                 return ydl._download_retcode
 933
 934             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 935             parser.error(
 936                 'You must provide at least one URL.\n'
 937                 'Type yt-dlp --help to see a list of all options.')
 938
 939         parser.destroy()
 940         try:
 941             if opts.load_info_filename is not None:
 942                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 943             else:
 944                 return ydl.download(all_urls)
 945         except DownloadCancelled:
 946             ydl.to_screen('Aborting remaining downloads')
 947             return 101
 948
 949
 950 def main(argv=None):
 951     try:
 952         _exit(*variadic(_real_main(argv)))
 953     except DownloadError:
 954         _exit(1)
 955     except SameFileError as e:
 956         _exit(f'ERROR: {e}')
 957     except KeyboardInterrupt:
 958         _exit('\nERROR: Interrupted by user')
 959     except BrokenPipeError as e:
 960         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 961         devnull = os.open(os.devnull, os.O_WRONLY)
 962         os.dup2(devnull, sys.stdout.fileno())
 963         _exit(f'\nERROR: {e}')
 964     except optparse.OptParseError as e:
 965         _exit(2, f'\n{e}')
 966
 967
 968 from .extractor import gen_extractors, list_extractors
 969
 970 __all__ = [
 971     'main',
 972     'YoutubeDL',
 973     'parse_options',
 974     'gen_extractors',
 975     'list_extractors',
 976 ]