yt_dlp/__init__.py

   1 #!/usr/bin/env python3
   2 f'You are using an unsupported version of Python. Only Python versions 3.6 and above are supported by yt-dlp'  # noqa: F541
   3
   4 __license__ = 'Public Domain'
   5
   6 import itertools
   7 import optparse
   8 import os
   9 import re
  10 import sys
  11
  12 from .compat import compat_getpass, compat_shlex_quote
  13 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
  14 from .downloader import FileDownloader
  15 from .extractor import GenericIE, list_extractor_classes
  16 from .extractor.adobepass import MSO_INFO
  17 from .extractor.common import InfoExtractor
  18 from .options import parseOpts
  19 from .postprocessor import (
  20     FFmpegExtractAudioPP,
  21     FFmpegSubtitlesConvertorPP,
  22     FFmpegThumbnailsConvertorPP,
  23     FFmpegVideoConvertorPP,
  24     FFmpegVideoRemuxerPP,
  25     MetadataFromFieldPP,
  26     MetadataParserPP,
  27 )
  28 from .update import run_update
  29 from .utils import (
  30     NO_DEFAULT,
  31     POSTPROCESS_WHEN,
  32     DateRange,
  33     DownloadCancelled,
  34     DownloadError,
  35     GeoUtils,
  36     SameFileError,
  37     decodeOption,
  38     download_range_func,
  39     expand_path,
  40     float_or_none,
  41     int_or_none,
  42     match_filter_func,
  43     parse_duration,
  44     preferredencoding,
  45     read_batch_urls,
  46     read_stdin,
  47     render_table,
  48     setproctitle,
  49     std_headers,
  50     traverse_obj,
  51     variadic,
  52     write_string,
  53 )
  54 from .YoutubeDL import YoutubeDL
  55
  56
  57 def _exit(status=0, *args):
  58     for msg in args:
  59         sys.stderr.write(msg)
  60     raise SystemExit(status)
  61
  62
  63 def get_urls(urls, batchfile, verbose):
  64     # Batch file verification
  65     batch_urls = []
  66     if batchfile is not None:
  67         try:
  68             batch_urls = read_batch_urls(
  69                 read_stdin('URLs') if batchfile == '-'
  70                 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
  71             if verbose:
  72                 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
  73         except OSError:
  74             _exit(f'ERROR: batch file {batchfile} could not be read')
  75     _enc = preferredencoding()
  76     return [
  77         url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
  78         for url in batch_urls + urls]
  79
  80
  81 def print_extractor_information(opts, urls):
  82     out = ''
  83     if opts.list_extractors:
  84         urls = dict.fromkeys(urls, False)
  85         for ie in list_extractor_classes(opts.age_limit):
  86             out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
  87             if ie == GenericIE:
  88                 matched_urls = [url for url, matched in urls.items() if not matched]
  89             else:
  90                 matched_urls = tuple(filter(ie.suitable, urls.keys()))
  91                 urls.update(dict.fromkeys(matched_urls, True))
  92             out += ''.join(f'  {url}\n' for url in matched_urls)
  93     elif opts.list_extractor_descriptions:
  94         _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
  95         out = '\n'.join(
  96             ie.description(markdown=False, search_examples=_SEARCHES)
  97             for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
  98     elif opts.ap_list_mso:
  99         out = 'Supported TV Providers:\n%s\n' % render_table(
 100             ['mso', 'mso name'],
 101             [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
 102     else:
 103         return False
 104     write_string(out, out=sys.stdout)
 105     return True
 106
 107
 108 def set_compat_opts(opts):
 109     def _unused_compat_opt(name):
 110         if name not in opts.compat_opts:
 111             return False
 112         opts.compat_opts.discard(name)
 113         opts.compat_opts.update(['*%s' % name])
 114         return True
 115
 116     def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
 117         attr = getattr(opts, opt_name)
 118         if compat_name in opts.compat_opts:
 119             if attr is None:
 120                 setattr(opts, opt_name, not default)
 121                 return True
 122             else:
 123                 if remove_compat:
 124                     _unused_compat_opt(compat_name)
 125                 return False
 126         elif attr is None:
 127             setattr(opts, opt_name, default)
 128         return None
 129
 130     set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
 131     set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
 132     set_default_compat('no-clean-infojson', 'clean_infojson')
 133     if 'no-attach-info-json' in opts.compat_opts:
 134         if opts.embed_infojson:
 135             _unused_compat_opt('no-attach-info-json')
 136         else:
 137             opts.embed_infojson = False
 138     if 'format-sort' in opts.compat_opts:
 139         opts.format_sort.extend(InfoExtractor.FormatSort.ytdl_default)
 140     _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
 141     _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
 142     if _video_multistreams_set is False and _audio_multistreams_set is False:
 143         _unused_compat_opt('multistreams')
 144     if 'filename' in opts.compat_opts:
 145         if opts.outtmpl.get('default') is None:
 146             opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
 147         else:
 148             _unused_compat_opt('filename')
 149
 150
 151 def validate_options(opts):
 152     def validate(cndn, name, value=None, msg=None):
 153         if cndn:
 154             return True
 155         raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
 156
 157     def validate_in(name, value, items, msg=None):
 158         return validate(value is None or value in items, name, value, msg)
 159
 160     def validate_regex(name, value, regex):
 161         return validate(value is None or re.match(regex, value), name, value)
 162
 163     def validate_positive(name, value, strict=False):
 164         return validate(value is None or value > 0 or (not strict and value == 0),
 165                         name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
 166
 167     def validate_minmax(min_val, max_val, min_name, max_name=None):
 168         if max_val is None or min_val is None or max_val >= min_val:
 169             return
 170         if not max_name:
 171             min_name, max_name = f'min {min_name}', f'max {min_name}'
 172         raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
 173
 174     # Usernames and passwords
 175     validate(not opts.usenetrc or (opts.username is None and opts.password is None),
 176              '.netrc', msg='using {name} conflicts with giving username/password')
 177     validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
 178     validate(opts.ap_password is None or opts.ap_username is not None,
 179              'TV Provider account username', msg='{name} missing')
 180     validate_in('TV Provider', opts.ap_mso, MSO_INFO,
 181                 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
 182
 183     # Numbers
 184     validate_positive('autonumber start', opts.autonumber_start)
 185     validate_positive('autonumber size', opts.autonumber_size, True)
 186     validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
 187     validate_positive('playlist start', opts.playliststart, True)
 188     if opts.playlistend != -1:
 189         validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
 190
 191     # Time ranges
 192     validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
 193     validate_positive('requests sleep interval', opts.sleep_interval_requests)
 194     validate_positive('sleep interval', opts.sleep_interval)
 195     validate_positive('max sleep interval', opts.max_sleep_interval)
 196     if opts.sleep_interval is None:
 197         validate(
 198             opts.max_sleep_interval is None, 'min sleep interval',
 199             msg='{name} must be specified; use --min-sleep-interval')
 200     elif opts.max_sleep_interval is None:
 201         opts.max_sleep_interval = opts.sleep_interval
 202     else:
 203         validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
 204
 205     if opts.wait_for_video is not None:
 206         min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
 207         validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
 208                  'time range to wait for video', opts.wait_for_video)
 209         validate_minmax(min_wait, max_wait, 'time range to wait for video')
 210         opts.wait_for_video = (min_wait, max_wait)
 211
 212     # Format sort
 213     for f in opts.format_sort:
 214         validate_regex('format sorting', f, InfoExtractor.FormatSort.regex)
 215
 216     # Postprocessor formats
 217     validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
 218     validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
 219     validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
 220     validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
 221     validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
 222     if opts.audioquality:
 223         opts.audioquality = opts.audioquality.strip('k').strip('K')
 224         # int_or_none prevents inf, nan
 225         validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
 226
 227     # Retries
 228     def parse_retries(name, value):
 229         if value is None:
 230             return None
 231         elif value in ('inf', 'infinite'):
 232             return float('inf')
 233         try:
 234             return int(value)
 235         except (TypeError, ValueError):
 236             validate(False, f'{name} retry count', value)
 237
 238     opts.retries = parse_retries('download', opts.retries)
 239     opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
 240     opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
 241     opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
 242
 243     # Retry sleep function
 244     def parse_sleep_func(expr):
 245         NUMBER_RE = r'\d+(?:\.\d+)?'
 246         op, start, limit, step, *_ = tuple(re.fullmatch(
 247             rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
 248             expr.strip()).groups()) + (None, None)
 249
 250         if op == 'exp':
 251             return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
 252         else:
 253             default_step = start if op or limit else 0
 254             return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
 255
 256     for key, expr in opts.retry_sleep.items():
 257         if not expr:
 258             del opts.retry_sleep[key]
 259             continue
 260         try:
 261             opts.retry_sleep[key] = parse_sleep_func(expr)
 262         except AttributeError:
 263             raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
 264
 265     # Bytes
 266     def parse_bytes(name, value):
 267         if value is None:
 268             return None
 269         numeric_limit = FileDownloader.parse_bytes(value)
 270         validate(numeric_limit is not None, 'rate limit', value)
 271         return numeric_limit
 272
 273     opts.ratelimit = parse_bytes('rate limit', opts.ratelimit)
 274     opts.throttledratelimit = parse_bytes('throttled rate limit', opts.throttledratelimit)
 275     opts.min_filesize = parse_bytes('min filesize', opts.min_filesize)
 276     opts.max_filesize = parse_bytes('max filesize', opts.max_filesize)
 277     opts.buffersize = parse_bytes('buffer size', opts.buffersize)
 278     opts.http_chunk_size = parse_bytes('http chunk size', opts.http_chunk_size)
 279
 280     # Output templates
 281     def validate_outtmpl(tmpl, msg):
 282         err = YoutubeDL.validate_outtmpl(tmpl)
 283         if err:
 284             raise ValueError(f'invalid {msg} "{tmpl}": {err}')
 285
 286     for k, tmpl in opts.outtmpl.items():
 287         validate_outtmpl(tmpl, f'{k} output template')
 288     for type_, tmpl_list in opts.forceprint.items():
 289         for tmpl in tmpl_list:
 290             validate_outtmpl(tmpl, f'{type_} print template')
 291     for type_, tmpl_list in opts.print_to_file.items():
 292         for tmpl, file in tmpl_list:
 293             validate_outtmpl(tmpl, f'{type_} print to file template')
 294             validate_outtmpl(file, f'{type_} print to file filename')
 295     validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
 296     for k, tmpl in opts.progress_template.items():
 297         k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
 298         validate_outtmpl(tmpl, f'{k} template')
 299
 300     outtmpl_default = opts.outtmpl.get('default')
 301     if outtmpl_default == '':
 302         opts.skip_download = None
 303         del opts.outtmpl['default']
 304     if outtmpl_default and not os.path.splitext(outtmpl_default)[1] and opts.extractaudio:
 305         raise ValueError(
 306             'Cannot download a video and extract audio into the same file! '
 307             f'Use "{outtmpl_default}.%(ext)s" instead of "{outtmpl_default}" as the output template')
 308
 309     def parse_chapters(name, value):
 310         chapters, ranges = [], []
 311         for regex in value or []:
 312             if regex.startswith('*'):
 313                 for range in regex[1:].split(','):
 314                     dur = tuple(map(parse_duration, range.strip().split('-')))
 315                     if len(dur) == 2 and all(t is not None for t in dur):
 316                         ranges.append(dur)
 317                     else:
 318                         raise ValueError(f'invalid {name} time range "{regex}". Must be of the form *start-end')
 319                 continue
 320             try:
 321                 chapters.append(re.compile(regex))
 322             except re.error as err:
 323                 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
 324         return chapters, ranges
 325
 326     opts.remove_chapters, opts.remove_ranges = parse_chapters('--remove-chapters', opts.remove_chapters)
 327     opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges))
 328
 329     # Cookies from browser
 330     if opts.cookiesfrombrowser:
 331         mobj = re.match(r'(?P<name>[^+:]+)(\s*\+\s*(?P<keyring>[^:]+))?(\s*:(?P<profile>.+))?', opts.cookiesfrombrowser)
 332         if mobj is None:
 333             raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
 334         browser_name, keyring, profile = mobj.group('name', 'keyring', 'profile')
 335         browser_name = browser_name.lower()
 336         if browser_name not in SUPPORTED_BROWSERS:
 337             raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
 338                              f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
 339         if keyring is not None:
 340             keyring = keyring.upper()
 341             if keyring not in SUPPORTED_KEYRINGS:
 342                 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
 343                                  f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
 344         opts.cookiesfrombrowser = (browser_name, profile, keyring)
 345
 346     # MetadataParser
 347     def metadataparser_actions(f):
 348         if isinstance(f, str):
 349             cmd = '--parse-metadata %s' % compat_shlex_quote(f)
 350             try:
 351                 actions = [MetadataFromFieldPP.to_action(f)]
 352             except Exception as err:
 353                 raise ValueError(f'{cmd} is invalid; {err}')
 354         else:
 355             cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
 356             actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
 357
 358         for action in actions:
 359             try:
 360                 MetadataParserPP.validate_action(*action)
 361             except Exception as err:
 362                 raise ValueError(f'{cmd} is invalid; {err}')
 363             yield action
 364
 365     parse_metadata = opts.parse_metadata or []
 366     if opts.metafromtitle is not None:
 367         parse_metadata.append('title:%s' % opts.metafromtitle)
 368     opts.parse_metadata = list(itertools.chain(*map(metadataparser_actions, parse_metadata)))
 369
 370     # Other options
 371     geo_bypass_code = opts.geo_bypass_ip_block or opts.geo_bypass_country
 372     if geo_bypass_code is not None:
 373         try:
 374             GeoUtils.random_ipv4(geo_bypass_code)
 375         except Exception:
 376             raise ValueError('unsupported geo-bypass country or ip-block')
 377
 378     opts.match_filter = match_filter_func(opts.match_filter)
 379
 380     if opts.download_archive is not None:
 381         opts.download_archive = expand_path(opts.download_archive)
 382
 383     if opts.user_agent is not None:
 384         opts.headers.setdefault('User-Agent', opts.user_agent)
 385     if opts.referer is not None:
 386         opts.headers.setdefault('Referer', opts.referer)
 387
 388     if opts.no_sponsorblock:
 389         opts.sponsorblock_mark = opts.sponsorblock_remove = set()
 390
 391     warnings, deprecation_warnings = [], []
 392
 393     # Common mistake: -f best
 394     if opts.format == 'best':
 395         warnings.append('.\n         '.join((
 396             '"-f best" selects the best pre-merged format which is often not the best option',
 397             'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
 398             'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
 399
 400     # --(postprocessor/downloader)-args without name
 401     def report_args_compat(name, value, key1, key2=None):
 402         if key1 in value and key2 not in value:
 403             warnings.append(f'{name} arguments given without specifying name. The arguments will be given to all {name}s')
 404             return True
 405         return False
 406
 407     report_args_compat('external downloader', opts.external_downloader_args, 'default')
 408     if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
 409         opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
 410         opts.postprocessor_args.setdefault('sponskrub', [])
 411
 412     def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
 413                         val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
 414         if val2 is NO_DEFAULT:
 415             val2 = getattr(opts, opt2)
 416         if not val2:
 417             return
 418
 419         if val1 is NO_DEFAULT:
 420             val1 = getattr(opts, opt1)
 421         if val1:
 422             warnings.append(f'{arg1} is ignored since {arg2} was given')
 423         setattr(opts, opt1, default)
 424
 425     # Conflicting options
 426     report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
 427     report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
 428     report_conflict('--exec-before-download', 'exec_before_dl_cmd',
 429                     '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
 430     report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
 431     report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
 432     report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
 433     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
 434     report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
 435     report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
 436                     val1=opts.sponskrub and opts.sponskrub_cut)
 437
 438     # Conflicts with --allow-unplayable-formats
 439     report_conflict('--add-metadata', 'addmetadata')
 440     report_conflict('--embed-chapters', 'addchapters')
 441     report_conflict('--embed-info-json', 'embed_infojson')
 442     report_conflict('--embed-subs', 'embedsubtitles')
 443     report_conflict('--embed-thumbnail', 'embedthumbnail')
 444     report_conflict('--extract-audio', 'extractaudio')
 445     report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
 446     report_conflict('--recode-video', 'recodevideo')
 447     report_conflict('--remove-chapters', 'remove_chapters', default=[])
 448     report_conflict('--remux-video', 'remuxvideo')
 449     report_conflict('--sponskrub', 'sponskrub')
 450     report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
 451     report_conflict('--xattrs', 'xattrs')
 452
 453     # Fully deprecated options
 454     def report_deprecation(val, old, new=None):
 455         if not val:
 456             return
 457         deprecation_warnings.append(
 458             f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
 459             else f'{old} is deprecated and may not work as expected')
 460
 461     report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
 462     report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
 463     # report_deprecation(opts.include_ads, '--include-ads')  # We may re-implement this in future
 464     # report_deprecation(opts.call_home, '--call-home')  # We may re-implement this in future
 465     # report_deprecation(opts.writeannotations, '--write-annotations')  # It's just that no website has it
 466
 467     # Dependent options
 468     opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
 469
 470     if opts.exec_before_dl_cmd:
 471         opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
 472
 473     if opts.useid:  # --id is not deprecated in youtube-dl
 474         opts.outtmpl['default'] = '%(id)s.%(ext)s'
 475
 476     if opts.overwrites:  # --force-overwrites implies --no-continue
 477         opts.continue_dl = False
 478
 479     if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
 480         # Add chapters when adding metadata or marking sponsors
 481         opts.addchapters = True
 482
 483     if opts.extractaudio and not opts.keepvideo and opts.format is None:
 484         # Do not unnecessarily download audio
 485         opts.format = 'bestaudio/best'
 486
 487     if opts.getcomments and opts.writeinfojson is None:
 488         # If JSON is not printed anywhere, but comments are requested, save it to file
 489         if not opts.dumpjson or opts.print_json or opts.dump_single_json:
 490             opts.writeinfojson = True
 491
 492     if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
 493         # --all-sub automatically sets --write-sub if --write-auto-sub is not given
 494         opts.writesubtitles = True
 495
 496     if opts.addmetadata and opts.embed_infojson is None:
 497         # If embedding metadata and infojson is present, embed it
 498         opts.embed_infojson = 'if_exists'
 499
 500     # Ask for passwords
 501     if opts.username is not None and opts.password is None:
 502         opts.password = compat_getpass('Type account password and press [Return]: ')
 503     if opts.ap_username is not None and opts.ap_password is None:
 504         opts.ap_password = compat_getpass('Type TV provider account password and press [Return]: ')
 505
 506     return warnings, deprecation_warnings
 507
 508
 509 def get_postprocessors(opts):
 510     yield from opts.add_postprocessors
 511
 512     if opts.parse_metadata:
 513         yield {
 514             'key': 'MetadataParser',
 515             'actions': opts.parse_metadata,
 516             'when': 'pre_process'
 517         }
 518     sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
 519     if sponsorblock_query:
 520         yield {
 521             'key': 'SponsorBlock',
 522             'categories': sponsorblock_query,
 523             'api': opts.sponsorblock_api,
 524             'when': 'after_filter'
 525         }
 526     if opts.convertsubtitles:
 527         yield {
 528             'key': 'FFmpegSubtitlesConvertor',
 529             'format': opts.convertsubtitles,
 530             'when': 'before_dl'
 531         }
 532     if opts.convertthumbnails:
 533         yield {
 534             'key': 'FFmpegThumbnailsConvertor',
 535             'format': opts.convertthumbnails,
 536             'when': 'before_dl'
 537         }
 538     if opts.extractaudio:
 539         yield {
 540             'key': 'FFmpegExtractAudio',
 541             'preferredcodec': opts.audioformat,
 542             'preferredquality': opts.audioquality,
 543             'nopostoverwrites': opts.nopostoverwrites,
 544         }
 545     if opts.remuxvideo:
 546         yield {
 547             'key': 'FFmpegVideoRemuxer',
 548             'preferedformat': opts.remuxvideo,
 549         }
 550     if opts.recodevideo:
 551         yield {
 552             'key': 'FFmpegVideoConvertor',
 553             'preferedformat': opts.recodevideo,
 554         }
 555     # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
 556     if opts.embedsubtitles:
 557         keep_subs = 'no-keep-subs' not in opts.compat_opts
 558         yield {
 559             'key': 'FFmpegEmbedSubtitle',
 560             # already_have_subtitle = True prevents the file from being deleted after embedding
 561             'already_have_subtitle': opts.writesubtitles and keep_subs
 562         }
 563         if not opts.writeautomaticsub and keep_subs:
 564             opts.writesubtitles = True
 565
 566     # ModifyChapters must run before FFmpegMetadataPP
 567     if opts.remove_chapters or sponsorblock_query:
 568         yield {
 569             'key': 'ModifyChapters',
 570             'remove_chapters_patterns': opts.remove_chapters,
 571             'remove_sponsor_segments': opts.sponsorblock_remove,
 572             'remove_ranges': opts.remove_ranges,
 573             'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
 574             'force_keyframes': opts.force_keyframes_at_cuts
 575         }
 576     # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
 577     # FFmpegExtractAudioPP as containers before conversion may not support
 578     # metadata (3gp, webm, etc.)
 579     # By default ffmpeg preserves metadata applicable for both
 580     # source and target containers. From this point the container won't change,
 581     # so metadata can be added here.
 582     if opts.addmetadata or opts.addchapters or opts.embed_infojson:
 583         yield {
 584             'key': 'FFmpegMetadata',
 585             'add_chapters': opts.addchapters,
 586             'add_metadata': opts.addmetadata,
 587             'add_infojson': opts.embed_infojson,
 588         }
 589     # Deprecated
 590     # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
 591     # but must be below EmbedSubtitle and FFmpegMetadata
 592     # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
 593     # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
 594     if opts.sponskrub is not False:
 595         yield {
 596             'key': 'SponSkrub',
 597             'path': opts.sponskrub_path,
 598             'args': opts.sponskrub_args,
 599             'cut': opts.sponskrub_cut,
 600             'force': opts.sponskrub_force,
 601             'ignoreerror': opts.sponskrub is None,
 602             '_from_cli': True,
 603         }
 604     if opts.embedthumbnail:
 605         yield {
 606             'key': 'EmbedThumbnail',
 607             # already_have_thumbnail = True prevents the file from being deleted after embedding
 608             'already_have_thumbnail': opts.writethumbnail
 609         }
 610         if not opts.writethumbnail:
 611             opts.writethumbnail = True
 612             opts.outtmpl['pl_thumbnail'] = ''
 613     if opts.split_chapters:
 614         yield {
 615             'key': 'FFmpegSplitChapters',
 616             'force_keyframes': opts.force_keyframes_at_cuts,
 617         }
 618     # XAttrMetadataPP should be run after post-processors that may change file contents
 619     if opts.xattrs:
 620         yield {'key': 'XAttrMetadata'}
 621     if opts.concat_playlist != 'never':
 622         yield {
 623             'key': 'FFmpegConcat',
 624             'only_multi_video': opts.concat_playlist != 'always',
 625             'when': 'playlist',
 626         }
 627     # Exec must be the last PP of each category
 628     for when, exec_cmd in opts.exec_cmd.items():
 629         yield {
 630             'key': 'Exec',
 631             'exec_cmd': exec_cmd,
 632             'when': when,
 633         }
 634
 635
 636 def parse_options(argv=None):
 637     """ @returns (parser, opts, urls, ydl_opts) """
 638     parser, opts, urls = parseOpts(argv)
 639     urls = get_urls(urls, opts.batchfile, opts.verbose)
 640
 641     set_compat_opts(opts)
 642     try:
 643         warnings, deprecation_warnings = validate_options(opts)
 644     except ValueError as err:
 645         parser.error(f'{err}\n')
 646
 647     postprocessors = list(get_postprocessors(opts))
 648
 649     print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[2:])
 650     any_getting = any(getattr(opts, k) for k in (
 651         'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
 652         'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
 653     ))
 654
 655     final_ext = (
 656         opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
 657         else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
 658         else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
 659         else None)
 660
 661     return parser, opts, urls, {
 662         'usenetrc': opts.usenetrc,
 663         'netrc_location': opts.netrc_location,
 664         'username': opts.username,
 665         'password': opts.password,
 666         'twofactor': opts.twofactor,
 667         'videopassword': opts.videopassword,
 668         'ap_mso': opts.ap_mso,
 669         'ap_username': opts.ap_username,
 670         'ap_password': opts.ap_password,
 671         'client_certificate': opts.client_certificate,
 672         'client_certificate_key': opts.client_certificate_key,
 673         'client_certificate_password': opts.client_certificate_password,
 674         'quiet': opts.quiet or any_getting or opts.print_json or bool(opts.forceprint),
 675         'no_warnings': opts.no_warnings,
 676         'forceurl': opts.geturl,
 677         'forcetitle': opts.gettitle,
 678         'forceid': opts.getid,
 679         'forcethumbnail': opts.getthumbnail,
 680         'forcedescription': opts.getdescription,
 681         'forceduration': opts.getduration,
 682         'forcefilename': opts.getfilename,
 683         'forceformat': opts.getformat,
 684         'forceprint': opts.forceprint,
 685         'print_to_file': opts.print_to_file,
 686         'forcejson': opts.dumpjson or opts.print_json,
 687         'dump_single_json': opts.dump_single_json,
 688         'force_write_download_archive': opts.force_write_download_archive,
 689         'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
 690         'skip_download': opts.skip_download,
 691         'format': opts.format,
 692         'allow_unplayable_formats': opts.allow_unplayable_formats,
 693         'ignore_no_formats_error': opts.ignore_no_formats_error,
 694         'format_sort': opts.format_sort,
 695         'format_sort_force': opts.format_sort_force,
 696         'allow_multiple_video_streams': opts.allow_multiple_video_streams,
 697         'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
 698         'check_formats': opts.check_formats,
 699         'listformats': opts.listformats,
 700         'listformats_table': opts.listformats_table,
 701         'outtmpl': opts.outtmpl,
 702         'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
 703         'paths': opts.paths,
 704         'autonumber_size': opts.autonumber_size,
 705         'autonumber_start': opts.autonumber_start,
 706         'restrictfilenames': opts.restrictfilenames,
 707         'windowsfilenames': opts.windowsfilenames,
 708         'ignoreerrors': opts.ignoreerrors,
 709         'force_generic_extractor': opts.force_generic_extractor,
 710         'ratelimit': opts.ratelimit,
 711         'throttledratelimit': opts.throttledratelimit,
 712         'overwrites': opts.overwrites,
 713         'retries': opts.retries,
 714         'file_access_retries': opts.file_access_retries,
 715         'fragment_retries': opts.fragment_retries,
 716         'extractor_retries': opts.extractor_retries,
 717         'retry_sleep_functions': opts.retry_sleep,
 718         'skip_unavailable_fragments': opts.skip_unavailable_fragments,
 719         'keep_fragments': opts.keep_fragments,
 720         'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
 721         'buffersize': opts.buffersize,
 722         'noresizebuffer': opts.noresizebuffer,
 723         'http_chunk_size': opts.http_chunk_size,
 724         'continuedl': opts.continue_dl,
 725         'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
 726         'progress_with_newline': opts.progress_with_newline,
 727         'progress_template': opts.progress_template,
 728         'playliststart': opts.playliststart,
 729         'playlistend': opts.playlistend,
 730         'playlistreverse': opts.playlist_reverse,
 731         'playlistrandom': opts.playlist_random,
 732         'noplaylist': opts.noplaylist,
 733         'logtostderr': opts.outtmpl.get('default') == '-',
 734         'consoletitle': opts.consoletitle,
 735         'nopart': opts.nopart,
 736         'updatetime': opts.updatetime,
 737         'writedescription': opts.writedescription,
 738         'writeannotations': opts.writeannotations,
 739         'writeinfojson': opts.writeinfojson,
 740         'allow_playlist_files': opts.allow_playlist_files,
 741         'clean_infojson': opts.clean_infojson,
 742         'getcomments': opts.getcomments,
 743         'writethumbnail': opts.writethumbnail is True,
 744         'write_all_thumbnails': opts.writethumbnail == 'all',
 745         'writelink': opts.writelink,
 746         'writeurllink': opts.writeurllink,
 747         'writewebloclink': opts.writewebloclink,
 748         'writedesktoplink': opts.writedesktoplink,
 749         'writesubtitles': opts.writesubtitles,
 750         'writeautomaticsub': opts.writeautomaticsub,
 751         'allsubtitles': opts.allsubtitles,
 752         'listsubtitles': opts.listsubtitles,
 753         'subtitlesformat': opts.subtitlesformat,
 754         'subtitleslangs': opts.subtitleslangs,
 755         'matchtitle': decodeOption(opts.matchtitle),
 756         'rejecttitle': decodeOption(opts.rejecttitle),
 757         'max_downloads': opts.max_downloads,
 758         'prefer_free_formats': opts.prefer_free_formats,
 759         'trim_file_name': opts.trim_file_name,
 760         'verbose': opts.verbose,
 761         'dump_intermediate_pages': opts.dump_intermediate_pages,
 762         'write_pages': opts.write_pages,
 763         'load_pages': opts.load_pages,
 764         'test': opts.test,
 765         'keepvideo': opts.keepvideo,
 766         'min_filesize': opts.min_filesize,
 767         'max_filesize': opts.max_filesize,
 768         'min_views': opts.min_views,
 769         'max_views': opts.max_views,
 770         'daterange': opts.date,
 771         'cachedir': opts.cachedir,
 772         'youtube_print_sig_code': opts.youtube_print_sig_code,
 773         'age_limit': opts.age_limit,
 774         'download_archive': opts.download_archive,
 775         'break_on_existing': opts.break_on_existing,
 776         'break_on_reject': opts.break_on_reject,
 777         'break_per_url': opts.break_per_url,
 778         'skip_playlist_after_errors': opts.skip_playlist_after_errors,
 779         'cookiefile': opts.cookiefile,
 780         'cookiesfrombrowser': opts.cookiesfrombrowser,
 781         'legacyserverconnect': opts.legacy_server_connect,
 782         'nocheckcertificate': opts.no_check_certificate,
 783         'prefer_insecure': opts.prefer_insecure,
 784         'http_headers': opts.headers,
 785         'proxy': opts.proxy,
 786         'socket_timeout': opts.socket_timeout,
 787         'bidi_workaround': opts.bidi_workaround,
 788         'debug_printtraffic': opts.debug_printtraffic,
 789         'prefer_ffmpeg': opts.prefer_ffmpeg,
 790         'include_ads': opts.include_ads,
 791         'default_search': opts.default_search,
 792         'dynamic_mpd': opts.dynamic_mpd,
 793         'extractor_args': opts.extractor_args,
 794         'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
 795         'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
 796         'encoding': opts.encoding,
 797         'extract_flat': opts.extract_flat,
 798         'live_from_start': opts.live_from_start,
 799         'wait_for_video': opts.wait_for_video,
 800         'mark_watched': opts.mark_watched,
 801         'merge_output_format': opts.merge_output_format,
 802         'final_ext': final_ext,
 803         'postprocessors': postprocessors,
 804         'fixup': opts.fixup,
 805         'source_address': opts.source_address,
 806         'call_home': opts.call_home,
 807         'sleep_interval_requests': opts.sleep_interval_requests,
 808         'sleep_interval': opts.sleep_interval,
 809         'max_sleep_interval': opts.max_sleep_interval,
 810         'sleep_interval_subtitles': opts.sleep_interval_subtitles,
 811         'external_downloader': opts.external_downloader,
 812         'download_ranges': opts.download_ranges,
 813         'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
 814         'list_thumbnails': opts.list_thumbnails,
 815         'playlist_items': opts.playlist_items,
 816         'xattr_set_filesize': opts.xattr_set_filesize,
 817         'match_filter': opts.match_filter,
 818         'no_color': opts.no_color,
 819         'ffmpeg_location': opts.ffmpeg_location,
 820         'hls_prefer_native': opts.hls_prefer_native,
 821         'hls_use_mpegts': opts.hls_use_mpegts,
 822         'hls_split_discontinuity': opts.hls_split_discontinuity,
 823         'external_downloader_args': opts.external_downloader_args,
 824         'postprocessor_args': opts.postprocessor_args,
 825         'cn_verification_proxy': opts.cn_verification_proxy,
 826         'geo_verification_proxy': opts.geo_verification_proxy,
 827         'geo_bypass': opts.geo_bypass,
 828         'geo_bypass_country': opts.geo_bypass_country,
 829         'geo_bypass_ip_block': opts.geo_bypass_ip_block,
 830         '_warnings': warnings,
 831         '_deprecation_warnings': deprecation_warnings,
 832         'compat_opts': opts.compat_opts,
 833     }
 834
 835
 836 def _real_main(argv=None):
 837     setproctitle('yt-dlp')
 838
 839     parser, opts, all_urls, ydl_opts = parse_options(argv)
 840
 841     # Dump user agent
 842     if opts.dump_user_agent:
 843         ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
 844         write_string(f'{ua}\n', out=sys.stdout)
 845         return
 846
 847     if print_extractor_information(opts, all_urls):
 848         return
 849
 850     with YoutubeDL(ydl_opts) as ydl:
 851         actual_use = all_urls or opts.load_info_filename
 852
 853         if opts.rm_cachedir:
 854             ydl.cache.remove()
 855
 856         if opts.update_self and run_update(ydl) and actual_use:
 857             # If updater returns True, exit. Required for windows
 858             return 100, 'ERROR: The program must exit for the update to complete'
 859
 860         if not actual_use:
 861             if opts.update_self or opts.rm_cachedir:
 862                 return ydl._download_retcode
 863
 864             ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
 865             parser.error(
 866                 'You must provide at least one URL.\n'
 867                 'Type yt-dlp --help to see a list of all options.')
 868
 869         parser.destroy()
 870         try:
 871             if opts.load_info_filename is not None:
 872                 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
 873             else:
 874                 return ydl.download(all_urls)
 875         except DownloadCancelled:
 876             ydl.to_screen('Aborting remaining downloads')
 877             return 101
 878
 879
 880 def main(argv=None):
 881     try:
 882         _exit(*variadic(_real_main(argv)))
 883     except DownloadError:
 884         _exit(1)
 885     except SameFileError as e:
 886         _exit(f'ERROR: {e}')
 887     except KeyboardInterrupt:
 888         _exit('\nERROR: Interrupted by user')
 889     except BrokenPipeError as e:
 890         # https://docs.python.org/3/library/signal.html#note-on-sigpipe
 891         devnull = os.open(os.devnull, os.O_WRONLY)
 892         os.dup2(devnull, sys.stdout.fileno())
 893         _exit(f'\nERROR: {e}')
 894     except optparse.OptParseError as e:
 895         _exit(2, f'\n{e}')
 896
 897
 898 from .extractor import gen_extractors, list_extractors
 899
 900 __all__ = [
 901     'main',
 902     'YoutubeDL',
 903     'parse_options',
 904     'gen_extractors',
 905     'list_extractors',
 906 ]