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 (
  17     parseOpts,
  18 )
  19 from .compat import (
  20     compat_getpass,
  21     compat_os_name,
  22     compat_shlex_quote,
  23     workaround_optparse_bug9161,
  24 )
  25 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  26 from .utils import (
  27     DateRange,
  28     decodeOption,
  29     DownloadCancelled,
  30     DownloadError,
  31     error_to_compat_str,
  32     expand_path,
  33     GeoUtils,
  34     float_or_none,
  35     int_or_none,
  36     match_filter_func,
  37     parse_duration,
  38     preferredencoding,
  39     read_batch_urls,
  40     render_table,
  41     SameFileError,
  42     setproctitle,
  43     std_headers,
  44     write_string,
  45 )
  46 from .update import run_update
  47 from .downloader import (
  48     FileDownloader,
  49 )
  50 from .extractor import gen_extractors, list_extractors
  51 from .extractor.common import InfoExtractor
  52 from .extractor.adobepass import MSO_INFO
  53 from .postprocessor import (
  54     FFmpegExtractAudioPP,
  55     FFmpegSubtitlesConvertorPP,
  56     FFmpegThumbnailsConvertorPP,
  57     FFmpegVideoConvertorPP,
  58     FFmpegVideoRemuxerPP,
  59     MetadataFromFieldPP,
  60     MetadataParserPP,
  61 )
  62 from .YoutubeDL import YoutubeDL
  63
  64
  65 def _real_main(argv=None):
  66     # Compatibility fixes for Windows
  67     if sys.platform == 'win32':
  68         # https://github.com/ytdl-org/youtube-dl/issues/820
  69         codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
  70
  71     workaround_optparse_bug9161()
  72
  73     setproctitle('yt-dlp')
  74
  75     parser, opts, args = parseOpts(argv)
  76     warnings, deprecation_warnings = [], []
  77
  78     # Set user agent
  79     if opts.user_agent is not None:
  80         std_headers['User-Agent'] = opts.user_agent
  81
  82     # Set referer
  83     if opts.referer is not None:
  84         std_headers['Referer'] = opts.referer
  85
  86     # Custom HTTP headers
  87     std_headers.update(opts.headers)
  88
  89     # Dump user agent
  90     if opts.dump_user_agent:
  91         write_string(std_headers['User-Agent'] + '\n', out=sys.stdout)
  92         sys.exit(0)
  93
  94     # Batch file verification
  95     batch_urls = []
  96     if opts.batchfile is not None:
  97         try:
  98             if opts.batchfile == '-':
  99                 write_string('Reading URLs from stdin - EOF (%s) to end:\n' % (
 100                     'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D'))
 101                 batchfd = sys.stdin
 102             else:
 103                 batchfd = io.open(
 104                     expand_path(opts.batchfile),
 105                     'r', encoding='utf-8', errors='ignore')
 106             batch_urls = read_batch_urls(batchfd)
 107             if opts.verbose:
 108                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
 109         except IOError:
 110             sys.exit('ERROR: batch file %s could not be read' % opts.batchfile)
 111     all_urls = batch_urls + [url.strip() for url in args]  # batch_urls are already striped in read_batch_urls
 112     _enc = preferredencoding()
 113     all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
 114
 115     if opts.list_extractors:
 116         for ie in list_extractors(opts.age_limit):
 117             write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n', out=sys.stdout)
 118             matchedUrls = [url for url in all_urls if ie.suitable(url)]
 119             for mu in matchedUrls:
 120                 write_string('  ' + mu + '\n', out=sys.stdout)
 121         sys.exit(0)
 122     if opts.list_extractor_descriptions:
 123         for ie in list_extractors(opts.age_limit):
 124             if not ie.working():
 125                 continue
 126             desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
 127             if desc is False:
 128                 continue
 129             if getattr(ie, 'SEARCH_KEY', None) is not None:
 130                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
 131                 _COUNTS = ('', '5', '10', 'all')
 132                 desc += f'; "{ie.SEARCH_KEY}:" prefix (Example: "{ie.SEARCH_KEY}{random.choice(_COUNTS)}:{random.choice(_SEARCHES)}")'
 133             write_string(desc + '\n', out=sys.stdout)
 134         sys.exit(0)
 135     if opts.ap_list_mso:
 136         table = [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]
 137         write_string('Supported TV Providers:\n' + render_table(['mso', 'mso name'], table) + '\n', out=sys.stdout)
 138         sys.exit(0)
 139
 140     # Conflicting, missing and erroneous options
 141     if opts.format == 'best':
 142         warnings.append('.\n         '.join((
 143             '"-f best" selects the best pre-merged format which is often not the best option',
 144             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 145             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 146     if opts.exec_cmd.get('before_dl') and opts.exec_before_dl_cmd:
 147         parser.error('using "--exec-before-download" conflicts with "--exec before_dl:"')
 148     if opts.usenetrc and (opts.username is not None or opts.password is not None):
 149         parser.error('using .netrc conflicts with giving username/password')
 150     if opts.password is not None and opts.username is None:
 151         parser.error('account username missing\n')
 152     if opts.ap_password is not None and opts.ap_username is None:
 153         parser.error('TV Provider account username missing\n')
 154     if opts.autonumber_size is not None:
 155         if opts.autonumber_size <= 0:
 156             parser.error('auto number size must be positive')
 157     if opts.autonumber_start is not None:
 158         if opts.autonumber_start < 0:
 159             parser.error('auto number start must be positive or 0')
 160     if opts.username is not None and opts.password is None:
 161         opts.password = compat_getpass('Type account password and press [Return]: ')
 162     if opts.ap_username is not None and opts.ap_password is None:
 163         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 164     if opts.ratelimit is not None:
 165         numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
 166         if numeric_limit is None:
 167             parser.error('invalid rate limit specified')
 168         opts.ratelimit = numeric_limit
 169     if opts.throttledratelimit is not None:
 170         numeric_limit = FileDownloader.parse_bytes(opts.throttledratelimit)
 171         if numeric_limit is None:
 172             parser.error('invalid rate limit specified')
 173         opts.throttledratelimit = numeric_limit
 174     if opts.min_filesize is not None:
 175         numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
 176         if numeric_limit is None:
 177             parser.error('invalid min_filesize specified')
 178         opts.min_filesize = numeric_limit
 179     if opts.max_filesize is not None:
 180         numeric_limit = FileDownloader.parse_bytes(opts.max_filesize)
 181         if numeric_limit is None:
 182             parser.error('invalid max_filesize specified')
 183         opts.max_filesize = numeric_limit
 184     if opts.sleep_interval is not None:
 185         if opts.sleep_interval < 0:
 186             parser.error('sleep interval must be positive or 0')
 187     if opts.max_sleep_interval is not None:
 188         if opts.max_sleep_interval < 0:
 189             parser.error('max sleep interval must be positive or 0')
 190         if opts.sleep_interval is None:
 191             parser.error('min sleep interval must be specified, use --min-sleep-interval')
 192         if opts.max_sleep_interval < opts.sleep_interval:
 193             parser.error('max sleep interval must be greater than or equal to min sleep interval')
 194     else:
 195         opts.max_sleep_interval = opts.sleep_interval
 196     if opts.sleep_interval_subtitles is not None:
 197         if opts.sleep_interval_subtitles < 0:
 198             parser.error('subtitles sleep interval must be positive or 0')
 199     if opts.sleep_interval_requests is not None:
 200         if opts.sleep_interval_requests < 0:
 201             parser.error('requests sleep interval must be positive or 0')
 202     if opts.ap_mso and opts.ap_mso not in MSO_INFO:
 203         parser.error('Unsupported TV Provider, use --ap-list-mso to get a list of supported TV Providers')
 204     if opts.overwrites:  # --yes-overwrites implies --no-continue
 205         opts.continue_dl = False
 206     if opts.concurrent_fragment_downloads <= 0:
 207         parser.error('Concurrent fragments must be positive')
 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         if min_wait is None or (max_wait is None and '-' in opts.wait_for_video):
 211             parser.error('Invalid time range to wait')
 212         elif max_wait is not None and max_wait < min_wait:
 213             parser.error('Minimum time range to wait must not be longer than the maximum')
 214         opts.wait_for_video = (min_wait, max_wait)
 215
 216     def parse_retries(retries, name=''):
 217         if retries in ('inf', 'infinite'):
 218             parsed_retries = float('inf')
 219         else:
 220             try:
 221                 parsed_retries = int(retries)
 222             except (TypeError, ValueError):
 223                 parser.error('invalid %sretry count specified' % name)
 224         return parsed_retries
 225     if opts.retries is not None:
 226         opts.retries = parse_retries(opts.retries)
 227     if opts.file_access_retries is not None:
 228         opts.file_access_retries = parse_retries(opts.file_access_retries, 'file access ')
 229     if opts.fragment_retries is not None:
 230         opts.fragment_retries = parse_retries(opts.fragment_retries, 'fragment ')
 231     if opts.extractor_retries is not None:
 232         opts.extractor_retries = parse_retries(opts.extractor_retries, 'extractor ')
 233     if opts.buffersize is not None:
 234         numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
 235         if numeric_buffersize is None:
 236             parser.error('invalid buffer size specified')
 237         opts.buffersize = numeric_buffersize
 238     if opts.http_chunk_size is not None:
 239         numeric_chunksize = FileDownloader.parse_bytes(opts.http_chunk_size)
 240         if not numeric_chunksize:
 241             parser.error('invalid http chunk size specified')
 242         opts.http_chunk_size = numeric_chunksize
 243     if opts.playliststart <= 0:
 244         raise parser.error('Playlist start must be positive')
 245     if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
 246         raise parser.error('Playlist end must be greater than playlist start')
 247     if opts.extractaudio:
 248         opts.audioformat = opts.audioformat.lower()
 249         if opts.audioformat not in ['best'] + list(FFmpegExtractAudioPP.SUPPORTED_EXTS):
 250             parser.error('invalid audio format specified')
 251     if opts.audioquality:
 252         opts.audioquality = opts.audioquality.strip('k').strip('K')
 253         audioquality = int_or_none(float_or_none(opts.audioquality))  # int_or_none prevents inf, nan
 254         if audioquality is None or audioquality < 0:
 255             parser.error('invalid audio quality specified')
 256     if opts.recodevideo is not None:
 257         opts.recodevideo = opts.recodevideo.replace(' ', '')
 258         if not re.match(FFmpegVideoConvertorPP.FORMAT_RE, opts.recodevideo):
 259             parser.error('invalid video remux format specified')
 260     if opts.remuxvideo is not None:
 261         opts.remuxvideo = opts.remuxvideo.replace(' ', '')
 262         if not re.match(FFmpegVideoRemuxerPP.FORMAT_RE, opts.remuxvideo):
 263             parser.error('invalid video remux format specified')
 264     if opts.convertsubtitles is not None:
 265         if opts.convertsubtitles not in FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS:
 266             parser.error('invalid subtitle format specified')
 267     if opts.convertthumbnails is not None:
 268         if opts.convertthumbnails not in FFmpegThumbnailsConvertorPP.SUPPORTED_EXTS:
 269             parser.error('invalid thumbnail format specified')
 270     if opts.cookiesfrombrowser is not None:
 271         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 272         if mobj is None:
 273             parser.error(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 274         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 275         browser_name = browser_name.lower()
 276         if browser_name not in SUPPORTED_BROWSERS:
 277             parser.error(f'unsupported browser specified for cookies: "{browser_name}". '
 278                          f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 279         if keyring is not None:
 280             keyring = keyring.upper()
 281             if keyring not in SUPPORTED_KEYRINGS:
 282                 parser.error(f'unsupported keyring specified for cookies: "{keyring}". '
 283                              f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 284         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 285     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 286     if geo_bypass_code is not None:
 287         try:
 288             GeoUtils.random_ipv4(geo_bypass_code)
 289         except Exception:
 290             parser.error('unsupported geo-bypass country or ip-block')
 291
 292     if opts.date is not None:
 293         date = DateRange.day(opts.date)
 294     else:
 295         date = DateRange(opts.dateafter, opts.datebefore)
 296
 297     compat_opts = opts.compat_opts
 298
 299     def report_conflict(arg1, arg2):
 300         warnings.append(f'{arg2} is ignored since {arg1} was given')
 301
 302     def _unused_compat_opt(name):
 303         if name not in compat_opts:
 304             return False
 305         compat_opts.discard(name)
 306         compat_opts.update(['*%s' % name])
 307         return True
 308
 309     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 310         attr = getattr(opts, opt_name)
 311         if compat_name in compat_opts:
 312             if attr is None:
 313                 setattr(opts, opt_name, not default)
 314                 return True
 315             else:
 316                 if remove_compat:
 317                     _unused_compat_opt(compat_name)
 318                 return False
 319         elif attr is None:
 320             setattr(opts, opt_name, default)
 321         return None
 322
 323     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 324     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 325     set_default_compat('no-clean-infojson', 'clean_infojson')
 326     if 'no-attach-info-json' in compat_opts:
 327         if opts.embed_infojson:
 328             _unused_compat_opt('no-attach-info-json')
 329         else:
 330             opts.embed_infojson = False
 331     if 'format-sort' in compat_opts:
 332         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 333     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 334     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 335     if _video_multistreams_set is False and _audio_multistreams_set is False:
 336         _unused_compat_opt('multistreams')
 337     outtmpl_default = opts.outtmpl.get('default')
 338     if opts.useid:
 339         if outtmpl_default is None:
 340             outtmpl_default = opts.outtmpl['default'] = '%(id)s.%(ext)s'
 341         else:
 342             report_conflict('--output', '--id')
 343     if 'filename' in compat_opts:
 344         if outtmpl_default is None:
 345             outtmpl_default = opts.outtmpl['default'] = '%(title)s-%(id)s.%(ext)s'
 346         else:
 347             _unused_compat_opt('filename')
 348
 349     def validate_outtmpl(tmpl, msg):
 350         err = YoutubeDL.validate_outtmpl(tmpl)
 351         if err:
 352             parser.error('invalid %s %r: %s' % (msg, tmpl, error_to_compat_str(err)))
 353
 354     for k, tmpl in opts.outtmpl.items():
 355         validate_outtmpl(tmpl, f'{k} output template')
 356     for type_, tmpl_list in opts.forceprint.items():
 357         for tmpl in tmpl_list:
 358             validate_outtmpl(tmpl, f'{type_} print template')
 359     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 360     for k, tmpl in opts.progress_template.items():
 361         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 362         validate_outtmpl(tmpl, f'{k} template')
 363
 364     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 365         opts.format = 'bestaudio/best'
 366
 367     if outtmpl_default is not None and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 368         parser.error('Cannot download a video and extract audio into the same'
 369                      ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
 370                      ' template'.format(outtmpl_default))
 371
 372     for f in opts.format_sort:
 373         if re.match(InfoExtractor.FormatSort.regex, f) is None:
 374             parser.error('invalid format sort string "%s" specified' % f)
 375
 376     def metadataparser_actions(f):
 377         if isinstance(f, str):
 378             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 379             try:
 380                 actions = [MetadataFromFieldPP.to_action(f)]
 381             except Exception as err:
 382                 parser.error(f'{cmd} is invalid; {err}')
 383         else:
 384             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 385             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 386
 387         for action in actions:
 388             try:
 389                 MetadataParserPP.validate_action(*action)
 390             except Exception as err:
 391                 parser.error(f'{cmd} is invalid; {err}')
 392             yield action
 393
 394     if opts.parse_metadata is None:
 395         opts.parse_metadata = []
 396     if opts.metafromtitle is not None:
 397         opts.parse_metadata.append('title:%s' % opts.metafromtitle)
 398     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, opts.parse_metadata)))
 399
 400     any_getting = (any(opts.forceprint.values()) or opts.dumpjson or opts.dump_single_json
 401                    or opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail
 402                    or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration)
 403
 404     any_printing = opts.print_json
 405     download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 406
 407     # If JSON is not printed anywhere, but comments are requested, save it to file
 408     printing_json = opts.dumpjson or opts.print_json or opts.dump_single_json
 409     if opts.getcomments and not printing_json:
 410         opts.writeinfojson = True
 411
 412     if opts.no_sponsorblock:
 413         opts.sponsorblock_mark = set()
 414         opts.sponsorblock_remove = set()
 415     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 416
 417     opts.remove_chapters = opts.remove_chapters or []
 418
 419     if (opts.remove_chapters or sponsorblock_query) and opts.sponskrub is not False:
 420         if opts.sponskrub:
 421             if opts.remove_chapters:
 422                 report_conflict('--remove-chapters', '--sponskrub')
 423             if opts.sponsorblock_mark:
 424                 report_conflict('--sponsorblock-mark', '--sponskrub')
 425             if opts.sponsorblock_remove:
 426                 report_conflict('--sponsorblock-remove', '--sponskrub')
 427         opts.sponskrub = False
 428     if opts.sponskrub_cut and opts.split_chapters and opts.sponskrub is not False:
 429         report_conflict('--split-chapter', '--sponskrub-cut')
 430         opts.sponskrub_cut = False
 431
 432     if opts.remuxvideo and opts.recodevideo:
 433         report_conflict('--recode-video', '--remux-video')
 434         opts.remuxvideo = False
 435
 436     if opts.allow_unplayable_formats:
 437         def report_unplayable_conflict(opt_name, arg, default=False, allowed=None):
 438             val = getattr(opts, opt_name)
 439             if (not allowed and val) or (allowed and not allowed(val)):
 440                 report_conflict('--allow-unplayable-formats', arg)
 441                 setattr(opts, opt_name, default)
 442
 443         report_unplayable_conflict('extractaudio', '--extract-audio')
 444         report_unplayable_conflict('remuxvideo', '--remux-video')
 445         report_unplayable_conflict('recodevideo', '--recode-video')
 446         report_unplayable_conflict('addmetadata', '--embed-metadata')
 447         report_unplayable_conflict('addchapters', '--embed-chapters')
 448         report_unplayable_conflict('embed_infojson', '--embed-info-json')
 449         opts.embed_infojson = False
 450         report_unplayable_conflict('embedsubtitles', '--embed-subs')
 451         report_unplayable_conflict('embedthumbnail', '--embed-thumbnail')
 452         report_unplayable_conflict('xattrs', '--xattrs')
 453         report_unplayable_conflict('fixup', '--fixup', default='never', allowed=lambda x: x in (None, 'never', 'ignore'))
 454         opts.fixup = 'never'
 455         report_unplayable_conflict('remove_chapters', '--remove-chapters', default=[])
 456         report_unplayable_conflict('sponsorblock_remove', '--sponsorblock-remove', default=set())
 457         report_unplayable_conflict('sponskrub', '--sponskrub', default=set())
 458         opts.sponskrub = False
 459
 460     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 461         opts.addchapters = True
 462
 463     # PostProcessors
 464     postprocessors = list(opts.add_postprocessors)
 465     if sponsorblock_query:
 466         postprocessors.append({
 467             'key': 'SponsorBlock',
 468             'categories': sponsorblock_query,
 469             'api': opts.sponsorblock_api,
 470             # Run this immediately after extraction is complete
 471             'when': 'pre_process'
 472         })
 473     if opts.parse_metadata:
 474         postprocessors.append({
 475             'key': 'MetadataParser',
 476             'actions': opts.parse_metadata,
 477             # Run this immediately after extraction is complete
 478             'when': 'pre_process'
 479         })
 480     if opts.convertsubtitles:
 481         postprocessors.append({
 482             'key': 'FFmpegSubtitlesConvertor',
 483             'format': opts.convertsubtitles,
 484             # Run this before the actual video download
 485             'when': 'before_dl'
 486         })
 487     if opts.convertthumbnails:
 488         postprocessors.append({
 489             'key': 'FFmpegThumbnailsConvertor',
 490             'format': opts.convertthumbnails,
 491             # Run this before the actual video download
 492             'when': 'before_dl'
 493         })
 494     if opts.extractaudio:
 495         postprocessors.append({
 496             'key': 'FFmpegExtractAudio',
 497             'preferredcodec': opts.audioformat,
 498             'preferredquality': opts.audioquality,
 499             'nopostoverwrites': opts.nopostoverwrites,
 500         })
 501     if opts.remuxvideo:
 502         postprocessors.append({
 503             'key': 'FFmpegVideoRemuxer',
 504             'preferedformat': opts.remuxvideo,
 505         })
 506     if opts.recodevideo:
 507         postprocessors.append({
 508             'key': 'FFmpegVideoConvertor',
 509             'preferedformat': opts.recodevideo,
 510         })
 511     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 512     if opts.embedsubtitles:
 513         already_have_subtitle = opts.writesubtitles and 'no-keep-subs' not in compat_opts
 514         postprocessors.append({
 515             'key': 'FFmpegEmbedSubtitle',
 516             # already_have_subtitle = True prevents the file from being deleted after embedding
 517             'already_have_subtitle': already_have_subtitle
 518         })
 519         if not opts.writeautomaticsub and 'no-keep-subs' not in compat_opts:
 520             opts.writesubtitles = True
 521     # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 522     # this was the old behaviour if only --all-sub was given.
 523     if opts.allsubtitles and not opts.writeautomaticsub:
 524         opts.writesubtitles = True
 525     # ModifyChapters must run before FFmpegMetadataPP
 526     remove_chapters_patterns, remove_ranges = [], []
 527     for regex in opts.remove_chapters:
 528         if regex.startswith('*'):
 529             dur = list(map(parse_duration, regex[1:].split('-')))
 530             if len(dur) == 2 and all(t is not None for t in dur):
 531                 remove_ranges.append(tuple(dur))
 532                 continue
 533             parser.error(f'invalid --remove-chapters time range {regex!r}. Must be of the form *start-end')
 534         try:
 535             remove_chapters_patterns.append(re.compile(regex))
 536         except re.error as err:
 537             parser.error(f'invalid --remove-chapters regex {regex!r} - {err}')
 538     if opts.remove_chapters or sponsorblock_query:
 539         postprocessors.append({
 540             'key': 'ModifyChapters',
 541             'remove_chapters_patterns': remove_chapters_patterns,
 542             'remove_sponsor_segments': opts.sponsorblock_remove,
 543             'remove_ranges': remove_ranges,
 544             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 545             'force_keyframes': opts.force_keyframes_at_cuts
 546         })
 547     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 548     # FFmpegExtractAudioPP as containers before conversion may not support
 549     # metadata (3gp, webm, etc.)
 550     # By default ffmpeg preserves metadata applicable for both
 551     # source and target containers. From this point the container won't change,
 552     # so metadata can be added here.
 553     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 554         if opts.embed_infojson is None:
 555             opts.embed_infojson = 'if_exists'
 556         postprocessors.append({
 557             'key': 'FFmpegMetadata',
 558             'add_chapters': opts.addchapters,
 559             'add_metadata': opts.addmetadata,
 560             'add_infojson': opts.embed_infojson,
 561         })
 562     # Deprecated
 563     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 564     # but must be below EmbedSubtitle and FFmpegMetadata
 565     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 566     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 567     if opts.sponskrub is not False:
 568         postprocessors.append({
 569             'key': 'SponSkrub',
 570             'path': opts.sponskrub_path,
 571             'args': opts.sponskrub_args,
 572             'cut': opts.sponskrub_cut,
 573             'force': opts.sponskrub_force,
 574             'ignoreerror': opts.sponskrub is None,
 575             '_from_cli': True,
 576         })
 577     if opts.embedthumbnail:
 578         postprocessors.append({
 579             'key': 'EmbedThumbnail',
 580             # already_have_thumbnail = True prevents the file from being deleted after embedding
 581             'already_have_thumbnail': opts.writethumbnail
 582         })
 583         if not opts.writethumbnail:
 584             opts.writethumbnail = True
 585             opts.outtmpl['pl_thumbnail'] = ''
 586     if opts.split_chapters:
 587         postprocessors.append({
 588             'key': 'FFmpegSplitChapters',
 589             'force_keyframes': opts.force_keyframes_at_cuts,
 590         })
 591     # XAttrMetadataPP should be run after post-processors that may change file contents
 592     if opts.xattrs:
 593         postprocessors.append({'key': 'XAttrMetadata'})
 594     # Exec must be the last PP of each category
 595     if opts.exec_before_dl_cmd:
 596         opts.exec_cmd.setdefault('before_dl', opts.exec_before_dl_cmd)
 597     for when, exec_cmd in opts.exec_cmd.items():
 598         postprocessors.append({
 599             'key': 'Exec',
 600             'exec_cmd': exec_cmd,
 601             # Run this only after the files have been moved to their final locations
 602             'when': when,
 603         })
 604
 605     def report_args_compat(arg, name):
 606         warnings.append('%s given without specifying name. The arguments will be given to all %s' % (arg, name))
 607
 608     if 'default' in opts.external_downloader_args:
 609         report_args_compat('--downloader-args', 'external downloaders')
 610
 611     if 'default-compat' in opts.postprocessor_args and 'default' not in opts.postprocessor_args:
 612         report_args_compat('--post-processor-args', 'post-processors')
 613         opts.postprocessor_args.setdefault('sponskrub', [])
 614         opts.postprocessor_args['default'] = opts.postprocessor_args['default-compat']
 615
 616     def report_deprecation(val, old, new=None):
 617         if not val:
 618             return
 619         deprecation_warnings.append(
 620             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 621             else f'{old} is deprecated and may not work as expected')
 622
 623     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 624     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 625     report_deprecation(opts.include_ads, '--include-ads')
 626     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 627     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 628
 629     final_ext = (
 630         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 631         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 632         else opts.audioformat if (opts.extractaudio and opts.audioformat != 'best')
 633         else None)
 634
 635     match_filter = (
 636         None if opts.match_filter is None
 637         else match_filter_func(opts.match_filter))
 638
 639     ydl_opts = {
 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         'forcejson': opts.dumpjson or opts.print_json,
 661         'dump_single_json': opts.dump_single_json,
 662         'force_write_download_archive': opts.force_write_download_archive,
 663         'simulate': (any_getting or None) if opts.simulate is None else opts.simulate,
 664         'skip_download': opts.skip_download,
 665         'format': opts.format,
 666         'allow_unplayable_formats': opts.allow_unplayable_formats,
 667         'ignore_no_formats_error': opts.ignore_no_formats_error,
 668         'format_sort': opts.format_sort,
 669         'format_sort_force': opts.format_sort_force,
 670         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 671         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 672         'check_formats': opts.check_formats,
 673         'listformats': opts.listformats,
 674         'listformats_table': opts.listformats_table,
 675         'outtmpl': opts.outtmpl,
 676         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 677         'paths': opts.paths,
 678         'autonumber_size': opts.autonumber_size,
 679         'autonumber_start': opts.autonumber_start,
 680         'restrictfilenames': opts.restrictfilenames,
 681         'windowsfilenames': opts.windowsfilenames,
 682         'ignoreerrors': opts.ignoreerrors,
 683         'force_generic_extractor': opts.force_generic_extractor,
 684         'ratelimit': opts.ratelimit,
 685         'throttledratelimit': opts.throttledratelimit,
 686         'overwrites': opts.overwrites,
 687         'retries': opts.retries,
 688         'file_access_retries': opts.file_access_retries,
 689         'fragment_retries': opts.fragment_retries,
 690         'extractor_retries': opts.extractor_retries,
 691         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 692         'keep_fragments': opts.keep_fragments,
 693         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 694         'buffersize': opts.buffersize,
 695         'noresizebuffer': opts.noresizebuffer,
 696         'http_chunk_size': opts.http_chunk_size,
 697         'continuedl': opts.continue_dl,
 698         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 699         'progress_with_newline': opts.progress_with_newline,
 700         'progress_template': opts.progress_template,
 701         'playliststart': opts.playliststart,
 702         'playlistend': opts.playlistend,
 703         'playlistreverse': opts.playlist_reverse,
 704         'playlistrandom': opts.playlist_random,
 705         'noplaylist': opts.noplaylist,
 706         'logtostderr': outtmpl_default == '-',
 707         'consoletitle': opts.consoletitle,
 708         'nopart': opts.nopart,
 709         'updatetime': opts.updatetime,
 710         'writedescription': opts.writedescription,
 711         'writeannotations': opts.writeannotations,
 712         'writeinfojson': opts.writeinfojson,
 713         'allow_playlist_files': opts.allow_playlist_files,
 714         'clean_infojson': opts.clean_infojson,
 715         'getcomments': opts.getcomments,
 716         'writethumbnail': opts.writethumbnail is True,
 717         'write_all_thumbnails': opts.writethumbnail == 'all',
 718         'writelink': opts.writelink,
 719         'writeurllink': opts.writeurllink,
 720         'writewebloclink': opts.writewebloclink,
 721         'writedesktoplink': opts.writedesktoplink,
 722         'writesubtitles': opts.writesubtitles,
 723         'writeautomaticsub': opts.writeautomaticsub,
 724         'allsubtitles': opts.allsubtitles,
 725         'listsubtitles': opts.listsubtitles,
 726         'subtitlesformat': opts.subtitlesformat,
 727         'subtitleslangs': opts.subtitleslangs,
 728         'matchtitle': decodeOption(opts.matchtitle),
 729         'rejecttitle': decodeOption(opts.rejecttitle),
 730         'max_downloads': opts.max_downloads,
 731         'prefer_free_formats': opts.prefer_free_formats,
 732         'trim_file_name': opts.trim_file_name,
 733         'verbose': opts.verbose,
 734         'dump_intermediate_pages': opts.dump_intermediate_pages,
 735         'write_pages': opts.write_pages,
 736         'test': opts.test,
 737         'keepvideo': opts.keepvideo,
 738         'min_filesize': opts.min_filesize,
 739         'max_filesize': opts.max_filesize,
 740         'min_views': opts.min_views,
 741         'max_views': opts.max_views,
 742         'daterange': date,
 743         'cachedir': opts.cachedir,
 744         'youtube_print_sig_code': opts.youtube_print_sig_code,
 745         'age_limit': opts.age_limit,
 746         'download_archive': download_archive_fn,
 747         'break_on_existing': opts.break_on_existing,
 748         'break_on_reject': opts.break_on_reject,
 749         'break_per_url': opts.break_per_url,
 750         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 751         'cookiefile': opts.cookiefile,
 752         'cookiesfrombrowser': opts.cookiesfrombrowser,
 753         'nocheckcertificate': opts.no_check_certificate,
 754         'prefer_insecure': opts.prefer_insecure,
 755         'proxy': opts.proxy,
 756         'socket_timeout': opts.socket_timeout,
 757         'bidi_workaround': opts.bidi_workaround,
 758         'debug_printtraffic': opts.debug_printtraffic,
 759         'prefer_ffmpeg': opts.prefer_ffmpeg,
 760         'include_ads': opts.include_ads,
 761         'default_search': opts.default_search,
 762         'dynamic_mpd': opts.dynamic_mpd,
 763         'extractor_args': opts.extractor_args,
 764         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 765         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 766         'encoding': opts.encoding,
 767         'extract_flat': opts.extract_flat,
 768         'live_from_start': opts.live_from_start,
 769         'wait_for_video': opts.wait_for_video,
 770         'mark_watched': opts.mark_watched,
 771         'merge_output_format': opts.merge_output_format,
 772         'final_ext': final_ext,
 773         'postprocessors': postprocessors,
 774         'fixup': opts.fixup,
 775         'source_address': opts.source_address,
 776         'call_home': opts.call_home,
 777         'sleep_interval_requests': opts.sleep_interval_requests,
 778         'sleep_interval': opts.sleep_interval,
 779         'max_sleep_interval': opts.max_sleep_interval,
 780         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 781         'external_downloader': opts.external_downloader,
 782         'list_thumbnails': opts.list_thumbnails,
 783         'playlist_items': opts.playlist_items,
 784         'xattr_set_filesize': opts.xattr_set_filesize,
 785         'match_filter': match_filter,
 786         'no_color': opts.no_color,
 787         'ffmpeg_location': opts.ffmpeg_location,
 788         'hls_prefer_native': opts.hls_prefer_native,
 789         'hls_use_mpegts': opts.hls_use_mpegts,
 790         'hls_split_discontinuity': opts.hls_split_discontinuity,
 791         'external_downloader_args': opts.external_downloader_args,
 792         'postprocessor_args': opts.postprocessor_args,
 793         'cn_verification_proxy': opts.cn_verification_proxy,
 794         'geo_verification_proxy': opts.geo_verification_proxy,
 795         'geo_bypass': opts.geo_bypass,
 796         'geo_bypass_country': opts.geo_bypass_country,
 797         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 798         '_warnings': warnings,
 799         '_deprecation_warnings': deprecation_warnings,
 800         'compat_opts': compat_opts,
 801     }
 802
 803     with YoutubeDL(ydl_opts) as ydl:
 804         actual_use = all_urls or opts.load_info_filename
 805
 806         # Remove cache dir
 807         if opts.rm_cachedir:
 808             ydl.cache.remove()
 809
 810         # Update version
 811         if opts.update_self:
 812             # If updater returns True, exit. Required for windows
 813             if run_update(ydl):
 814                 if actual_use:
 815                     sys.exit('ERROR: The program must exit for the update to complete')
 816                 sys.exit()
 817
 818         # Maybe do nothing
 819         if not actual_use:
 820             if opts.update_self or opts.rm_cachedir:
 821                 sys.exit()
 822
 823             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 824             parser.error(
 825                 'You must provide at least one URL.\n'
 826                 'Type yt-dlp --help to see a list of all options.')
 827
 828         try:
 829             if opts.load_info_filename is not None:
 830                 retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
 831             else:
 832                 retcode = ydl.download(all_urls)
 833         except DownloadCancelled:
 834             ydl.to_screen('Aborting remaining downloads')
 835             retcode = 101
 836
 837     sys.exit(retcode)
 838
 839
 840 def main(argv=None):
 841     try:
 842         _real_main(argv)
 843     except DownloadError:
 844         sys.exit(1)
 845     except SameFileError as e:
 846         sys.exit(f'ERROR: {e}')
 847     except KeyboardInterrupt:
 848         sys.exit('\nERROR: Interrupted by user')
 849     except BrokenPipeError as e:
 850         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 851         devnull = os.open(os.devnull, os.O_WRONLY)
 852         os.dup2(devnull, sys.stdout.fileno())
 853         sys.exit(f'\nERROR: {e}')
 854
 855
 856 __all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']