Release 2024.12.13
[yt-dlp.git] / yt_dlp / __init__.py
blob20111175b1c3f5c5696e7a55652b2924181c02cd
1 import sys
3 if sys.version_info < (3, 9):
4 raise ImportError(
5 f'You are using an unsupported version of Python. Only Python versions 3.9 and above are supported by yt-dlp') # noqa: F541
7 __license__ = 'The Unlicense'
9 import collections
10 import getpass
11 import itertools
12 import optparse
13 import os
14 import re
15 import traceback
17 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS, CookieLoadError
18 from .downloader.external import get_external_downloader
19 from .extractor import list_extractor_classes
20 from .extractor.adobepass import MSO_INFO
21 from .networking.impersonate import ImpersonateTarget
22 from .options import parseOpts
23 from .postprocessor import (
24 FFmpegExtractAudioPP,
25 FFmpegMergerPP,
26 FFmpegPostProcessor,
27 FFmpegSubtitlesConvertorPP,
28 FFmpegThumbnailsConvertorPP,
29 FFmpegVideoConvertorPP,
30 FFmpegVideoRemuxerPP,
31 MetadataFromFieldPP,
32 MetadataParserPP,
34 from .update import Updater
35 from .utils import (
36 Config,
37 NO_DEFAULT,
38 POSTPROCESS_WHEN,
39 DateRange,
40 DownloadCancelled,
41 DownloadError,
42 FormatSorter,
43 GeoUtils,
44 PlaylistEntries,
45 SameFileError,
46 download_range_func,
47 expand_path,
48 float_or_none,
49 format_field,
50 int_or_none,
51 join_nonempty,
52 match_filter_func,
53 parse_bytes,
54 parse_duration,
55 preferredencoding,
56 read_batch_urls,
57 read_stdin,
58 render_table,
59 setproctitle,
60 shell_quote,
61 traverse_obj,
62 variadic,
63 write_string,
65 from .utils.networking import std_headers
66 from .utils._utils import _UnsafeExtensionError
67 from .YoutubeDL import YoutubeDL
69 _IN_CLI = False
72 def _exit(status=0, *args):
73 for msg in args:
74 sys.stderr.write(msg)
75 raise SystemExit(status)
78 def get_urls(urls, batchfile, verbose):
79 """
80 @param verbose -1: quiet, 0: normal, 1: verbose
81 """
82 batch_urls = []
83 if batchfile is not None:
84 try:
85 batch_urls = read_batch_urls(
86 read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
87 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
88 if verbose == 1:
89 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
90 except OSError:
91 _exit(f'ERROR: batch file {batchfile} could not be read')
92 _enc = preferredencoding()
93 return [
94 url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
95 for url in batch_urls + urls]
98 def print_extractor_information(opts, urls):
99 out = ''
100 if opts.list_extractors:
101 # Importing GenericIE is currently slow since it imports YoutubeIE
102 from .extractor.generic import GenericIE
104 urls = dict.fromkeys(urls, False)
105 for ie in list_extractor_classes(opts.age_limit):
106 out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
107 if ie == GenericIE:
108 matched_urls = [url for url, matched in urls.items() if not matched]
109 else:
110 matched_urls = tuple(filter(ie.suitable, urls.keys()))
111 urls.update(dict.fromkeys(matched_urls, True))
112 out += ''.join(f' {url}\n' for url in matched_urls)
113 elif opts.list_extractor_descriptions:
114 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
115 out = '\n'.join(
116 ie.description(markdown=False, search_examples=_SEARCHES)
117 for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
118 elif opts.ap_list_mso:
119 out = 'Supported TV Providers:\n{}\n'.format(render_table(
120 ['mso', 'mso name'],
121 [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]))
122 else:
123 return False
124 write_string(out, out=sys.stdout)
125 return True
128 def set_compat_opts(opts):
129 def _unused_compat_opt(name):
130 if name not in opts.compat_opts:
131 return False
132 opts.compat_opts.discard(name)
133 opts.compat_opts.update([f'*{name}'])
134 return True
136 def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
137 attr = getattr(opts, opt_name)
138 if compat_name in opts.compat_opts:
139 if attr is None:
140 setattr(opts, opt_name, not default)
141 return True
142 else:
143 if remove_compat:
144 _unused_compat_opt(compat_name)
145 return False
146 elif attr is None:
147 setattr(opts, opt_name, default)
148 return None
150 set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
151 set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
152 set_default_compat('no-clean-infojson', 'clean_infojson')
153 if 'no-attach-info-json' in opts.compat_opts:
154 if opts.embed_infojson:
155 _unused_compat_opt('no-attach-info-json')
156 else:
157 opts.embed_infojson = False
158 if 'format-sort' in opts.compat_opts:
159 opts.format_sort.extend(FormatSorter.ytdl_default)
160 elif 'prefer-vp9-sort' in opts.compat_opts:
161 opts.format_sort.extend(FormatSorter._prefer_vp9_sort)
163 _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
164 _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
165 if _video_multistreams_set is False and _audio_multistreams_set is False:
166 _unused_compat_opt('multistreams')
167 if 'filename' in opts.compat_opts:
168 if opts.outtmpl.get('default') is None:
169 opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
170 else:
171 _unused_compat_opt('filename')
174 def validate_options(opts):
175 def validate(cndn, name, value=None, msg=None):
176 if cndn:
177 return True
178 raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
180 def validate_in(name, value, items, msg=None):
181 return validate(value is None or value in items, name, value, msg)
183 def validate_regex(name, value, regex):
184 return validate(value is None or re.match(regex, value), name, value)
186 def validate_positive(name, value, strict=False):
187 return validate(value is None or value > 0 or (not strict and value == 0),
188 name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
190 def validate_minmax(min_val, max_val, min_name, max_name=None):
191 if max_val is None or min_val is None or max_val >= min_val:
192 return
193 if not max_name:
194 min_name, max_name = f'min {min_name}', f'max {min_name}'
195 raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
197 # Usernames and passwords
198 validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
199 msg='{name}, netrc command and username/password are mutually exclusive options')
200 validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
201 validate(opts.ap_password is None or opts.ap_username is not None,
202 'TV Provider account username', msg='{name} missing')
203 validate_in('TV Provider', opts.ap_mso, MSO_INFO,
204 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
206 # Numbers
207 validate_positive('autonumber start', opts.autonumber_start)
208 validate_positive('autonumber size', opts.autonumber_size, True)
209 validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
210 validate_positive('playlist start', opts.playliststart, True)
211 if opts.playlistend != -1:
212 validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
214 # Time ranges
215 validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
216 validate_positive('requests sleep interval', opts.sleep_interval_requests)
217 validate_positive('sleep interval', opts.sleep_interval)
218 validate_positive('max sleep interval', opts.max_sleep_interval)
219 if opts.sleep_interval is None:
220 validate(
221 opts.max_sleep_interval is None, 'min sleep interval',
222 msg='{name} must be specified; use --min-sleep-interval')
223 elif opts.max_sleep_interval is None:
224 opts.max_sleep_interval = opts.sleep_interval
225 else:
226 validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
228 if opts.wait_for_video is not None:
229 min_wait, max_wait, *_ = map(parse_duration, [*opts.wait_for_video.split('-', 1), None])
230 validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
231 'time range to wait for video', opts.wait_for_video)
232 validate_minmax(min_wait, max_wait, 'time range to wait for video')
233 opts.wait_for_video = (min_wait, max_wait)
235 # Format sort
236 for f in opts.format_sort:
237 validate_regex('format sorting', f, FormatSorter.regex)
239 # Postprocessor formats
240 if opts.convertsubtitles == 'none':
241 opts.convertsubtitles = None
242 if opts.convertthumbnails == 'none':
243 opts.convertthumbnails = None
245 validate_regex('merge output format', opts.merge_output_format,
246 r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
247 validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
248 validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
249 validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
250 validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
251 validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
252 if opts.audioquality:
253 opts.audioquality = opts.audioquality.strip('k').strip('K')
254 # int_or_none prevents inf, nan
255 validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
257 # Retries
258 def parse_retries(name, value):
259 if value is None:
260 return None
261 elif value in ('inf', 'infinite'):
262 return float('inf')
263 try:
264 return int(value)
265 except (TypeError, ValueError):
266 validate(False, f'{name} retry count', value)
268 opts.retries = parse_retries('download', opts.retries)
269 opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
270 opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
271 opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
273 # Retry sleep function
274 def parse_sleep_func(expr):
275 NUMBER_RE = r'\d+(?:\.\d+)?'
276 op, start, limit, step, *_ = (*tuple(re.fullmatch(
277 rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
278 expr.strip()).groups()), None, None)
280 if op == 'exp':
281 return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
282 else:
283 default_step = start if op or limit else 0
284 return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
286 for key, expr in opts.retry_sleep.items():
287 if not expr:
288 del opts.retry_sleep[key]
289 continue
290 try:
291 opts.retry_sleep[key] = parse_sleep_func(expr)
292 except AttributeError:
293 raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
295 # Bytes
296 def validate_bytes(name, value):
297 if value is None:
298 return None
299 numeric_limit = parse_bytes(value)
300 validate(numeric_limit is not None, 'rate limit', value)
301 return numeric_limit
303 opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
304 opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
305 opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
306 opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
307 opts.buffersize = validate_bytes('buffer size', opts.buffersize)
308 opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
310 # Output templates
311 def validate_outtmpl(tmpl, msg):
312 err = YoutubeDL.validate_outtmpl(tmpl)
313 if err:
314 raise ValueError(f'invalid {msg} "{tmpl}": {err}')
316 for k, tmpl in opts.outtmpl.items():
317 validate_outtmpl(tmpl, f'{k} output template')
318 for type_, tmpl_list in opts.forceprint.items():
319 for tmpl in tmpl_list:
320 validate_outtmpl(tmpl, f'{type_} print template')
321 for type_, tmpl_list in opts.print_to_file.items():
322 for tmpl, file in tmpl_list:
323 validate_outtmpl(tmpl, f'{type_} print to file template')
324 validate_outtmpl(file, f'{type_} print to file filename')
325 validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
326 for k, tmpl in opts.progress_template.items():
327 k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
328 validate_outtmpl(tmpl, f'{k} template')
330 outtmpl_default = opts.outtmpl.get('default')
331 if outtmpl_default == '':
332 opts.skip_download = None
333 del opts.outtmpl['default']
335 def parse_chapters(name, value, advanced=False):
336 parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
337 TIMESTAMP_RE = r'''(?x)(?:
338 (?P<start_sign>-?)(?P<start>[^-]+)
339 )?\s*-\s*(?:
340 (?P<end_sign>-?)(?P<end>[^-]+)
341 )?'''
343 chapters, ranges, from_url = [], [], False
344 for regex in value or []:
345 if advanced and regex == '*from-url':
346 from_url = True
347 continue
348 elif not regex.startswith('*'):
349 try:
350 chapters.append(re.compile(regex))
351 except re.error as err:
352 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
353 continue
355 for range_ in map(str.strip, regex[1:].split(',')):
356 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
357 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
358 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
360 err = None
361 if None in (dur or [None]):
362 err = 'Must be of the form "*start-end"'
363 elif not advanced and any(signs):
364 err = 'Negative timestamps are not allowed'
365 else:
366 dur[0] *= -1 if signs[0] else 1
367 dur[1] *= -1 if signs[1] else 1
368 if dur[1] == float('-inf'):
369 err = '"-inf" is not a valid end'
370 if err:
371 raise ValueError(f'invalid {name} time range "{regex}". {err}')
372 ranges.append(dur)
374 return chapters, ranges, from_url
376 opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
377 opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
379 # Cookies from browser
380 if opts.cookiesfrombrowser:
381 container = None
382 mobj = re.fullmatch(r'''(?x)
383 (?P<name>[^+:]+)
384 (?:\s*\+\s*(?P<keyring>[^:]+))?
385 (?:\s*:\s*(?!:)(?P<profile>.+?))?
386 (?:\s*::\s*(?P<container>.+))?
387 ''', opts.cookiesfrombrowser)
388 if mobj is None:
389 raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
390 browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
391 browser_name = browser_name.lower()
392 if browser_name not in SUPPORTED_BROWSERS:
393 raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
394 f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
395 if keyring is not None:
396 keyring = keyring.upper()
397 if keyring not in SUPPORTED_KEYRINGS:
398 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
399 f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
400 opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
402 if opts.impersonate is not None:
403 opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
405 # MetadataParser
406 def metadataparser_actions(f):
407 if isinstance(f, str):
408 cmd = f'--parse-metadata {shell_quote(f)}'
409 try:
410 actions = [MetadataFromFieldPP.to_action(f)]
411 except Exception as err:
412 raise ValueError(f'{cmd} is invalid; {err}')
413 else:
414 cmd = f'--replace-in-metadata {shell_quote(f)}'
415 actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
417 for action in actions:
418 try:
419 MetadataParserPP.validate_action(*action)
420 except Exception as err:
421 raise ValueError(f'{cmd} is invalid; {err}')
422 yield action
424 if opts.metafromtitle is not None:
425 opts.parse_metadata.setdefault('pre_process', []).append(f'title:{opts.metafromtitle}')
426 opts.parse_metadata = {
427 k: list(itertools.chain(*map(metadataparser_actions, v)))
428 for k, v in opts.parse_metadata.items()
431 # Other options
432 if opts.playlist_items is not None:
433 try:
434 tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
435 except Exception as err:
436 raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
438 opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
439 if opts.geo_bypass.lower() not in ('default', 'never'):
440 try:
441 GeoUtils.random_ipv4(opts.geo_bypass)
442 except Exception:
443 raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
444 if len(opts.geo_bypass) == 2:
445 opts.geo_bypass_country = opts.geo_bypass
446 else:
447 opts.geo_bypass_ip_block = opts.geo_bypass
448 opts.geo_bypass = opts.geo_bypass.lower() != 'never'
450 opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
452 if opts.download_archive is not None:
453 opts.download_archive = expand_path(opts.download_archive)
455 if opts.ffmpeg_location is not None:
456 opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
458 if opts.user_agent is not None:
459 opts.headers.setdefault('User-Agent', opts.user_agent)
460 if opts.referer is not None:
461 opts.headers.setdefault('Referer', opts.referer)
463 if opts.no_sponsorblock:
464 opts.sponsorblock_mark = opts.sponsorblock_remove = set()
466 default_downloader = None
467 for proto, path in opts.external_downloader.items():
468 if path == 'native':
469 continue
470 ed = get_external_downloader(path)
471 if ed is None:
472 raise ValueError(
473 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
474 elif ed and proto == 'default':
475 default_downloader = ed.get_basename()
477 for policy in opts.color.values():
478 if policy not in ('always', 'auto', 'auto-tty', 'no_color', 'no_color-tty', 'never'):
479 raise ValueError(f'"{policy}" is not a valid color policy')
481 warnings, deprecation_warnings = [], []
483 # Common mistake: -f best
484 if opts.format == 'best':
485 warnings.append('.\n '.join((
486 '"-f best" selects the best pre-merged format which is often not the best option',
487 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
488 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
490 # --(postprocessor/downloader)-args without name
491 def report_args_compat(name, value, key1, key2=None, where=None):
492 if key1 in value and key2 not in value:
493 warnings.append(f'{name.title()} arguments given without specifying name. '
494 f'The arguments will be given to {where or f"all {name}s"}')
495 return True
496 return False
498 if report_args_compat('external downloader', opts.external_downloader_args,
499 'default', where=default_downloader) and default_downloader:
500 # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
501 opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
503 if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
504 opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
505 opts.postprocessor_args.setdefault('sponskrub', [])
507 def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
508 val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
509 if val2 is NO_DEFAULT:
510 val2 = getattr(opts, opt2)
511 if not val2:
512 return
514 if val1 is NO_DEFAULT:
515 val1 = getattr(opts, opt1)
516 if val1:
517 warnings.append(f'{arg1} is ignored since {arg2} was given')
518 setattr(opts, opt1, default)
520 # Conflicting options
521 report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
522 report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
523 report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
524 report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
525 report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
526 report_conflict('--exec-before-download', 'exec_before_dl_cmd',
527 '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
528 report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
529 report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
530 report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
531 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
532 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
533 report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
534 val1=opts.sponskrub and opts.sponskrub_cut)
536 # Conflicts with --allow-unplayable-formats
537 report_conflict('--embed-metadata', 'addmetadata')
538 report_conflict('--embed-chapters', 'addchapters')
539 report_conflict('--embed-info-json', 'embed_infojson')
540 report_conflict('--embed-subs', 'embedsubtitles')
541 report_conflict('--embed-thumbnail', 'embedthumbnail')
542 report_conflict('--extract-audio', 'extractaudio')
543 report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
544 report_conflict('--recode-video', 'recodevideo')
545 report_conflict('--remove-chapters', 'remove_chapters', default=[])
546 report_conflict('--remux-video', 'remuxvideo')
547 report_conflict('--sponskrub', 'sponskrub')
548 report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
549 report_conflict('--xattrs', 'xattrs')
551 # Fully deprecated options
552 def report_deprecation(val, old, new=None):
553 if not val:
554 return
555 deprecation_warnings.append(
556 f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
557 else f'{old} is deprecated and may not work as expected')
559 report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
560 report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
561 # report_deprecation(opts.include_ads, '--include-ads') # We may re-implement this in future
562 # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
563 # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
565 # Dependent options
566 opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
568 if opts.exec_before_dl_cmd:
569 opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
571 if opts.useid: # --id is not deprecated in youtube-dl
572 opts.outtmpl['default'] = '%(id)s.%(ext)s'
574 if opts.overwrites: # --force-overwrites implies --no-continue
575 opts.continue_dl = False
577 if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
578 # Add chapters when adding metadata or marking sponsors
579 opts.addchapters = True
581 if opts.extractaudio and not opts.keepvideo and opts.format is None:
582 # Do not unnecessarily download audio
583 opts.format = 'bestaudio/best'
585 if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
586 # If JSON is not printed anywhere, but comments are requested, save it to file
587 if not opts.dumpjson or opts.print_json or opts.dump_single_json:
588 opts.writeinfojson = True
590 if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
591 # --all-sub automatically sets --write-sub if --write-auto-sub is not given
592 opts.writesubtitles = True
594 if opts.addmetadata and opts.embed_infojson is None:
595 # If embedding metadata and infojson is present, embed it
596 opts.embed_infojson = 'if_exists'
598 # Ask for passwords
599 if opts.username is not None and opts.password is None:
600 opts.password = getpass.getpass('Type account password and press [Return]: ')
601 if opts.ap_username is not None and opts.ap_password is None:
602 opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
604 # compat option changes global state destructively; only allow from cli
605 if 'allow-unsafe-ext' in opts.compat_opts:
606 warnings.append(
607 'Using allow-unsafe-ext opens you up to potential attacks. '
608 'Use with great care!')
609 _UnsafeExtensionError.sanitize_extension = lambda x, prepend=False: x
611 return warnings, deprecation_warnings
614 def get_postprocessors(opts):
615 yield from opts.add_postprocessors
617 for when, actions in opts.parse_metadata.items():
618 yield {
619 'key': 'MetadataParser',
620 'actions': actions,
621 'when': when,
623 sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
624 if sponsorblock_query:
625 yield {
626 'key': 'SponsorBlock',
627 'categories': sponsorblock_query,
628 'api': opts.sponsorblock_api,
629 'when': 'after_filter',
631 if opts.convertsubtitles:
632 yield {
633 'key': 'FFmpegSubtitlesConvertor',
634 'format': opts.convertsubtitles,
635 'when': 'before_dl',
637 if opts.convertthumbnails:
638 yield {
639 'key': 'FFmpegThumbnailsConvertor',
640 'format': opts.convertthumbnails,
641 'when': 'before_dl',
643 if opts.extractaudio:
644 yield {
645 'key': 'FFmpegExtractAudio',
646 'preferredcodec': opts.audioformat,
647 'preferredquality': opts.audioquality,
648 'nopostoverwrites': opts.nopostoverwrites,
650 if opts.remuxvideo:
651 yield {
652 'key': 'FFmpegVideoRemuxer',
653 'preferedformat': opts.remuxvideo,
655 if opts.recodevideo:
656 yield {
657 'key': 'FFmpegVideoConvertor',
658 'preferedformat': opts.recodevideo,
660 # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
661 if opts.embedsubtitles:
662 keep_subs = 'no-keep-subs' not in opts.compat_opts
663 yield {
664 'key': 'FFmpegEmbedSubtitle',
665 # already_have_subtitle = True prevents the file from being deleted after embedding
666 'already_have_subtitle': opts.writesubtitles and keep_subs,
668 if not opts.writeautomaticsub and keep_subs:
669 opts.writesubtitles = True
671 # ModifyChapters must run before FFmpegMetadataPP
672 if opts.remove_chapters or sponsorblock_query:
673 yield {
674 'key': 'ModifyChapters',
675 'remove_chapters_patterns': opts.remove_chapters,
676 'remove_sponsor_segments': opts.sponsorblock_remove,
677 'remove_ranges': opts.remove_ranges,
678 'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
679 'force_keyframes': opts.force_keyframes_at_cuts,
681 # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
682 # FFmpegExtractAudioPP as containers before conversion may not support
683 # metadata (3gp, webm, etc.)
684 # By default ffmpeg preserves metadata applicable for both
685 # source and target containers. From this point the container won't change,
686 # so metadata can be added here.
687 if opts.addmetadata or opts.addchapters or opts.embed_infojson:
688 yield {
689 'key': 'FFmpegMetadata',
690 'add_chapters': opts.addchapters,
691 'add_metadata': opts.addmetadata,
692 'add_infojson': opts.embed_infojson,
694 # Deprecated
695 # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
696 # but must be below EmbedSubtitle and FFmpegMetadata
697 # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
698 # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
699 if opts.sponskrub is not False:
700 yield {
701 'key': 'SponSkrub',
702 'path': opts.sponskrub_path,
703 'args': opts.sponskrub_args,
704 'cut': opts.sponskrub_cut,
705 'force': opts.sponskrub_force,
706 'ignoreerror': opts.sponskrub is None,
707 '_from_cli': True,
709 if opts.embedthumbnail:
710 yield {
711 'key': 'EmbedThumbnail',
712 # already_have_thumbnail = True prevents the file from being deleted after embedding
713 'already_have_thumbnail': opts.writethumbnail,
715 if not opts.writethumbnail:
716 opts.writethumbnail = True
717 opts.outtmpl['pl_thumbnail'] = ''
718 if opts.split_chapters:
719 yield {
720 'key': 'FFmpegSplitChapters',
721 'force_keyframes': opts.force_keyframes_at_cuts,
723 # XAttrMetadataPP should be run after post-processors that may change file contents
724 if opts.xattrs:
725 yield {'key': 'XAttrMetadata'}
726 if opts.concat_playlist != 'never':
727 yield {
728 'key': 'FFmpegConcat',
729 'only_multi_video': opts.concat_playlist != 'always',
730 'when': 'playlist',
732 # Exec must be the last PP of each category
733 for when, exec_cmd in opts.exec_cmd.items():
734 yield {
735 'key': 'Exec',
736 'exec_cmd': exec_cmd,
737 'when': when,
741 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
744 def parse_options(argv=None):
745 """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
746 parser, opts, urls = parseOpts(argv)
747 urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
749 set_compat_opts(opts)
750 try:
751 warnings, deprecation_warnings = validate_options(opts)
752 except ValueError as err:
753 parser.error(f'{err}\n')
755 postprocessors = list(get_postprocessors(opts))
757 print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
758 any_getting = any(getattr(opts, k) for k in (
759 'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
760 'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl',
762 if opts.quiet is None:
763 opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
765 playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
766 write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
767 and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
768 if not any((
769 opts.extract_flat,
770 opts.dump_single_json,
771 opts.forceprint.get('playlist'),
772 opts.print_to_file.get('playlist'),
773 write_playlist_infojson,
775 if not playlist_pps:
776 opts.extract_flat = 'discard'
777 elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
778 opts.extract_flat = 'discard_in_playlist'
780 final_ext = (
781 opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
782 else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
783 else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
784 else None)
786 return ParsedOptions(parser, opts, urls, {
787 'usenetrc': opts.usenetrc,
788 'netrc_location': opts.netrc_location,
789 'netrc_cmd': opts.netrc_cmd,
790 'username': opts.username,
791 'password': opts.password,
792 'twofactor': opts.twofactor,
793 'videopassword': opts.videopassword,
794 'ap_mso': opts.ap_mso,
795 'ap_username': opts.ap_username,
796 'ap_password': opts.ap_password,
797 'client_certificate': opts.client_certificate,
798 'client_certificate_key': opts.client_certificate_key,
799 'client_certificate_password': opts.client_certificate_password,
800 'quiet': opts.quiet,
801 'no_warnings': opts.no_warnings,
802 'forceurl': opts.geturl,
803 'forcetitle': opts.gettitle,
804 'forceid': opts.getid,
805 'forcethumbnail': opts.getthumbnail,
806 'forcedescription': opts.getdescription,
807 'forceduration': opts.getduration,
808 'forcefilename': opts.getfilename,
809 'forceformat': opts.getformat,
810 'forceprint': opts.forceprint,
811 'print_to_file': opts.print_to_file,
812 'forcejson': opts.dumpjson or opts.print_json,
813 'dump_single_json': opts.dump_single_json,
814 'force_write_download_archive': opts.force_write_download_archive,
815 'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
816 'skip_download': opts.skip_download,
817 'format': opts.format,
818 'allow_unplayable_formats': opts.allow_unplayable_formats,
819 'ignore_no_formats_error': opts.ignore_no_formats_error,
820 'format_sort': opts.format_sort,
821 'format_sort_force': opts.format_sort_force,
822 'allow_multiple_video_streams': opts.allow_multiple_video_streams,
823 'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
824 'check_formats': opts.check_formats,
825 'listformats': opts.listformats,
826 'listformats_table': opts.listformats_table,
827 'outtmpl': opts.outtmpl,
828 'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
829 'paths': opts.paths,
830 'autonumber_size': opts.autonumber_size,
831 'autonumber_start': opts.autonumber_start,
832 'restrictfilenames': opts.restrictfilenames,
833 'windowsfilenames': opts.windowsfilenames,
834 'ignoreerrors': opts.ignoreerrors,
835 'force_generic_extractor': opts.force_generic_extractor,
836 'allowed_extractors': opts.allowed_extractors or ['default'],
837 'ratelimit': opts.ratelimit,
838 'throttledratelimit': opts.throttledratelimit,
839 'overwrites': opts.overwrites,
840 'retries': opts.retries,
841 'file_access_retries': opts.file_access_retries,
842 'fragment_retries': opts.fragment_retries,
843 'extractor_retries': opts.extractor_retries,
844 'retry_sleep_functions': opts.retry_sleep,
845 'skip_unavailable_fragments': opts.skip_unavailable_fragments,
846 'keep_fragments': opts.keep_fragments,
847 'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
848 'buffersize': opts.buffersize,
849 'noresizebuffer': opts.noresizebuffer,
850 'http_chunk_size': opts.http_chunk_size,
851 'continuedl': opts.continue_dl,
852 'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
853 'progress_with_newline': opts.progress_with_newline,
854 'progress_template': opts.progress_template,
855 'progress_delta': opts.progress_delta,
856 'playliststart': opts.playliststart,
857 'playlistend': opts.playlistend,
858 'playlistreverse': opts.playlist_reverse,
859 'playlistrandom': opts.playlist_random,
860 'lazy_playlist': opts.lazy_playlist,
861 'noplaylist': opts.noplaylist,
862 'logtostderr': opts.outtmpl.get('default') == '-',
863 'consoletitle': opts.consoletitle,
864 'nopart': opts.nopart,
865 'updatetime': opts.updatetime,
866 'writedescription': opts.writedescription,
867 'writeannotations': opts.writeannotations,
868 'writeinfojson': opts.writeinfojson,
869 'allow_playlist_files': opts.allow_playlist_files,
870 'clean_infojson': opts.clean_infojson,
871 'getcomments': opts.getcomments,
872 'writethumbnail': opts.writethumbnail is True,
873 'write_all_thumbnails': opts.writethumbnail == 'all',
874 'writelink': opts.writelink,
875 'writeurllink': opts.writeurllink,
876 'writewebloclink': opts.writewebloclink,
877 'writedesktoplink': opts.writedesktoplink,
878 'writesubtitles': opts.writesubtitles,
879 'writeautomaticsub': opts.writeautomaticsub,
880 'allsubtitles': opts.allsubtitles,
881 'listsubtitles': opts.listsubtitles,
882 'subtitlesformat': opts.subtitlesformat,
883 'subtitleslangs': opts.subtitleslangs,
884 'matchtitle': opts.matchtitle,
885 'rejecttitle': opts.rejecttitle,
886 'max_downloads': opts.max_downloads,
887 'prefer_free_formats': opts.prefer_free_formats,
888 'trim_file_name': opts.trim_file_name,
889 'verbose': opts.verbose,
890 'dump_intermediate_pages': opts.dump_intermediate_pages,
891 'write_pages': opts.write_pages,
892 'load_pages': opts.load_pages,
893 'test': opts.test,
894 'keepvideo': opts.keepvideo,
895 'min_filesize': opts.min_filesize,
896 'max_filesize': opts.max_filesize,
897 'min_views': opts.min_views,
898 'max_views': opts.max_views,
899 'daterange': opts.date,
900 'cachedir': opts.cachedir,
901 'youtube_print_sig_code': opts.youtube_print_sig_code,
902 'age_limit': opts.age_limit,
903 'download_archive': opts.download_archive,
904 'break_on_existing': opts.break_on_existing,
905 'break_on_reject': opts.break_on_reject,
906 'break_per_url': opts.break_per_url,
907 'skip_playlist_after_errors': opts.skip_playlist_after_errors,
908 'cookiefile': opts.cookiefile,
909 'cookiesfrombrowser': opts.cookiesfrombrowser,
910 'legacyserverconnect': opts.legacy_server_connect,
911 'nocheckcertificate': opts.no_check_certificate,
912 'prefer_insecure': opts.prefer_insecure,
913 'enable_file_urls': opts.enable_file_urls,
914 'http_headers': opts.headers,
915 'proxy': opts.proxy,
916 'socket_timeout': opts.socket_timeout,
917 'bidi_workaround': opts.bidi_workaround,
918 'debug_printtraffic': opts.debug_printtraffic,
919 'prefer_ffmpeg': opts.prefer_ffmpeg,
920 'include_ads': opts.include_ads,
921 'default_search': opts.default_search,
922 'dynamic_mpd': opts.dynamic_mpd,
923 'extractor_args': opts.extractor_args,
924 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
925 'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
926 'encoding': opts.encoding,
927 'extract_flat': opts.extract_flat,
928 'live_from_start': opts.live_from_start,
929 'wait_for_video': opts.wait_for_video,
930 'mark_watched': opts.mark_watched,
931 'merge_output_format': opts.merge_output_format,
932 'final_ext': final_ext,
933 'postprocessors': postprocessors,
934 'fixup': opts.fixup,
935 'source_address': opts.source_address,
936 'impersonate': opts.impersonate,
937 'call_home': opts.call_home,
938 'sleep_interval_requests': opts.sleep_interval_requests,
939 'sleep_interval': opts.sleep_interval,
940 'max_sleep_interval': opts.max_sleep_interval,
941 'sleep_interval_subtitles': opts.sleep_interval_subtitles,
942 'external_downloader': opts.external_downloader,
943 'download_ranges': opts.download_ranges,
944 'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
945 'list_thumbnails': opts.list_thumbnails,
946 'playlist_items': opts.playlist_items,
947 'xattr_set_filesize': opts.xattr_set_filesize,
948 'match_filter': opts.match_filter,
949 'color': opts.color,
950 'ffmpeg_location': opts.ffmpeg_location,
951 'hls_prefer_native': opts.hls_prefer_native,
952 'hls_use_mpegts': opts.hls_use_mpegts,
953 'hls_split_discontinuity': opts.hls_split_discontinuity,
954 'external_downloader_args': opts.external_downloader_args,
955 'postprocessor_args': opts.postprocessor_args,
956 'cn_verification_proxy': opts.cn_verification_proxy,
957 'geo_verification_proxy': opts.geo_verification_proxy,
958 'geo_bypass': opts.geo_bypass,
959 'geo_bypass_country': opts.geo_bypass_country,
960 'geo_bypass_ip_block': opts.geo_bypass_ip_block,
961 '_warnings': warnings,
962 '_deprecation_warnings': deprecation_warnings,
963 'compat_opts': opts.compat_opts,
967 def _real_main(argv=None):
968 setproctitle('yt-dlp')
970 parser, opts, all_urls, ydl_opts = parse_options(argv)
972 # HACK: Set the plugin dirs early on
973 # TODO(coletdjnz): remove when plugin globals system is implemented
974 if opts.plugin_dirs is not None:
975 Config._plugin_dirs = list(map(expand_path, opts.plugin_dirs))
977 # Dump user agent
978 if opts.dump_user_agent:
979 ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
980 write_string(f'{ua}\n', out=sys.stdout)
981 return
983 if print_extractor_information(opts, all_urls):
984 return
986 # We may need ffmpeg_location without having access to the YoutubeDL instance
987 # See https://github.com/yt-dlp/yt-dlp/issues/2191
988 if opts.ffmpeg_location:
989 FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
991 with YoutubeDL(ydl_opts) as ydl:
992 pre_process = opts.update_self or opts.rm_cachedir
993 actual_use = all_urls or opts.load_info_filename
995 if opts.rm_cachedir:
996 ydl.cache.remove()
998 try:
999 updater = Updater(ydl, opts.update_self)
1000 if opts.update_self and updater.update() and actual_use:
1001 if updater.cmd:
1002 return updater.restart()
1003 # This code is reachable only for zip variant in py < 3.10
1004 # It makes sense to exit here, but the old behavior is to continue
1005 ydl.report_warning('Restart yt-dlp to use the updated version')
1006 # return 100, 'ERROR: The program must exit for the update to complete'
1007 except Exception:
1008 traceback.print_exc()
1009 ydl._download_retcode = 100
1011 if opts.list_impersonate_targets:
1013 known_targets = [
1014 # List of simplified targets we know are supported,
1015 # to help users know what dependencies may be required.
1016 (ImpersonateTarget('chrome'), 'curl_cffi'),
1017 (ImpersonateTarget('edge'), 'curl_cffi'),
1018 (ImpersonateTarget('safari'), 'curl_cffi'),
1021 available_targets = ydl._get_available_impersonate_targets()
1023 def make_row(target, handler):
1024 return [
1025 join_nonempty(target.client.title(), target.version, delim='-') or '-',
1026 join_nonempty((target.os or '').title(), target.os_version, delim='-') or '-',
1027 handler,
1030 rows = [make_row(target, handler) for target, handler in available_targets]
1032 for known_target, known_handler in known_targets:
1033 if not any(
1034 known_target in target and handler == known_handler
1035 for target, handler in available_targets
1037 rows.append([
1038 ydl._format_out(text, ydl.Styles.SUPPRESS)
1039 for text in make_row(known_target, f'{known_handler} (not available)')
1042 ydl.to_screen('[info] Available impersonate targets')
1043 ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
1044 return
1046 if not actual_use:
1047 if pre_process:
1048 return ydl._download_retcode
1050 args = sys.argv[1:] if argv is None else argv
1051 ydl.warn_if_short_id(args)
1053 # Show a useful error message and wait for keypress if not launched from shell on Windows
1054 if not args and os.name == 'nt' and getattr(sys, 'frozen', False):
1055 import ctypes.wintypes
1056 import msvcrt
1058 kernel32 = ctypes.WinDLL('Kernel32')
1060 buffer = (1 * ctypes.wintypes.DWORD)()
1061 attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
1062 # If we only have a single process attached, then the executable was double clicked
1063 # When using `pyinstaller` with `--onefile`, two processes get attached
1064 is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
1065 if attached_processes == 1 or (is_onefile and attached_processes == 2):
1066 print(parser._generate_error_message(
1067 'Do not double-click the executable, instead call it from a command line.\n'
1068 'Please read the README for further information on how to use yt-dlp: '
1069 'https://github.com/yt-dlp/yt-dlp#readme'))
1070 msvcrt.getch()
1071 _exit(2)
1072 parser.error(
1073 'You must provide at least one URL.\n'
1074 'Type yt-dlp --help to see a list of all options.')
1076 parser.destroy()
1077 try:
1078 if opts.load_info_filename is not None:
1079 if all_urls:
1080 ydl.report_warning('URLs are ignored due to --load-info-json')
1081 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
1082 else:
1083 return ydl.download(all_urls)
1084 except DownloadCancelled:
1085 ydl.to_screen('Aborting remaining downloads')
1086 return 101
1089 def main(argv=None):
1090 global _IN_CLI
1091 _IN_CLI = True
1092 try:
1093 _exit(*variadic(_real_main(argv)))
1094 except (CookieLoadError, DownloadError):
1095 _exit(1)
1096 except SameFileError as e:
1097 _exit(f'ERROR: {e}')
1098 except KeyboardInterrupt:
1099 _exit('\nERROR: Interrupted by user')
1100 except BrokenPipeError as e:
1101 # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1102 devnull = os.open(os.devnull, os.O_WRONLY)
1103 os.dup2(devnull, sys.stdout.fileno())
1104 _exit(f'\nERROR: {e}')
1105 except optparse.OptParseError as e:
1106 _exit(2, f'\n{e}')
1109 from .extractor import gen_extractors, list_extractors
1111 __all__ = [
1112 'YoutubeDL',
1113 'gen_extractors',
1114 'list_extractors',
1115 'main',
1116 'parse_options',