yt_dlp/__init__.py

   1 #!/usr/bin/env python3
   2 # coding: utf-8
   3
   4 f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp'  # noqa: F541
   5
   6 __license__ = 'Public Domain'
   7
   8 import io
   9 import itertools
  10 import os
  11 import random
  12 import re
  13 import sys
  14
  15 from .options import parseOpts
  16 from .compat import (
  17     compat_getpass,
  18     compat_os_name,
  19     compat_shlex_quote,
  20 )
  21 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  22 from .utils import (
  23     DateRange,
  24     decodeOption,
  25     DownloadCancelled,
  26     DownloadError,
  27     expand_path,
  28     float_or_none,
  29     GeoUtils,
  30     int_or_none,
  31     match_filter_func,
  32     NO_DEFAULT,
  33     parse_duration,
  34     preferredencoding,
  35     read_batch_urls,
  36     render_table,
  37     SameFileError,
  38     setproctitle,
  39     std_headers,
  40     traverse_obj,
  41     write_string,
  42 )
  43 from .update import run_update
  44 from .downloader import FileDownloader
  45 from .extractor import gen_extractors, list_extractors
  46 from .extractor.common import InfoExtractor
  47 from .extractor.adobepass import MSO_INFO
  48 from .postprocessor import (
  49     FFmpegExtractAudioPP,
  50     FFmpegSubtitlesConvertorPP,
  51     FFmpegThumbnailsConvertorPP,
  52     FFmpegVideoConvertorPP,
  53     FFmpegVideoRemuxerPP,
  54     MetadataFromFieldPP,
  55     MetadataParserPP,
  56 )
  57 from .YoutubeDL import YoutubeDL
  58
  59
  60 def get_urls(urls, batchfile, verbose):
  61     # Batch file verification
  62     batch_urls = []
  63     if batchfile is not None:
  64         try:
  65             if batchfile == '-':
  66                 write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
  67                     'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
  68                 batchfd = sys.stdin
  69             else:
  70                 batchfd = io.open(
  71                     expand_path(batchfile),
  72                     'r', encoding='utf-8', errors='ignore')
  73             batch_urls = read_batch_urls(batchfd)
  74             if verbose:
  75                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  76         except IOError:
  77             sys.exit('ERROR: batch file %s could not be read' % batchfile)
  78     _enc = preferredencoding()
  79     return [
  80         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  81         for url in batch_urls + urls]
  82
  83
  84 def print_extractor_information(opts, urls):
  85     if opts.list_extractors:
  86         for ie in list_extractors(opts.age_limit):
  87             write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n', out=sys.stdout)
  88             matchedUrls = [url for url in urls if ie.suitable(url)]
  89             for mu in matchedUrls:
  90                 write_string('  ' + mu + '\n', out=sys.stdout)
  91     elif opts.list_extractor_descriptions:
  92         for ie in list_extractors(opts.age_limit):
  93             if not ie.working():
  94                 continue
  95             if ie.IE_DESC is False:
  96                 continue
  97             desc = ie.IE_DESC or ie.IE_NAME
  98             if getattr(ie, 'SEARCH_KEY', None) is not None:
  99                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 100                 _COUNTS = ('', '5', '10', 'all')
 101                 desc += f'; "{ie.SEARCH_KEY}:" prefix (Example: "{ie.SEARCH_KEY}{random.choice(_COUNTS)}:{random.choice(_SEARCHES)}")'
 102             write_string(desc + '\n', out=sys.stdout)
 103     elif opts.ap_list_mso:
 104         table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]
 105         write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout)
 106     else:
 107         return False
 108     return True
 109
 110
 111 def set_compat_opts(opts):
 112     def _unused_compat_opt(name):
 113         if name not in opts.compat_opts:
 114             return False
 115         opts.compat_opts.discard(name)
 116         opts.compat_opts.update(['*%s' % name])
 117         return True
 118
 119     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 120         attr = getattr(opts, opt_name)
 121         if compat_name in opts.compat_opts:
 122             if attr is None:
 123                 setattr(opts, opt_name, not default)
 124                 return True
 125             else:
 126                 if remove_compat:
 127                     _unused_compat_opt(compat_name)
 128                 return False
 129         elif attr is None:
 130             setattr(opts, opt_name, default)
 131         return None
 132
 133     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 134     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 135     set_default_compat('no-clean-infojson', 'clean_infojson')
 136     if 'no-attach-info-json' in opts.compat_opts:
 137         if opts.embed_infojson:
 138             _unused_compat_opt('no-attach-info-json')
 139         else:
 140             opts.embed_infojson = False
 141     if 'format-sort' in opts.compat_opts:
 142         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 143     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 144     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 145     if _video_multistreams_set is False and _audio_multistreams_set is False:
 146         _unused_compat_opt('multistreams')
 147     if 'filename' in opts.compat_opts:
 148         if opts.outtmpl.get('default') is None:
 149             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 150         else:
 151             _unused_compat_opt('filename')
 152
 153
 154 def validate_options(opts):
 155     def validate(cndn, name, value=None, msg=None):
 156         if cndn:
 157             return True
 158         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 159
 160     def validate_in(name, value, items, msg=None):
 161         return validate(value is None or value in items, name, value, msg)
 162
 163     def validate_regex(name, value, regex):
 164         return validate(value is None or re.match(regex, value), name, value)
 165
 166     def validate_positive(name, value, strict=False):
 167         return validate(value is None or value > 0 or (not strict and value == 0),
 168                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 169
 170     def validate_minmax(min_val, max_val, min_name, max_name=None):
 171         if max_val is None or min_val is None or max_val >= min_val:
 172             return
 173         if not max_name:
 174             min_name, max_name = f'min {min_name}', f'max {min_name}'
 175         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 176
 177     # Usernames and passwords
 178     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 179              '.netrc', msg='using {name} conflicts with giving username/password')
 180     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 181     validate(opts.ap_password is None or opts.ap_username is not None,
 182              'TV Provider account username', msg='{name} missing')
 183     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 184                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 185
 186     # Numbers
 187     validate_positive('autonumber start', opts.autonumber_start)
 188     validate_positive('autonumber size', opts.autonumber_size, True)
 189     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 190     validate_positive('playlist start', opts.playliststart, True)
 191     if opts.playlistend != -1:
 192         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 193
 194     # Time ranges
 195     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 196     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 197     validate_positive('sleep interval', opts.sleep_interval)
 198     validate_positive('max sleep interval', opts.max_sleep_interval)
 199     if opts.sleep_interval is None:
 200         validate(
 201             opts.max_sleep_interval is None, 'min sleep interval',
 202             msg='{name} must be specified; use --min-sleep-interval')
 203     elif opts.max_sleep_interval is None:
 204         opts.max_sleep_interval = opts.sleep_interval
 205     else:
 206         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 207
 208     if opts.wait_for_video is not None:
 209         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 210         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 211                  'time range to wait for video', opts.wait_for_video)
 212         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 213         opts.wait_for_video = (min_wait, max_wait)
 214
 215     # Format sort
 216     for f in opts.format_sort:
 217         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 218
 219     # Postprocessor formats
 220     validate_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
 221     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 222     validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
 223     if opts.recodevideo is not None:
 224         opts.recodevideo = opts.recodevideo.replace(' ', '')
 225         validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 226     if opts.remuxvideo is not None:
 227         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 228         validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 229     if opts.audioquality:
 230         opts.audioquality = opts.audioquality.strip('k').strip('K')
 231         # int_or_none prevents inf, nan
 232         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 233
 234     # Retries
 235     def parse_retries(name, value):
 236         if value is None:
 237             return None
 238         elif value in ('inf', 'infinite'):
 239             return float('inf')
 240         try:
 241             return int(value)
 242         except (TypeError, ValueError):
 243             validate(False, f'{name} retry count', value)
 244
 245     opts.retries = parse_retries('download', opts.retries)
 246     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 247     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 248     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 249
 250     # Bytes
 251     def parse_bytes(name, value):
 252         if value is None:
 253             return None
 254         numeric_limit = FileDownloader.parse_bytes(value)
 255         validate(numeric_limit is not None, 'rate limit', value)
 256         return numeric_limit
 257
 258     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 259     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 260     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 261     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 262     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 263     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 264
 265     # Output templates
 266     def validate_outtmpl(tmpl, msg):
 267         err = YoutubeDL.validate_outtmpl(tmpl)
 268         if err:
 269             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 270
 271     for k, tmpl in opts.outtmpl.items():
 272         validate_outtmpl(tmpl, f'{k} output template')
 273     for type_, tmpl_list in opts.forceprint.items():
 274         for tmpl in tmpl_list:
 275             validate_outtmpl(tmpl, f'{type_} print template')
 276     for type_, tmpl_list in opts.print_to_file.items():
 277         for tmpl, file in tmpl_list:
 278             validate_outtmpl(tmpl, f'{type_} print to file template')
 279             validate_outtmpl(file, f'{type_} print to file filename')
 280     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 281     for k, tmpl in opts.progress_template.items():
 282         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 283         validate_outtmpl(tmpl, f'{k} template')
 284
 285     outtmpl_default = opts.outtmpl.get('default')
 286     if outtmpl_default == '':
 287         opts.skip_download = None
 288         del opts.outtmpl['default']
 289     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 290         raise ValueError(
 291             'Cannot download a video and extract audio into the same file! '
 292             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 293
 294     # Remove chapters
 295     remove_chapters_patterns, opts.remove_ranges = [], []
 296     for regex in opts.remove_chapters or []:
 297         if regex.startswith('*'):
 298             dur = list(map(parse_duration, regex[1:].split('-')))
 299             if len(dur) == 2 and all(t is not None for t in dur):
 300                 opts.remove_ranges.append(tuple(dur))
 301                 continue
 302             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 303         try:
 304             remove_chapters_patterns.append(re.compile(regex))
 305         except re.error as err:
 306             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 307     opts.remove_chapters = remove_chapters_patterns
 308
 309     # Cookies from browser
 310     if opts.cookiesfrombrowser:
 311         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 312         if mobj is None:
 313             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 314         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 315         browser_name = browser_name.lower()
 316         if browser_name not in SUPPORTED_BROWSERS:
 317             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 318                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 319         if keyring is not None:
 320             keyring = keyring.upper()
 321             if keyring not in SUPPORTED_KEYRINGS:
 322                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 323                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 324         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 325
 326     # MetadataParser
 327     def metadataparser_actions(f):
 328         if isinstance(f, str):
 329             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 330             try:
 331                 actions = [MetadataFromFieldPP.to_action(f)]
 332             except Exception as err:
 333                 raise ValueError(f'{cmd} is invalid; {err}')
 334         else:
 335             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 336             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 337
 338         for action in actions:
 339             try:
 340                 MetadataParserPP.validate_action(*action)
 341             except Exception as err:
 342                 raise ValueError(f'{cmd} is invalid; {err}')
 343             yield action
 344
 345     parse_metadata = opts.parse_metadata or []
 346     if opts.metafromtitle is not None:
 347         parse_metadata.append('title:%s' % opts.metafromtitle)
 348     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 349
 350     # Other options
 351     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 352     if geo_bypass_code is not None:
 353         try:
 354             GeoUtils.random_ipv4(geo_bypass_code)
 355         except Exception:
 356             raise ValueError('unsupported geo-bypass country or ip-block')
 357
 358     opts.match_filter = match_filter_func(opts.match_filter)
 359
 360     if opts.download_archive is not None:
 361         opts.download_archive = expand_path(opts.download_archive)
 362
 363     if opts.user_agent is not None:
 364         opts.headers.setdefault('User-Agent', opts.user_agent)
 365     if opts.referer is not None:
 366         opts.headers.setdefault('Referer', opts.referer)
 367
 368     if opts.no_sponsorblock:
 369         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 370
 371     warnings, deprecation_warnings = [], []
 372
 373     # Common mistake: -f best
 374     if opts.format == 'best':
 375         warnings.append('.\n         '.join((
 376             '"-f best" selects the best pre-merged format which is often not the best option',
 377             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 378             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 379
 380     # --(postprocessor/downloader)-args without name
 381     def report_args_compat(name, value, key1, key2=None):
 382         if key1 in value and key2 not in value:
 383             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 384             return True
 385         return False
 386
 387     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 388     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 389         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 390         opts.postprocessor_args.setdefault('sponskrub', [])
 391
 392     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 393                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 394         if val2 is NO_DEFAULT:
 395             val2 = getattr(opts, opt2)
 396         if not val2:
 397             return
 398
 399         if val1 is NO_DEFAULT:
 400             val1 = getattr(opts, opt1)
 401         if val1:
 402             warnings.append(f'{arg1} is ignored since {arg2} was given')
 403         setattr(opts, opt1, default)
 404
 405     # Conflicting options
 406     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 407     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 408     report_conflict('--exec-before-download', 'exec_before_dl_cmd', '"--exec before_dl:"', 'exec_cmd', opts.exec_cmd.get('before_dl'))
 409     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 410     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 411     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 412     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 413     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 414     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters', val1=opts.sponskrub and opts.sponskrub_cut)
 415
 416     # Conflicts with --allow-unplayable-formats
 417     report_conflict('--add-metadata', 'addmetadata')
 418     report_conflict('--embed-chapters', 'addchapters')
 419     report_conflict('--embed-info-json', 'embed_infojson')
 420     report_conflict('--embed-subs', 'embedsubtitles')
 421     report_conflict('--embed-thumbnail', 'embedthumbnail')
 422     report_conflict('--extract-audio', 'extractaudio')
 423     report_conflict('--fixup', 'fixup', val1=(opts.fixup or '').lower() in ('', 'never', 'ignore'), default='never')
 424     report_conflict('--recode-video', 'recodevideo')
 425     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 426     report_conflict('--remux-video', 'remuxvideo')
 427     report_conflict('--sponskrub', 'sponskrub')
 428     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 429     report_conflict('--xattrs', 'xattrs')
 430
 431     # Fully deprecated options
 432     def report_deprecation(val, old, new=None):
 433         if not val:
 434             return
 435         deprecation_warnings.append(
 436             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 437             else f'{old} is deprecated and may not work as expected')
 438
 439     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 440     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 441     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 442     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 443     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 444
 445     # Dependent options
 446     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 447
 448     if opts.exec_before_dl_cmd:
 449         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 450
 451     if opts.useid:  # --id is not deprecated in youtube-dl
 452         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 453
 454     if opts.overwrites:  # --force-overwrites implies --no-continue
 455         opts.continue_dl = False
 456
 457     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 458         # Add chapters when adding metadata or marking sponsors
 459         opts.addchapters = True
 460
 461     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 462         # Do not unnecessarily download audio
 463         opts.format = 'bestaudio/best'
 464
 465     if opts.getcomments and opts.writeinfojson is None:
 466         # If JSON is not printed anywhere, but comments are requested, save it to file
 467         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 468             opts.writeinfojson = True
 469
 470     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 471         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 472         opts.writesubtitles = True
 473
 474     if opts.addmetadata and opts.embed_infojson is None:
 475         # If embedding metadata and infojson is present, embed it
 476         opts.embed_infojson = 'if_exists'
 477
 478     # Ask for passwords
 479     if opts.username is not None and opts.password is None:
 480         opts.password = compat_getpass('Type account password and press [Return]: ')
 481     if opts.ap_username is not None and opts.ap_password is None:
 482         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 483
 484     return warnings, deprecation_warnings
 485
 486
 487 def get_postprocessors(opts):
 488     yield from opts.add_postprocessors
 489
 490     if opts.parse_metadata:
 491         yield {
 492             'key': 'MetadataParser',
 493             'actions': opts.parse_metadata,
 494             'when': 'pre_process'
 495         }
 496     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 497     if sponsorblock_query:
 498         yield {
 499             'key': 'SponsorBlock',
 500             'categories': sponsorblock_query,
 501             'api': opts.sponsorblock_api,
 502             'when': 'after_filter'
 503         }
 504     if opts.convertsubtitles:
 505         yield {
 506             'key': 'FFmpegSubtitlesConvertor',
 507             'format': opts.convertsubtitles,
 508             'when': 'before_dl'
 509         }
 510     if opts.convertthumbnails:
 511         yield {
 512             'key': 'FFmpegThumbnailsConvertor',
 513             'format': opts.convertthumbnails,
 514             'when': 'before_dl'
 515         }
 516     if opts.extractaudio:
 517         yield {
 518             'key': 'FFmpegExtractAudio',
 519             'preferredcodec': opts.audioformat,
 520             'preferredquality': opts.audioquality,
 521             'nopostoverwrites': opts.nopostoverwrites,
 522         }
 523     if opts.remuxvideo:
 524         yield {
 525             'key': 'FFmpegVideoRemuxer',
 526             'preferedformat': opts.remuxvideo,
 527         }
 528     if opts.recodevideo:
 529         yield {
 530             'key': 'FFmpegVideoConvertor',
 531             'preferedformat': opts.recodevideo,
 532         }
 533     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 534     if opts.embedsubtitles:
 535         keep_subs = 'no-keep-subs' not in opts.compat_opts
 536         yield {
 537             'key': 'FFmpegEmbedSubtitle',
 538             # already_have_subtitle = True prevents the file from being deleted after embedding
 539             'already_have_subtitle': opts.writesubtitles and keep_subs
 540         }
 541         if not opts.writeautomaticsub and keep_subs:
 542             opts.writesubtitles = True
 543
 544     # ModifyChapters must run before FFmpegMetadataPP
 545     if opts.remove_chapters or sponsorblock_query:
 546         yield {
 547             'key': 'ModifyChapters',
 548             'remove_chapters_patterns': opts.remove_chapters,
 549             'remove_sponsor_segments': opts.sponsorblock_remove,
 550             'remove_ranges': opts.remove_ranges,
 551             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 552             'force_keyframes': opts.force_keyframes_at_cuts
 553         }
 554     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 555     # FFmpegExtractAudioPP as containers before conversion may not support
 556     # metadata (3gp, webm, etc.)
 557     # By default ffmpeg preserves metadata applicable for both
 558     # source and target containers. From this point the container won't change,
 559     # so metadata can be added here.
 560     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 561         yield {
 562             'key': 'FFmpegMetadata',
 563             'add_chapters': opts.addchapters,
 564             'add_metadata': opts.addmetadata,
 565             'add_infojson': opts.embed_infojson,
 566         }
 567     # Deprecated
 568     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 569     # but must be below EmbedSubtitle and FFmpegMetadata
 570     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 571     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 572     if opts.sponskrub is not False:
 573         yield {
 574             'key': 'SponSkrub',
 575             'path': opts.sponskrub_path,
 576             'args': opts.sponskrub_args,
 577             'cut': opts.sponskrub_cut,
 578             'force': opts.sponskrub_force,
 579             'ignoreerror': opts.sponskrub is None,
 580             '_from_cli': True,
 581         }
 582     if opts.embedthumbnail:
 583         yield {
 584             'key': 'EmbedThumbnail',
 585             # already_have_thumbnail = True prevents the file from being deleted after embedding
 586             'already_have_thumbnail': opts.writethumbnail
 587         }
 588         if not opts.writethumbnail:
 589             opts.writethumbnail = True
 590             opts.outtmpl['pl_thumbnail'] = ''
 591     if opts.split_chapters:
 592         yield {
 593             'key': 'FFmpegSplitChapters',
 594             'force_keyframes': opts.force_keyframes_at_cuts,
 595         }
 596     # XAttrMetadataPP should be run after post-processors that may change file contents
 597     if opts.xattrs:
 598         yield {'key': 'XAttrMetadata'}
 599     if opts.concat_playlist != 'never':
 600         yield {
 601             'key': 'FFmpegConcat',
 602             'only_multi_video': opts.concat_playlist != 'always',
 603             'when': 'playlist',
 604         }
 605     # Exec must be the last PP of each category
 606     for when, exec_cmd in opts.exec_cmd.items():
 607         yield {
 608             'key': 'Exec',
 609             'exec_cmd': exec_cmd,
 610             'when': when,
 611         }
 612
 613
 614 def parse_options(argv=None):
 615     """ @returns (parser, opts, urls, ydl_opts) """
 616     parser, opts, urls = parseOpts(argv)
 617     urls = get_urls(urls, opts.batchfile, opts.verbose)
 618
 619     set_compat_opts(opts)
 620     try:
 621         warnings, deprecation_warnings = validate_options(opts)
 622     except ValueError as err:
 623         parser.error(f'{err}\n')
 624
 625     postprocessors = list(get_postprocessors(opts))
 626
 627     any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
 628                    or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
 629                    or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
 630
 631     any_printing = opts.print_json
 632
 633     final_ext = (
 634         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 635         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 636         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 637         else None)
 638
 639     return parser, opts, urls, {
 640         'usenetrc': opts.usenetrc,
 641         'netrc_location': opts.netrc_location,
 642         'username': opts.username,
 643         'password': opts.password,
 644         'twofactor': opts.twofactor,
 645         'videopassword': opts.videopassword,
 646         'ap_mso': opts.ap_mso,
 647         'ap_username': opts.ap_username,
 648         'ap_password': opts.ap_password,
 649         'quiet': (opts.quiet or any_getting or any_printing),
 650         'no_warnings': opts.no_warnings,
 651         'forceurl': opts.geturl,
 652         'forcetitle': opts.gettitle,
 653         'forceid': opts.getid,
 654         'forcethumbnail': opts.getthumbnail,
 655         'forcedescription': opts.getdescription,
 656         'forceduration': opts.getduration,
 657         'forcefilename': opts.getfilename,
 658         'forceformat': opts.getformat,
 659         'forceprint': opts.forceprint,
 660         'print_to_file': opts.print_to_file,
 661         'forcejson': opts.dumpjson or opts.print_json,
 662         'dump_single_json': opts.dump_single_json,
 663         'force_write_download_archive': opts.force_write_download_archive,
 664         'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
 665         'skip_download': opts.skip_download,
 666         'format': opts.format,
 667         'allow_unplayable_formats': opts.allow_unplayable_formats,
 668         'ignore_no_formats_error': opts.ignore_no_formats_error,
 669         'format_sort': opts.format_sort,
 670         'format_sort_force': opts.format_sort_force,
 671         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 672         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 673         'check_formats': opts.check_formats,
 674         'listformats': opts.listformats,
 675         'listformats_table': opts.listformats_table,
 676         'outtmpl': opts.outtmpl,
 677         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 678         'paths': opts.paths,
 679         'autonumber_size': opts.autonumber_size,
 680         'autonumber_start': opts.autonumber_start,
 681         'restrictfilenames': opts.restrictfilenames,
 682         'windowsfilenames': opts.windowsfilenames,
 683         'ignoreerrors': opts.ignoreerrors,
 684         'force_generic_extractor': opts.force_generic_extractor,
 685         'ratelimit': opts.ratelimit,
 686         'throttledratelimit': opts.throttledratelimit,
 687         'overwrites': opts.overwrites,
 688         'retries': opts.retries,
 689         'file_access_retries': opts.file_access_retries,
 690         'fragment_retries': opts.fragment_retries,
 691         'extractor_retries': opts.extractor_retries,
 692         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 693         'keep_fragments': opts.keep_fragments,
 694         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 695         'buffersize': opts.buffersize,
 696         'noresizebuffer': opts.noresizebuffer,
 697         'http_chunk_size': opts.http_chunk_size,
 698         'continuedl': opts.continue_dl,
 699         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 700         'progress_with_newline': opts.progress_with_newline,
 701         'progress_template': opts.progress_template,
 702         'playliststart': opts.playliststart,
 703         'playlistend': opts.playlistend,
 704         'playlistreverse': opts.playlist_reverse,
 705         'playlistrandom': opts.playlist_random,
 706         'noplaylist': opts.noplaylist,
 707         'logtostderr': opts.outtmpl.get('default') == '-',
 708         'consoletitle': opts.consoletitle,
 709         'nopart': opts.nopart,
 710         'updatetime': opts.updatetime,
 711         'writedescription': opts.writedescription,
 712         'writeannotations': opts.writeannotations,
 713         'writeinfojson': opts.writeinfojson,
 714         'allow_playlist_files': opts.allow_playlist_files,
 715         'clean_infojson': opts.clean_infojson,
 716         'getcomments': opts.getcomments,
 717         'writethumbnail': opts.writethumbnail is True,
 718         'write_all_thumbnails': opts.writethumbnail == 'all',
 719         'writelink': opts.writelink,
 720         'writeurllink': opts.writeurllink,
 721         'writewebloclink': opts.writewebloclink,
 722         'writedesktoplink': opts.writedesktoplink,
 723         'writesubtitles': opts.writesubtitles,
 724         'writeautomaticsub': opts.writeautomaticsub,
 725         'allsubtitles': opts.allsubtitles,
 726         'listsubtitles': opts.listsubtitles,
 727         'subtitlesformat': opts.subtitlesformat,
 728         'subtitleslangs': opts.subtitleslangs,
 729         'matchtitle': decodeOption(opts.matchtitle),
 730         'rejecttitle': decodeOption(opts.rejecttitle),
 731         'max_downloads': opts.max_downloads,
 732         'prefer_free_formats': opts.prefer_free_formats,
 733         'trim_file_name': opts.trim_file_name,
 734         'verbose': opts.verbose,
 735         'dump_intermediate_pages': opts.dump_intermediate_pages,
 736         'write_pages': opts.write_pages,
 737         'test': opts.test,
 738         'keepvideo': opts.keepvideo,
 739         'min_filesize': opts.min_filesize,
 740         'max_filesize': opts.max_filesize,
 741         'min_views': opts.min_views,
 742         'max_views': opts.max_views,
 743         'daterange': opts.date,
 744         'cachedir': opts.cachedir,
 745         'youtube_print_sig_code': opts.youtube_print_sig_code,
 746         'age_limit': opts.age_limit,
 747         'download_archive': opts.download_archive,
 748         'break_on_existing': opts.break_on_existing,
 749         'break_on_reject': opts.break_on_reject,
 750         'break_per_url': opts.break_per_url,
 751         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 752         'cookiefile': opts.cookiefile,
 753         'cookiesfrombrowser': opts.cookiesfrombrowser,
 754         'legacyserverconnect': opts.legacy_server_connect,
 755         'nocheckcertificate': opts.no_check_certificate,
 756         'prefer_insecure': opts.prefer_insecure,
 757         'http_headers': opts.headers,
 758         'proxy': opts.proxy,
 759         'socket_timeout': opts.socket_timeout,
 760         'bidi_workaround': opts.bidi_workaround,
 761         'debug_printtraffic': opts.debug_printtraffic,
 762         'prefer_ffmpeg': opts.prefer_ffmpeg,
 763         'include_ads': opts.include_ads,
 764         'default_search': opts.default_search,
 765         'dynamic_mpd': opts.dynamic_mpd,
 766         'extractor_args': opts.extractor_args,
 767         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 768         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 769         'encoding': opts.encoding,
 770         'extract_flat': opts.extract_flat,
 771         'live_from_start': opts.live_from_start,
 772         'wait_for_video': opts.wait_for_video,
 773         'mark_watched': opts.mark_watched,
 774         'merge_output_format': opts.merge_output_format,
 775         'final_ext': final_ext,
 776         'postprocessors': postprocessors,
 777         'fixup': opts.fixup,
 778         'source_address': opts.source_address,
 779         'call_home': opts.call_home,
 780         'sleep_interval_requests': opts.sleep_interval_requests,
 781         'sleep_interval': opts.sleep_interval,
 782         'max_sleep_interval': opts.max_sleep_interval,
 783         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 784         'external_downloader': opts.external_downloader,
 785         'list_thumbnails': opts.list_thumbnails,
 786         'playlist_items': opts.playlist_items,
 787         'xattr_set_filesize': opts.xattr_set_filesize,
 788         'match_filter': opts.match_filter,
 789         'no_color': opts.no_color,
 790         'ffmpeg_location': opts.ffmpeg_location,
 791         'hls_prefer_native': opts.hls_prefer_native,
 792         'hls_use_mpegts': opts.hls_use_mpegts,
 793         'hls_split_discontinuity': opts.hls_split_discontinuity,
 794         'external_downloader_args': opts.external_downloader_args,
 795         'postprocessor_args': opts.postprocessor_args,
 796         'cn_verification_proxy': opts.cn_verification_proxy,
 797         'geo_verification_proxy': opts.geo_verification_proxy,
 798         'geo_bypass': opts.geo_bypass,
 799         'geo_bypass_country': opts.geo_bypass_country,
 800         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 801         '_warnings': warnings,
 802         '_deprecation_warnings': deprecation_warnings,
 803         'compat_opts': opts.compat_opts,
 804     }
 805
 806
 807 def _real_main(argv=None):
 808     setproctitle('yt-dlp')
 809
 810     parser, opts, all_urls, ydl_opts = parse_options(argv)
 811
 812     # Dump user agent
 813     if opts.dump_user_agent:
 814         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 815         write_string(f'{ua}\n', out=sys.stdout)
 816         sys.exit(0)
 817
 818     if print_extractor_information(opts, all_urls):
 819         sys.exit(0)
 820
 821     with YoutubeDL(ydl_opts) as ydl:
 822         actual_use = all_urls or opts.load_info_filename
 823
 824         # Remove cache dir
 825         if opts.rm_cachedir:
 826             ydl.cache.remove()
 827
 828         # Update version
 829         if opts.update_self:
 830             # If updater returns True, exit. Required for windows
 831             if run_update(ydl):
 832                 if actual_use:
 833                     sys.exit('ERROR: The program must exit for the update to complete')
 834                 sys.exit()
 835
 836         # Maybe do nothing
 837         if not actual_use:
 838             if opts.update_self or opts.rm_cachedir:
 839                 sys.exit()
 840
 841             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 842             parser.error(
 843                 'You must provide at least one URL.\n'
 844                 'Type yt-dlp --help to see a list of all options.')
 845
 846         try:
 847             if opts.load_info_filename is not None:
 848                 retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
 849             else:
 850                 retcode = ydl.download(all_urls)
 851         except DownloadCancelled:
 852             ydl.to_screen('Aborting remaining downloads')
 853             retcode = 101
 854
 855     sys.exit(retcode)
 856
 857
 858 def main(argv=None):
 859     try:
 860         _real_main(argv)
 861     except DownloadError:
 862         sys.exit(1)
 863     except SameFileError as e:
 864         sys.exit(f'ERROR: {e}')
 865     except KeyboardInterrupt:
 866         sys.exit('\nERROR: Interrupted by user')
 867     except BrokenPipeError as e:
 868         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 869         devnull = os.open(os.devnull, os.O_WRONLY)
 870         os.dup2(devnull, sys.stdout.fileno())
 871         sys.exit(f'\nERROR: {e}')
 872
 873
 874 __all__ = [
 875     'main',
 876     'YoutubeDL',
 877     'parse_options',
 878     'gen_extractors',
 879     'list_extractors',
 880 ]