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.sleep_interval is None:
 202         validate(
 203             opts.max_sleep_interval is None, 'min sleep interval',
 204             msg='{name} must be specified; use --min-sleep-interval')
 205     elif opts.max_sleep_interval is None:
 206         opts.max_sleep_interval = opts.sleep_interval
 207     else:
 208         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 209
 210     if opts.wait_for_video is not None:
 211         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 212         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 213                  'time range to wait for video', opts.wait_for_video)
 214         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 215         opts.wait_for_video = (min_wait, max_wait)
 216
 217     # Format sort
 218     for f in opts.format_sort:
 219         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 220
 221     # Postprocessor formats
 222     validate_in('audio format', opts.audioformat, ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS))
 223     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 224     validate_in('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS)
 225     if opts.recodevideo is not None:
 226         opts.recodevideo = opts.recodevideo.replace(' ', '')
 227         validate_regex('video recode format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 228     if opts.remuxvideo is not None:
 229         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 230         validate_regex('video remux format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 231     if opts.audioquality:
 232         opts.audioquality = opts.audioquality.strip('k').strip('K')
 233         # int_or_none prevents inf, nan
 234         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 235
 236     # Retries
 237     def parse_retries(name, value):
 238         if value is None:
 239             return None
 240         elif value in ('inf', 'infinite'):
 241             return float('inf')
 242         try:
 243             return int(value)
 244         except (TypeError, ValueError):
 245             validate(False, f'{name} retry count', value)
 246
 247     opts.retries = parse_retries('download', opts.retries)
 248     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 249     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 250     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 251
 252     # Bytes
 253     def parse_bytes(name, value):
 254         if value is None:
 255             return None
 256         numeric_limit = FileDownloader.parse_bytes(value)
 257         validate(numeric_limit is not None, 'rate limit', value)
 258         return numeric_limit
 259
 260     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 261     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 262     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 263     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 264     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 265     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 266
 267     # Output templates
 268     def validate_outtmpl(tmpl, msg):
 269         err = YoutubeDL.validate_outtmpl(tmpl)
 270         if err:
 271             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 272
 273     for k, tmpl in opts.outtmpl.items():
 274         validate_outtmpl(tmpl, f'{k} output template')
 275     for type_, tmpl_list in opts.forceprint.items():
 276         for tmpl in tmpl_list:
 277             validate_outtmpl(tmpl, f'{type_} print template')
 278     for type_, tmpl_list in opts.print_to_file.items():
 279         for tmpl, file in tmpl_list:
 280             validate_outtmpl(tmpl, f'{type_} print to file template')
 281             validate_outtmpl(file, f'{type_} print to file filename')
 282     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 283     for k, tmpl in opts.progress_template.items():
 284         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 285         validate_outtmpl(tmpl, f'{k} template')
 286
 287     outtmpl_default = opts.outtmpl.get('default')
 288     if outtmpl_default == '':
 289         opts.skip_download = None
 290         del opts.outtmpl['default']
 291     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 292         raise ValueError(
 293             'Cannot download a video and extract audio into the same file! '
 294             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 295
 296     # Remove chapters
 297     remove_chapters_patterns, opts.remove_ranges = [], []
 298     for regex in opts.remove_chapters or []:
 299         if regex.startswith('*'):
 300             dur = list(map(parse_duration, regex[1:].split('-')))
 301             if len(dur) == 2 and all(t is not None for t in dur):
 302                 opts.remove_ranges.append(tuple(dur))
 303                 continue
 304             raise ValueError(f'invalid --remove-chapters time range "{regex}". Must be of the form *start-end')
 305         try:
 306             remove_chapters_patterns.append(re.compile(regex))
 307         except re.error as err:
 308             raise ValueError(f'invalid --remove-chapters regex "{regex}" - {err}')
 309     opts.remove_chapters = remove_chapters_patterns
 310
 311     # Cookies from browser
 312     if opts.cookiesfrombrowser:
 313         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 314         if mobj is None:
 315             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 316         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 317         browser_name = browser_name.lower()
 318         if browser_name not in SUPPORTED_BROWSERS:
 319             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 320                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 321         if keyring is not None:
 322             keyring = keyring.upper()
 323             if keyring not in SUPPORTED_KEYRINGS:
 324                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 325                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 326         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 327
 328     # MetadataParser
 329     def metadataparser_actions(f):
 330         if isinstance(f, str):
 331             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 332             try:
 333                 actions = [MetadataFromFieldPP.to_action(f)]
 334             except Exception as err:
 335                 raise ValueError(f'{cmd} is invalid; {err}')
 336         else:
 337             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 338             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 339
 340         for action in actions:
 341             try:
 342                 MetadataParserPP.validate_action(*action)
 343             except Exception as err:
 344                 raise ValueError(f'{cmd} is invalid; {err}')
 345             yield action
 346
 347     parse_metadata = opts.parse_metadata or []
 348     if opts.metafromtitle is not None:
 349         parse_metadata.append('title:%s' % opts.metafromtitle)
 350     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 351
 352     # Other options
 353     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 354     if geo_bypass_code is not None:
 355         try:
 356             GeoUtils.random_ipv4(geo_bypass_code)
 357         except Exception:
 358             raise ValueError('unsupported geo-bypass country or ip-block')
 359
 360     opts.match_filter = match_filter_func(opts.match_filter)
 361     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 362
 363     if opts.download_archive is not None:
 364         opts.download_archive = expand_path(opts.download_archive)
 365
 366     if opts.user_agent is not None:
 367         opts.headers.setdefault('User-Agent', opts.user_agent)
 368     if opts.referer is not None:
 369         opts.headers.setdefault('Referer', opts.referer)
 370
 371     if opts.no_sponsorblock:
 372         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 373
 374     warnings, deprecation_warnings = [], []
 375
 376     # Common mistake: -f best
 377     if opts.format == 'best':
 378         warnings.append('.\n         '.join((
 379             '"-f best" selects the best pre-merged format which is often not the best option',
 380             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 381             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 382
 383     # --(post-processor/downloader)-args without name
 384     def report_args_compat(name, value, key1, key2=None):
 385         if key1 in value and key2 not in value:
 386             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 387             return True
 388         return False
 389
 390     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 391     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 392         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 393         opts.postprocessor_args.setdefault('sponskrub', [])
 394
 395     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 396                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 397         if val2 is NO_DEFAULT:
 398             val2 = getattr(opts, opt2)
 399         if not val2:
 400             return
 401
 402         if val1 is NO_DEFAULT:
 403             val1 = getattr(opts, opt1)
 404         if val1:
 405             warnings.append(f'{arg1} is ignored since {arg2} was given')
 406         setattr(opts, opt1, default)
 407
 408     # Conflicting options
 409     report_conflict('--date-after', 'dateafter', '--date', 'date', default=None)
 410     report_conflict('--date-before', 'datebefore', '--date', 'date', default=None)
 411     report_conflict('--exec-before-download', 'exec_before_dl_cmd', '"--exec before_dl:"', 'exec_cmd', opts.exec_cmd.get('before_dl'))
 412     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 413     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 414     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 415     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 416     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 417     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters', val1=opts.sponskrub and opts.sponskrub_cut)
 418
 419     # Conflicts with --allow-unplayable-formats
 420     report_conflict('--add-metadata', 'addmetadata')
 421     report_conflict('--embed-chapters', 'addchapters')
 422     report_conflict('--embed-info-json', 'embed_infojson')
 423     report_conflict('--embed-subs', 'embedsubtitles')
 424     report_conflict('--embed-thumbnail', 'embedthumbnail')
 425     report_conflict('--extract-audio', 'extractaudio')
 426     report_conflict('--fixup', 'fixup', val1=(opts.fixup or '').lower() in ('', 'never', 'ignore'), default='never')
 427     report_conflict('--recode-video', 'recodevideo')
 428     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 429     report_conflict('--remux-video', 'remuxvideo')
 430     report_conflict('--sponskrub', 'sponskrub')
 431     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 432     report_conflict('--xattrs', 'xattrs')
 433
 434     # Fully deprecated options
 435     def report_deprecation(val, old, new=None):
 436         if not val:
 437             return
 438         deprecation_warnings.append(
 439             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 440             else f'{old} is deprecated and may not work as expected')
 441
 442     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 443     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 444     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 445     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 446     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 447
 448     # Dependent options
 449     if opts.exec_before_dl_cmd:
 450         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 451
 452     if opts.useid:  # --id is not deprecated in youtube-dl
 453         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 454
 455     if opts.overwrites:  # --force-overwrites implies --no-continue
 456         opts.continue_dl = False
 457
 458     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 459         # Add chapters when adding metadata or marking sponsors
 460         opts.addchapters = True
 461
 462     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 463         # Do not unnecessarily download audio
 464         opts.format = 'bestaudio/best'
 465
 466     if opts.getcomments and opts.writeinfojson is None:
 467         # If JSON is not printed anywhere, but comments are requested, save it to file
 468         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 469             opts.writeinfojson = True
 470
 471     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 472         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 473         opts.writesubtitles = True
 474
 475     if opts.addmetadata and opts.embed_infojson is None:
 476         # If embedding metadata and infojson is present, embed it
 477         opts.embed_infojson = 'if_exists'
 478
 479     # Ask for passwords
 480     if opts.username is not None and opts.password is None:
 481         opts.password = compat_getpass('Type account password and press [Return]: ')
 482     if opts.ap_username is not None and opts.ap_password is None:
 483         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 484
 485     return warnings, deprecation_warnings
 486
 487
 488 def get_postprocessors(opts):
 489     yield from opts.add_postprocessors
 490
 491     if opts.parse_metadata:
 492         yield {
 493             'key': 'MetadataParser',
 494             'actions': opts.parse_metadata,
 495             'when': 'pre_process'
 496         }
 497     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 498     if sponsorblock_query:
 499         yield {
 500             'key': 'SponsorBlock',
 501             'categories': sponsorblock_query,
 502             'api': opts.sponsorblock_api,
 503             'when': 'after_filter'
 504         }
 505     if opts.convertsubtitles:
 506         yield {
 507             'key': 'FFmpegSubtitlesConvertor',
 508             'format': opts.convertsubtitles,
 509             'when': 'before_dl'
 510         }
 511     if opts.convertthumbnails:
 512         yield {
 513             'key': 'FFmpegThumbnailsConvertor',
 514             'format': opts.convertthumbnails,
 515             'when': 'before_dl'
 516         }
 517     if opts.extractaudio:
 518         yield {
 519             'key': 'FFmpegExtractAudio',
 520             'preferredcodec': opts.audioformat,
 521             'preferredquality': opts.audioquality,
 522             'nopostoverwrites': opts.nopostoverwrites,
 523         }
 524     if opts.remuxvideo:
 525         yield {
 526             'key': 'FFmpegVideoRemuxer',
 527             'preferedformat': opts.remuxvideo,
 528         }
 529     if opts.recodevideo:
 530         yield {
 531             'key': 'FFmpegVideoConvertor',
 532             'preferedformat': opts.recodevideo,
 533         }
 534     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 535     if opts.embedsubtitles:
 536         keep_subs = 'no-keep-subs' not in opts.compat_opts
 537         yield {
 538             'key': 'FFmpegEmbedSubtitle',
 539             # already_have_subtitle = True prevents the file from being deleted after embedding
 540             'already_have_subtitle': opts.writesubtitles and keep_subs
 541         }
 542         if not opts.writeautomaticsub and keep_subs:
 543             opts.writesubtitles = True
 544
 545     # ModifyChapters must run before FFmpegMetadataPP
 546     if opts.remove_chapters or sponsorblock_query:
 547         yield {
 548             'key': 'ModifyChapters',
 549             'remove_chapters_patterns': opts.remove_chapters,
 550             'remove_sponsor_segments': opts.sponsorblock_remove,
 551             'remove_ranges': opts.remove_ranges,
 552             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 553             'force_keyframes': opts.force_keyframes_at_cuts
 554         }
 555     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 556     # FFmpegExtractAudioPP as containers before conversion may not support
 557     # metadata (3gp, webm, etc.)
 558     # By default ffmpeg preserves metadata applicable for both
 559     # source and target containers. From this point the container won't change,
 560     # so metadata can be added here.
 561     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 562         yield {
 563             'key': 'FFmpegMetadata',
 564             'add_chapters': opts.addchapters,
 565             'add_metadata': opts.addmetadata,
 566             'add_infojson': opts.embed_infojson,
 567         }
 568     # Deprecated
 569     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 570     # but must be below EmbedSubtitle and FFmpegMetadata
 571     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 572     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 573     if opts.sponskrub is not False:
 574         yield {
 575             'key': 'SponSkrub',
 576             'path': opts.sponskrub_path,
 577             'args': opts.sponskrub_args,
 578             'cut': opts.sponskrub_cut,
 579             'force': opts.sponskrub_force,
 580             'ignoreerror': opts.sponskrub is None,
 581             '_from_cli': True,
 582         }
 583     if opts.embedthumbnail:
 584         yield {
 585             'key': 'EmbedThumbnail',
 586             # already_have_thumbnail = True prevents the file from being deleted after embedding
 587             'already_have_thumbnail': opts.writethumbnail
 588         }
 589         if not opts.writethumbnail:
 590             opts.writethumbnail = True
 591             opts.outtmpl['pl_thumbnail'] = ''
 592     if opts.split_chapters:
 593         yield {
 594             'key': 'FFmpegSplitChapters',
 595             'force_keyframes': opts.force_keyframes_at_cuts,
 596         }
 597     # XAttrMetadataPP should be run after post-processors that may change file contents
 598     if opts.xattrs:
 599         yield {'key': 'XAttrMetadata'}
 600     if opts.concat_playlist != 'never':
 601         yield {
 602             'key': 'FFmpegConcat',
 603             'only_multi_video': opts.concat_playlist != 'always',
 604             'when': 'playlist',
 605         }
 606     # Exec must be the last PP of each category
 607     for when, exec_cmd in opts.exec_cmd.items():
 608         yield {
 609             'key': 'Exec',
 610             'exec_cmd': exec_cmd,
 611             'when': when,
 612         }
 613
 614
 615 def parse_options(argv=None):
 616     """ @returns (parser, opts, urls, ydl_opts) """
 617     parser, opts, urls = parseOpts(argv)
 618     urls = get_urls(urls, opts.batchfile, opts.verbose)
 619
 620     set_compat_opts(opts)
 621     try:
 622         warnings, deprecation_warnings = validate_options(opts)
 623     except ValueError as err:
 624         parser.error(f'{err}\n')
 625
 626     postprocessors = list(get_postprocessors(opts))
 627
 628     any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
 629                    or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
 630                    or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
 631
 632     any_printing = opts.print_json
 633
 634     final_ext = (
 635         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 636         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 637         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 638         else None)
 639
 640     return parser, opts, urls, {
 641         'usenetrc': opts.usenetrc,
 642         'netrc_location': opts.netrc_location,
 643         'username': opts.username,
 644         'password': opts.password,
 645         'twofactor': opts.twofactor,
 646         'videopassword': opts.videopassword,
 647         'ap_mso': opts.ap_mso,
 648         'ap_username': opts.ap_username,
 649         'ap_password': opts.ap_password,
 650         'quiet': (opts.quiet or any_getting or any_printing),
 651         'no_warnings': opts.no_warnings,
 652         'forceurl': opts.geturl,
 653         'forcetitle': opts.gettitle,
 654         'forceid': opts.getid,
 655         'forcethumbnail': opts.getthumbnail,
 656         'forcedescription': opts.getdescription,
 657         'forceduration': opts.getduration,
 658         'forcefilename': opts.getfilename,
 659         'forceformat': opts.getformat,
 660         'forceprint': opts.forceprint,
 661         'print_to_file': opts.print_to_file,
 662         'forcejson': opts.dumpjson or opts.print_json,
 663         'dump_single_json': opts.dump_single_json,
 664         'force_write_download_archive': opts.force_write_download_archive,
 665         'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
 666         'skip_download': opts.skip_download,
 667         'format': opts.format,
 668         'allow_unplayable_formats': opts.allow_unplayable_formats,
 669         'ignore_no_formats_error': opts.ignore_no_formats_error,
 670         'format_sort': opts.format_sort,
 671         'format_sort_force': opts.format_sort_force,
 672         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 673         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 674         'check_formats': opts.check_formats,
 675         'listformats': opts.listformats,
 676         'listformats_table': opts.listformats_table,
 677         'outtmpl': opts.outtmpl,
 678         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 679         'paths': opts.paths,
 680         'autonumber_size': opts.autonumber_size,
 681         'autonumber_start': opts.autonumber_start,
 682         'restrictfilenames': opts.restrictfilenames,
 683         'windowsfilenames': opts.windowsfilenames,
 684         'ignoreerrors': opts.ignoreerrors,
 685         'force_generic_extractor': opts.force_generic_extractor,
 686         'ratelimit': opts.ratelimit,
 687         'throttledratelimit': opts.throttledratelimit,
 688         'overwrites': opts.overwrites,
 689         'retries': opts.retries,
 690         'file_access_retries': opts.file_access_retries,
 691         'fragment_retries': opts.fragment_retries,
 692         'extractor_retries': opts.extractor_retries,
 693         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 694         'keep_fragments': opts.keep_fragments,
 695         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 696         'buffersize': opts.buffersize,
 697         'noresizebuffer': opts.noresizebuffer,
 698         'http_chunk_size': opts.http_chunk_size,
 699         'continuedl': opts.continue_dl,
 700         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 701         'progress_with_newline': opts.progress_with_newline,
 702         'progress_template': opts.progress_template,
 703         'playliststart': opts.playliststart,
 704         'playlistend': opts.playlistend,
 705         'playlistreverse': opts.playlist_reverse,
 706         'playlistrandom': opts.playlist_random,
 707         'noplaylist': opts.noplaylist,
 708         'logtostderr': opts.outtmpl.get('default') == '-',
 709         'consoletitle': opts.consoletitle,
 710         'nopart': opts.nopart,
 711         'updatetime': opts.updatetime,
 712         'writedescription': opts.writedescription,
 713         'writeannotations': opts.writeannotations,
 714         'writeinfojson': opts.writeinfojson,
 715         'allow_playlist_files': opts.allow_playlist_files,
 716         'clean_infojson': opts.clean_infojson,
 717         'getcomments': opts.getcomments,
 718         'writethumbnail': opts.writethumbnail is True,
 719         'write_all_thumbnails': opts.writethumbnail == 'all',
 720         'writelink': opts.writelink,
 721         'writeurllink': opts.writeurllink,
 722         'writewebloclink': opts.writewebloclink,
 723         'writedesktoplink': opts.writedesktoplink,
 724         'writesubtitles': opts.writesubtitles,
 725         'writeautomaticsub': opts.writeautomaticsub,
 726         'allsubtitles': opts.allsubtitles,
 727         'listsubtitles': opts.listsubtitles,
 728         'subtitlesformat': opts.subtitlesformat,
 729         'subtitleslangs': opts.subtitleslangs,
 730         'matchtitle': decodeOption(opts.matchtitle),
 731         'rejecttitle': decodeOption(opts.rejecttitle),
 732         'max_downloads': opts.max_downloads,
 733         'prefer_free_formats': opts.prefer_free_formats,
 734         'trim_file_name': opts.trim_file_name,
 735         'verbose': opts.verbose,
 736         'dump_intermediate_pages': opts.dump_intermediate_pages,
 737         'write_pages': opts.write_pages,
 738         'test': opts.test,
 739         'keepvideo': opts.keepvideo,
 740         'min_filesize': opts.min_filesize,
 741         'max_filesize': opts.max_filesize,
 742         'min_views': opts.min_views,
 743         'max_views': opts.max_views,
 744         'daterange': opts.date,
 745         'cachedir': opts.cachedir,
 746         'youtube_print_sig_code': opts.youtube_print_sig_code,
 747         'age_limit': opts.age_limit,
 748         'download_archive': opts.download_archive,
 749         'break_on_existing': opts.break_on_existing,
 750         'break_on_reject': opts.break_on_reject,
 751         'break_per_url': opts.break_per_url,
 752         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 753         'cookiefile': opts.cookiefile,
 754         'cookiesfrombrowser': opts.cookiesfrombrowser,
 755         'legacyserverconnect': opts.legacy_server_connect,
 756         'nocheckcertificate': opts.no_check_certificate,
 757         'prefer_insecure': opts.prefer_insecure,
 758         'http_headers': opts.headers,
 759         'proxy': opts.proxy,
 760         'socket_timeout': opts.socket_timeout,
 761         'bidi_workaround': opts.bidi_workaround,
 762         'debug_printtraffic': opts.debug_printtraffic,
 763         'prefer_ffmpeg': opts.prefer_ffmpeg,
 764         'include_ads': opts.include_ads,
 765         'default_search': opts.default_search,
 766         'dynamic_mpd': opts.dynamic_mpd,
 767         'extractor_args': opts.extractor_args,
 768         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 769         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 770         'encoding': opts.encoding,
 771         'extract_flat': opts.extract_flat,
 772         'live_from_start': opts.live_from_start,
 773         'wait_for_video': opts.wait_for_video,
 774         'mark_watched': opts.mark_watched,
 775         'merge_output_format': opts.merge_output_format,
 776         'final_ext': final_ext,
 777         'postprocessors': postprocessors,
 778         'fixup': opts.fixup,
 779         'source_address': opts.source_address,
 780         'call_home': opts.call_home,
 781         'sleep_interval_requests': opts.sleep_interval_requests,
 782         'sleep_interval': opts.sleep_interval,
 783         'max_sleep_interval': opts.max_sleep_interval,
 784         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 785         'external_downloader': opts.external_downloader,
 786         'list_thumbnails': opts.list_thumbnails,
 787         'playlist_items': opts.playlist_items,
 788         'xattr_set_filesize': opts.xattr_set_filesize,
 789         'match_filter': opts.match_filter,
 790         'no_color': opts.no_color,
 791         'ffmpeg_location': opts.ffmpeg_location,
 792         'hls_prefer_native': opts.hls_prefer_native,
 793         'hls_use_mpegts': opts.hls_use_mpegts,
 794         'hls_split_discontinuity': opts.hls_split_discontinuity,
 795         'external_downloader_args': opts.external_downloader_args,
 796         'postprocessor_args': opts.postprocessor_args,
 797         'cn_verification_proxy': opts.cn_verification_proxy,
 798         'geo_verification_proxy': opts.geo_verification_proxy,
 799         'geo_bypass': opts.geo_bypass,
 800         'geo_bypass_country': opts.geo_bypass_country,
 801         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 802         '_warnings': warnings,
 803         '_deprecation_warnings': deprecation_warnings,
 804         'compat_opts': opts.compat_opts,
 805     }
 806
 807
 808 def _real_main(argv=None):
 809     # Compatibility fixes for Windows
 810     if sys.platform == 'win32':
 811         # https://github.com/ytdl-org/youtube-dl/issues/820
 812         codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
 813
 814     workaround_optparse_bug9161()
 815
 816     setproctitle('yt-dlp')
 817
 818     parser, opts, all_urls, ydl_opts = parse_options(argv)
 819
 820     # Dump user agent
 821     if opts.dump_user_agent:
 822         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 823         write_string(f'{ua}\n', out=sys.stdout)
 824
 825     if print_extractor_information(opts, all_urls):
 826         sys.exit(0)
 827
 828     with YoutubeDL(ydl_opts) as ydl:
 829         actual_use = all_urls or opts.load_info_filename
 830
 831         # Remove cache dir
 832         if opts.rm_cachedir:
 833             ydl.cache.remove()
 834
 835         # Update version
 836         if opts.update_self:
 837             # If updater returns True, exit. Required for windows
 838             if run_update(ydl):
 839                 if actual_use:
 840                     sys.exit('ERROR: The program must exit for the update to complete')
 841                 sys.exit()
 842
 843         # Maybe do nothing
 844         if not actual_use:
 845             if opts.update_self or opts.rm_cachedir:
 846                 sys.exit()
 847
 848             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 849             parser.error(
 850                 'You must provide at least one URL.\n'
 851                 'Type yt-dlp --help to see a list of all options.')
 852
 853         try:
 854             if opts.load_info_filename is not None:
 855                 retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
 856             else:
 857                 retcode = ydl.download(all_urls)
 858         except DownloadCancelled:
 859             ydl.to_screen('Aborting remaining downloads')
 860             retcode = 101
 861
 862     sys.exit(retcode)
 863
 864
 865 def main(argv=None):
 866     try:
 867         _real_main(argv)
 868     except DownloadError:
 869         sys.exit(1)
 870     except SameFileError as e:
 871         sys.exit(f'ERROR: {e}')
 872     except KeyboardInterrupt:
 873         sys.exit('\nERROR: Interrupted by user')
 874     except BrokenPipeError as e:
 875         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 876         devnull = os.open(os.devnull, os.O_WRONLY)
 877         os.dup2(devnull, sys.stdout.fileno())
 878         sys.exit(f'\nERROR: {e}')
 879
 880
 881 __all__ = [
 882     'main',
 883     'YoutubeDL',
 884     'parse_options',
 885     'gen_extractors',
 886     'list_extractors',
 887 ]