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 codecs
   9 import io
  10 import itertools
  11 import os
  12 import random
  13 import re
  14 import sys
  15
  16 from .options import parseOpts
  17 from .compat import (
  18     compat_getpass,
  19     compat_os_name,
  20     compat_shlex_quote,
  21     workaround_optparse_bug9161,
  22 )
  23 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  24 from .utils import (
  25     DateRange,
  26     decodeOption,
  27     DownloadCancelled,
  28     DownloadError,
  29     expand_path,
  30     float_or_none,
  31     GeoUtils,
  32     int_or_none,
  33     match_filter_func,
  34     NO_DEFAULT,
  35     parse_duration,
  36     preferredencoding,
  37     read_batch_urls,
  38     render_table,
  39     SameFileError,
  40     setproctitle,
  41     std_headers,
  42     traverse_obj,
  43     write_string,
  44 )
  45 from .update import run_update
  46 from .downloader import FileDownloader
  47 from .extractor import gen_extractors, list_extractors
  48 from .extractor.common import InfoExtractor
  49 from .extractor.adobepass import MSO_INFO
  50 from .postprocessor import (
  51     FFmpegExtractAudioPP,
  52     FFmpegSubtitlesConvertorPP,
  53     FFmpegThumbnailsConvertorPP,
  54     FFmpegVideoConvertorPP,
  55     FFmpegVideoRemuxerPP,
  56     MetadataFromFieldPP,
  57     MetadataParserPP,
  58 )
  59 from .YoutubeDL import YoutubeDL
  60
  61
  62 def get_urls(urls, batchfile, verbose):
  63     # Batch file verification
  64     batch_urls = []
  65     if batchfile is not None:
  66         try:
  67             if batchfile == '-':
  68                 write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
  69                     'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
  70                 batchfd = sys.stdin
  71             else:
  72                 batchfd = io.open(
  73                     expand_path(batchfile),
  74                     'r', encoding='utf-8', errors='ignore')
  75             batch_urls = read_batch_urls(batchfd)
  76             if verbose:
  77                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  78         except IOError:
  79             sys.exit('ERROR: batch file %s could not be read' % batchfile)
  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     if opts.list_extractors:
  88         for ie in list_extractors(opts.age_limit):
  89             write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n', out=sys.stdout)
  90             matchedUrls = [url for url in urls if ie.suitable(url)]
  91             for mu in matchedUrls:
  92                 write_string('  ' + mu + '\n', out=sys.stdout)
  93     elif opts.list_extractor_descriptions:
  94         for ie in list_extractors(opts.age_limit):
  95             if not ie.working():
  96                 continue
  97             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
  98             if desc is False:
  99                 continue
 100             if getattr(ie, 'SEARCH_KEY', None) is not None:
 101                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 102                 _COUNTS = ('', '5', '10', 'all')
 103                 desc += f'; "{ie.SEARCH_KEY}:" prefix (Example: "{ie.SEARCH_KEY}{random.choice(_COUNTS)}:{random.choice(_SEARCHES)}")'
 104             write_string(desc + '\n', out=sys.stdout)
 105     elif opts.ap_list_mso:
 106         table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]
 107         write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout)
 108     else:
 109         return False
 110     return True
 111
 112
 113 def set_compat_opts(opts):
 114     def _unused_compat_opt(name):
 115         if name not in opts.compat_opts:
 116             return False
 117         opts.compat_opts.discard(name)
 118         opts.compat_opts.update(['*%s' % name])
 119         return True
 120
 121     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 122         attr = getattr(opts, opt_name)
 123         if compat_name in opts.compat_opts:
 124             if attr is None:
 125                 setattr(opts, opt_name, not default)
 126                 return True
 127             else:
 128                 if remove_compat:
 129                     _unused_compat_opt(compat_name)
 130                 return False
 131         elif attr is None:
 132             setattr(opts, opt_name, default)
 133         return None
 134
 135     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 136     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 137     set_default_compat('no-clean-infojson', 'clean_infojson')
 138     if 'no-attach-info-json' in opts.compat_opts:
 139         if opts.embed_infojson:
 140             _unused_compat_opt('no-attach-info-json')
 141         else:
 142             opts.embed_infojson = False
 143     if 'format-sort' in opts.compat_opts:
 144         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 145     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 146     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 147     if _video_multistreams_set is False and _audio_multistreams_set is False:
 148         _unused_compat_opt('multistreams')
 149     if 'filename' in opts.compat_opts:
 150         if opts.outtmpl.get('default') is None:
 151             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 152         else:
 153             _unused_compat_opt('filename')
 154
 155
 156 def validate_options(opts):
 157     def validate(cndn, name, value=None, msg=None):
 158         if cndn:
 159             return True
 160         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 161
 162     def validate_in(name, value, items, msg=None):
 163         return validate(value is None or value in items, name, value, msg)
 164
 165     def validate_regex(name, value, regex):
 166         return validate(value is None or re.match(regex, value), name, value)
 167
 168     def validate_positive(name, value, strict=False):
 169         return validate(value is None or value > 0 or (not strict and value == 0),
 170                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 171
 172     def validate_minmax(min_val, max_val, min_name, max_name=None):
 173         if max_val is None or min_val is None or max_val >= min_val:
 174             return
 175         if not max_name:
 176             min_name, max_name = f'min {min_name}', f'max {min_name}'
 177         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 178
 179     # Usernames and passwords
 180     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 181              '.netrc', msg='using {name} conflicts with giving username/password')
 182     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 183     validate(opts.ap_password is None or opts.ap_username is not None,
 184              'TV Provider account username', msg='{name} missing')
 185     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 186                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 187
 188     # Numbers
 189     validate_positive('autonumber start', opts.autonumber_start)
 190     validate_positive('autonumber size', opts.autonumber_size, True)
 191     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 192     validate_positive('playlist start', opts.playliststart, True)
 193     if opts.playlistend != -1:
 194         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 195
 196     # Time ranges
 197     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 198     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 199     validate_positive('sleep interval', opts.sleep_interval)
 200     validate_positive('max sleep interval', opts.max_sleep_interval)
 201     if opts.max_sleep_interval is not None:
 202         validate(
 203             opts.sleep_interval is not None, 'min sleep interval',
 204             msg='{name} must be specified; use --min-sleep-interval')
 205         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 206
 207     if opts.wait_for_video is not None:
 208         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 209         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 210                  'time range to wait for video', opts.wait_for_video)
 211         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 212         opts.wait_for_video = (min_wait, max_wait)
 213
 214     # Format sort
 215     for f in opts.format_sort:
 216         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 217
 218     # Postprocessor formats
 219     validate_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
 220     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 221     validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
 222     if opts.recodevideo is not None:
 223         opts.recodevideo = opts.recodevideo.replace(' ', '')
 224         validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 225     if opts.remuxvideo is not None:
 226         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 227         validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 228     if opts.audioquality:
 229         opts.audioquality = opts.audioquality.strip('k').strip('K')
 230         # int_or_none prevents inf, nan
 231         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 232
 233     # Retries
 234     def parse_retries(name, value):
 235         if value is None:
 236             return None
 237         elif value in ('inf', 'infinite'):
 238             return float('inf')
 239         try:
 240             return int(value)
 241         except (TypeError, ValueError):
 242             validate(False, f'{name} retry count', value)
 243
 244     opts.retries = parse_retries('download', opts.retries)
 245     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 246     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 247     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 248
 249     # Bytes
 250     def parse_bytes(name, value):
 251         if value is None:
 252             return None
 253         numeric_limit = FileDownloader.parse_bytes(value)
 254         validate(numeric_limit is not None, 'rate limit', value)
 255         return numeric_limit
 256
 257     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 258     opts.ratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 259     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 260     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 261     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 262     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 263
 264     # Output templates
 265     def validate_outtmpl(tmpl, msg):
 266         err = YoutubeDL.validate_outtmpl(tmpl)
 267         if err:
 268             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 269
 270     for k, tmpl in opts.outtmpl.items():
 271         validate_outtmpl(tmpl, f'{k} output template')
 272     for type_, tmpl_list in opts.forceprint.items():
 273         for tmpl in tmpl_list:
 274             validate_outtmpl(tmpl, f'{type_} print template')
 275     for type_, tmpl_list in opts.print_to_file.items():
 276         for tmpl, file in tmpl_list:
 277             validate_outtmpl(tmpl, f'{type_} print to file template')
 278             validate_outtmpl(file, f'{type_} print to file filename')
 279     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 280     for k, tmpl in opts.progress_template.items():
 281         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 282         validate_outtmpl(tmpl, f'{k} template')
 283
 284     outtmpl_default = opts.outtmpl.get('default')
 285     if outtmpl_default == '':
 286         opts.skip_download = None
 287         del opts.outtmpl['default']
 288     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 289         raise ValueError(
 290             'Cannot download a video and extract audio into the same file! '
 291             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 292
 293     # Remove chapters
 294     remove_chapters_patterns, opts.remove_ranges = [], []
 295     for regex in opts.remove_chapters or []:
 296         if regex.startswith('*'):
 297             dur = list(map(parse_duration, regex[1:].split('-')))
 298             if len(dur) == 2 and all(t is not None for t in dur):
 299                 opts.remove_ranges.append(tuple(dur))
 300                 continue
 301             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 302         try:
 303             remove_chapters_patterns.append(re.compile(regex))
 304         except re.error as err:
 305             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 306     opts.remove_chapters = remove_chapters_patterns
 307
 308     # Cookies from browser
 309     if opts.cookiesfrombrowser:
 310         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 311         if mobj is None:
 312             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 313         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 314         browser_name = browser_name.lower()
 315         if browser_name not in SUPPORTED_BROWSERS:
 316             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 317                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 318         if keyring is not None:
 319             keyring = keyring.upper()
 320             if keyring not in SUPPORTED_KEYRINGS:
 321                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 322                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 323         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 324
 325     # MetadataParser
 326     def metadataparser_actions(f):
 327         if isinstance(f, str):
 328             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 329             try:
 330                 actions = [MetadataFromFieldPP.to_action(f)]
 331             except Exception as err:
 332                 raise ValueError(f'{cmd} is invalid; {err}')
 333         else:
 334             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 335             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 336
 337         for action in actions:
 338             try:
 339                 MetadataParserPP.validate_action(*action)
 340             except Exception as err:
 341                 raise ValueError(f'{cmd} is invalid; {err}')
 342             yield action
 343
 344     parse_metadata = opts.parse_metadata or []
 345     if opts.metafromtitle is not None:
 346         parse_metadata.append('title:%s' % opts.metafromtitle)
 347     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 348
 349     # Other options
 350     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 351     if geo_bypass_code is not None:
 352         try:
 353             GeoUtils.random_ipv4(geo_bypass_code)
 354         except Exception:
 355             raise ValueError('unsupported geo-bypass country or ip-block')
 356
 357     opts.match_filter = match_filter_func(opts.match_filter)
 358     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 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     # --(post-processor/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('--date-after', 'dateafter', '--date', 'date', default=None)
 407     report_conflict('--date-before', '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     if opts.exec_before_dl_cmd:
 447         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 448
 449     if opts.useid:  # --id is not deprecated in youtube-dl
 450         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 451
 452     if opts.overwrites:  # --force-overwrites implies --no-continue
 453         opts.continue_dl = False
 454
 455     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 456         # Add chapters when adding metadata or marking sponsors
 457         opts.addchapters = True
 458
 459     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 460         # Do not unnecessarily download audio
 461         opts.format = 'bestaudio/best'
 462
 463     if opts.getcomments and opts.writeinfojson is None:
 464         # If JSON is not printed anywhere, but comments are requested, save it to file
 465         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 466             opts.writeinfojson = True
 467
 468     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 469         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 470         opts.writesubtitles = True
 471
 472     if opts.addmetadata and opts.embed_infojson is None:
 473         # If embedding metadata and infojson is present, embed it
 474         opts.embed_infojson = 'if_exists'
 475
 476     # Ask for passwords
 477     if opts.username is not None and opts.password is None:
 478         opts.password = compat_getpass('Type account password and press [Return]: ')
 479     if opts.ap_username is not None and opts.ap_password is None:
 480         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 481
 482     return warnings, deprecation_warnings
 483
 484
 485 def get_postprocessors(opts):
 486     yield from opts.add_postprocessors
 487
 488     if opts.parse_metadata:
 489         yield {
 490             'key': 'MetadataParser',
 491             'actions': opts.parse_metadata,
 492             'when': 'pre_process'
 493         }
 494     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 495     if sponsorblock_query:
 496         yield {
 497             'key': 'SponsorBlock',
 498             'categories': sponsorblock_query,
 499             'api': opts.sponsorblock_api,
 500             'when': 'after_filter'
 501         }
 502     if opts.convertsubtitles:
 503         yield {
 504             'key': 'FFmpegSubtitlesConvertor',
 505             'format': opts.convertsubtitles,
 506             'when': 'before_dl'
 507         }
 508     if opts.convertthumbnails:
 509         yield {
 510             'key': 'FFmpegThumbnailsConvertor',
 511             'format': opts.convertthumbnails,
 512             'when': 'before_dl'
 513         }
 514     if opts.extractaudio:
 515         yield {
 516             'key': 'FFmpegExtractAudio',
 517             'preferredcodec': opts.audioformat,
 518             'preferredquality': opts.audioquality,
 519             'nopostoverwrites': opts.nopostoverwrites,
 520         }
 521     if opts.remuxvideo:
 522         yield {
 523             'key': 'FFmpegVideoRemuxer',
 524             'preferedformat': opts.remuxvideo,
 525         }
 526     if opts.recodevideo:
 527         yield {
 528             'key': 'FFmpegVideoConvertor',
 529             'preferedformat': opts.recodevideo,
 530         }
 531     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 532     if opts.embedsubtitles:
 533         keep_subs = 'no-keep-subs' not in opts.compat_opts
 534         yield {
 535             'key': 'FFmpegEmbedSubtitle',
 536             # already_have_subtitle = True prevents the file from being deleted after embedding
 537             'already_have_subtitle': opts.writesubtitles and keep_subs
 538         }
 539         if not opts.writeautomaticsub and keep_subs:
 540             opts.writesubtitles = True
 541
 542     # ModifyChapters must run before FFmpegMetadataPP
 543     if opts.remove_chapters or sponsorblock_query:
 544         yield {
 545             'key': 'ModifyChapters',
 546             'remove_chapters_patterns': opts.remove_chapters,
 547             'remove_sponsor_segments': opts.sponsorblock_remove,
 548             'remove_ranges': opts.remove_ranges,
 549             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 550             'force_keyframes': opts.force_keyframes_at_cuts
 551         }
 552     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 553     # FFmpegExtractAudioPP as containers before conversion may not support
 554     # metadata (3gp, webm, etc.)
 555     # By default ffmpeg preserves metadata applicable for both
 556     # source and target containers. From this point the container won't change,
 557     # so metadata can be added here.
 558     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 559         yield {
 560             'key': 'FFmpegMetadata',
 561             'add_chapters': opts.addchapters,
 562             'add_metadata': opts.addmetadata,
 563             'add_infojson': opts.embed_infojson,
 564         }
 565     # Deprecated
 566     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 567     # but must be below EmbedSubtitle and FFmpegMetadata
 568     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 569     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 570     if opts.sponskrub is not False:
 571         yield {
 572             'key': 'SponSkrub',
 573             'path': opts.sponskrub_path,
 574             'args': opts.sponskrub_args,
 575             'cut': opts.sponskrub_cut,
 576             'force': opts.sponskrub_force,
 577             'ignoreerror': opts.sponskrub is None,
 578             '_from_cli': True,
 579         }
 580     if opts.embedthumbnail:
 581         yield {
 582             'key': 'EmbedThumbnail',
 583             # already_have_thumbnail = True prevents the file from being deleted after embedding
 584             'already_have_thumbnail': opts.writethumbnail
 585         }
 586         if not opts.writethumbnail:
 587             opts.writethumbnail = True
 588             opts.outtmpl['pl_thumbnail'] = ''
 589     if opts.split_chapters:
 590         yield {
 591             'key': 'FFmpegSplitChapters',
 592             'force_keyframes': opts.force_keyframes_at_cuts,
 593         }
 594     # XAttrMetadataPP should be run after post-processors that may change file contents
 595     if opts.xattrs:
 596         yield {'key': 'XAttrMetadata'}
 597     if opts.concat_playlist != 'never':
 598         yield {
 599             'key': 'FFmpegConcat',
 600             'only_multi_video': opts.concat_playlist != 'always',
 601             'when': 'playlist',
 602         }
 603     # Exec must be the last PP of each category
 604     for when, exec_cmd in opts.exec_cmd.items():
 605         yield {
 606             'key': 'Exec',
 607             'exec_cmd': exec_cmd,
 608             'when': when,
 609         }
 610
 611
 612 def parse_options(argv=None):
 613     """ @returns (parser, opts, urls, ydl_opts) """
 614     parser, opts, urls = parseOpts(argv)
 615     urls = get_urls(urls, opts.batchfile, opts.verbose)
 616
 617     set_compat_opts(opts)
 618     try:
 619         warnings, deprecation_warnings = validate_options(opts)
 620     except ValueError as err:
 621         parser.error(f'{err}\n')
 622
 623     postprocessors = list(get_postprocessors(opts))
 624
 625     any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
 626                    or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
 627                    or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
 628
 629     any_printing = opts.print_json
 630
 631     final_ext = (
 632         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 633         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 634         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 635         else None)
 636
 637     return parser, opts, urls, {
 638         'usenetrc': opts.usenetrc,
 639         'netrc_location': opts.netrc_location,
 640         'username': opts.username,
 641         'password': opts.password,
 642         'twofactor': opts.twofactor,
 643         'videopassword': opts.videopassword,
 644         'ap_mso': opts.ap_mso,
 645         'ap_username': opts.ap_username,
 646         'ap_password': opts.ap_password,
 647         'quiet': (opts.quiet or any_getting or any_printing),
 648         'no_warnings': opts.no_warnings,
 649         'forceurl': opts.geturl,
 650         'forcetitle': opts.gettitle,
 651         'forceid': opts.getid,
 652         'forcethumbnail': opts.getthumbnail,
 653         'forcedescription': opts.getdescription,
 654         'forceduration': opts.getduration,
 655         'forcefilename': opts.getfilename,
 656         'forceformat': opts.getformat,
 657         'forceprint': opts.forceprint,
 658         'print_to_file': opts.print_to_file,
 659         'forcejson': opts.dumpjson or opts.print_json,
 660         'dump_single_json': opts.dump_single_json,
 661         'force_write_download_archive': opts.force_write_download_archive,
 662         'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
 663         'skip_download': opts.skip_download,
 664         'format': opts.format,
 665         'allow_unplayable_formats': opts.allow_unplayable_formats,
 666         'ignore_no_formats_error': opts.ignore_no_formats_error,
 667         'format_sort': opts.format_sort,
 668         'format_sort_force': opts.format_sort_force,
 669         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 670         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 671         'check_formats': opts.check_formats,
 672         'listformats': opts.listformats,
 673         'listformats_table': opts.listformats_table,
 674         'outtmpl': opts.outtmpl,
 675         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 676         'paths': opts.paths,
 677         'autonumber_size': opts.autonumber_size,
 678         'autonumber_start': opts.autonumber_start,
 679         'restrictfilenames': opts.restrictfilenames,
 680         'windowsfilenames': opts.windowsfilenames,
 681         'ignoreerrors': opts.ignoreerrors,
 682         'force_generic_extractor': opts.force_generic_extractor,
 683         'ratelimit': opts.ratelimit,
 684         'throttledratelimit': opts.throttledratelimit,
 685         'overwrites': opts.overwrites,
 686         'retries': opts.retries,
 687         'file_access_retries': opts.file_access_retries,
 688         'fragment_retries': opts.fragment_retries,
 689         'extractor_retries': opts.extractor_retries,
 690         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 691         'keep_fragments': opts.keep_fragments,
 692         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 693         'buffersize': opts.buffersize,
 694         'noresizebuffer': opts.noresizebuffer,
 695         'http_chunk_size': opts.http_chunk_size,
 696         'continuedl': opts.continue_dl,
 697         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 698         'progress_with_newline': opts.progress_with_newline,
 699         'progress_template': opts.progress_template,
 700         'playliststart': opts.playliststart,
 701         'playlistend': opts.playlistend,
 702         'playlistreverse': opts.playlist_reverse,
 703         'playlistrandom': opts.playlist_random,
 704         'noplaylist': opts.noplaylist,
 705         'logtostderr': opts.outtmpl.get('default') == '-',
 706         'consoletitle': opts.consoletitle,
 707         'nopart': opts.nopart,
 708         'updatetime': opts.updatetime,
 709         'writedescription': opts.writedescription,
 710         'writeannotations': opts.writeannotations,
 711         'writeinfojson': opts.writeinfojson,
 712         'allow_playlist_files': opts.allow_playlist_files,
 713         'clean_infojson': opts.clean_infojson,
 714         'getcomments': opts.getcomments,
 715         'writethumbnail': opts.writethumbnail is True,
 716         'write_all_thumbnails': opts.writethumbnail == 'all',
 717         'writelink': opts.writelink,
 718         'writeurllink': opts.writeurllink,
 719         'writewebloclink': opts.writewebloclink,
 720         'writedesktoplink': opts.writedesktoplink,
 721         'writesubtitles': opts.writesubtitles,
 722         'writeautomaticsub': opts.writeautomaticsub,
 723         'allsubtitles': opts.allsubtitles,
 724         'listsubtitles': opts.listsubtitles,
 725         'subtitlesformat': opts.subtitlesformat,
 726         'subtitleslangs': opts.subtitleslangs,
 727         'matchtitle': decodeOption(opts.matchtitle),
 728         'rejecttitle': decodeOption(opts.rejecttitle),
 729         'max_downloads': opts.max_downloads,
 730         'prefer_free_formats': opts.prefer_free_formats,
 731         'trim_file_name': opts.trim_file_name,
 732         'verbose': opts.verbose,
 733         'dump_intermediate_pages': opts.dump_intermediate_pages,
 734         'write_pages': opts.write_pages,
 735         'test': opts.test,
 736         'keepvideo': opts.keepvideo,
 737         'min_filesize': opts.min_filesize,
 738         'max_filesize': opts.max_filesize,
 739         'min_views': opts.min_views,
 740         'max_views': opts.max_views,
 741         'daterange': opts.date,
 742         'cachedir': opts.cachedir,
 743         'youtube_print_sig_code': opts.youtube_print_sig_code,
 744         'age_limit': opts.age_limit,
 745         'download_archive': opts.download_archive,
 746         'break_on_existing': opts.break_on_existing,
 747         'break_on_reject': opts.break_on_reject,
 748         'break_per_url': opts.break_per_url,
 749         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 750         'cookiefile': opts.cookiefile,
 751         'cookiesfrombrowser': opts.cookiesfrombrowser,
 752         'legacyserverconnect': opts.legacy_server_connect,
 753         'nocheckcertificate': opts.no_check_certificate,
 754         'prefer_insecure': opts.prefer_insecure,
 755         'http_headers': opts.headers,
 756         'proxy': opts.proxy,
 757         'socket_timeout': opts.socket_timeout,
 758         'bidi_workaround': opts.bidi_workaround,
 759         'debug_printtraffic': opts.debug_printtraffic,
 760         'prefer_ffmpeg': opts.prefer_ffmpeg,
 761         'include_ads': opts.include_ads,
 762         'default_search': opts.default_search,
 763         'dynamic_mpd': opts.dynamic_mpd,
 764         'extractor_args': opts.extractor_args,
 765         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 766         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 767         'encoding': opts.encoding,
 768         'extract_flat': opts.extract_flat,
 769         'live_from_start': opts.live_from_start,
 770         'wait_for_video': opts.wait_for_video,
 771         'mark_watched': opts.mark_watched,
 772         'merge_output_format': opts.merge_output_format,
 773         'final_ext': final_ext,
 774         'postprocessors': postprocessors,
 775         'fixup': opts.fixup,
 776         'source_address': opts.source_address,
 777         'call_home': opts.call_home,
 778         'sleep_interval_requests': opts.sleep_interval_requests,
 779         'sleep_interval': opts.sleep_interval,
 780         'max_sleep_interval': opts.max_sleep_interval,
 781         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 782         'external_downloader': opts.external_downloader,
 783         'list_thumbnails': opts.list_thumbnails,
 784         'playlist_items': opts.playlist_items,
 785         'xattr_set_filesize': opts.xattr_set_filesize,
 786         'match_filter': opts.match_filter,
 787         'no_color': opts.no_color,
 788         'ffmpeg_location': opts.ffmpeg_location,
 789         'hls_prefer_native': opts.hls_prefer_native,
 790         'hls_use_mpegts': opts.hls_use_mpegts,
 791         'hls_split_discontinuity': opts.hls_split_discontinuity,
 792         'external_downloader_args': opts.external_downloader_args,
 793         'postprocessor_args': opts.postprocessor_args,
 794         'cn_verification_proxy': opts.cn_verification_proxy,
 795         'geo_verification_proxy': opts.geo_verification_proxy,
 796         'geo_bypass': opts.geo_bypass,
 797         'geo_bypass_country': opts.geo_bypass_country,
 798         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 799         '_warnings': warnings,
 800         '_deprecation_warnings': deprecation_warnings,
 801         'compat_opts': opts.compat_opts,
 802     }
 803
 804
 805 def _real_main(argv=None):
 806     # Compatibility fixes for Windows
 807     if sys.platform == 'win32':
 808         # https://github.com/ytdl-org/youtube-dl/issues/820
 809         codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
 810
 811     workaround_optparse_bug9161()
 812
 813     setproctitle('yt-dlp')
 814
 815     parser, opts, all_urls, ydl_opts = parse_options(argv)
 816
 817     # Dump user agent
 818     if opts.dump_user_agent:
 819         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 820         write_string(f'{ua}\n', out=sys.stdout)
 821
 822     if print_extractor_information(opts, all_urls):
 823         sys.exit(0)
 824
 825     with YoutubeDL(ydl_opts) as ydl:
 826         actual_use = all_urls or opts.load_info_filename
 827
 828         # Remove cache dir
 829         if opts.rm_cachedir:
 830             ydl.cache.remove()
 831
 832         # Update version
 833         if opts.update_self:
 834             # If updater returns True, exit. Required for windows
 835             if run_update(ydl):
 836                 if actual_use:
 837                     sys.exit('ERROR: The program must exit for the update to complete')
 838                 sys.exit()
 839
 840         # Maybe do nothing
 841         if not actual_use:
 842             if opts.update_self or opts.rm_cachedir:
 843                 sys.exit()
 844
 845             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 846             parser.error(
 847                 'You must provide at least one URL.\n'
 848                 'Type yt-dlp --help to see a list of all options.')
 849
 850         try:
 851             if opts.load_info_filename is not None:
 852                 retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
 853             else:
 854                 retcode = ydl.download(all_urls)
 855         except DownloadCancelled:
 856             ydl.to_screen('Aborting remaining downloads')
 857             retcode = 101
 858
 859     sys.exit(retcode)
 860
 861
 862 def main(argv=None):
 863     try:
 864         _real_main(argv)
 865     except DownloadError:
 866         sys.exit(1)
 867     except SameFileError as e:
 868         sys.exit(f'ERROR: {e}')
 869     except KeyboardInterrupt:
 870         sys.exit('\nERROR: Interrupted by user')
 871     except BrokenPipeError as e:
 872         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 873         devnull = os.open(os.devnull, os.O_WRONLY)
 874         os.dup2(devnull, sys.stdout.fileno())
 875         sys.exit(f'\nERROR: {e}')
 876
 877
 878 __all__ = [
 879     'main',
 880     'YoutubeDL',
 881     'parse_options',
 882     'gen_extractors',
 883     'list_extractors',
 884 ]