Add option `--plugin-dirs` (#11277)
[yt-dlp3.git] / yt_dlp / __init__.py
blobd976f5bbcb7d0f53a924ccdeece726a166b5790c
1 import sys
3 if sys.version_info < (3, 8):
4 raise ImportError(
5 f'You are using an unsupported version of Python. Only Python versions 3.8 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 .compat import compat_os_name
18 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS, CookieLoadError
19 from .downloader.external import get_external_downloader
20 from .extractor import list_extractor_classes
21 from .extractor.adobepass import MSO_INFO
22 from .networking.impersonate import ImpersonateTarget
23 from .options import parseOpts
24 from .postprocessor import (
25 FFmpegExtractAudioPP,
26 FFmpegMergerPP,
27 FFmpegPostProcessor,
28 FFmpegSubtitlesConvertorPP,
29 FFmpegThumbnailsConvertorPP,
30 FFmpegVideoConvertorPP,
31 FFmpegVideoRemuxerPP,
32 MetadataFromFieldPP,
33 MetadataParserPP,
35 from .update import Updater
36 from .utils import (
37 Config,
38 NO_DEFAULT,
39 POSTPROCESS_WHEN,
40 DateRange,
41 DownloadCancelled,
42 DownloadError,
43 FormatSorter,
44 GeoUtils,
45 PlaylistEntries,
46 SameFileError,
47 decodeOption,
48 download_range_func,
49 expand_path,
50 float_or_none,
51 format_field,
52 int_or_none,
53 join_nonempty,
54 match_filter_func,
55 parse_bytes,
56 parse_duration,
57 preferredencoding,
58 read_batch_urls,
59 read_stdin,
60 render_table,
61 setproctitle,
62 shell_quote,
63 traverse_obj,
64 variadic,
65 write_string,
67 from .utils.networking import std_headers
68 from .utils._utils import _UnsafeExtensionError
69 from .YoutubeDL import YoutubeDL
71 _IN_CLI = False
74 def _exit(status=0, *args):
75 for msg in args:
76 sys.stderr.write(msg)
77 raise SystemExit(status)
80 def get_urls(urls, batchfile, verbose):
81 """
82 @param verbose -1: quiet, 0: normal, 1: verbose
83 """
84 batch_urls = []
85 if batchfile is not None:
86 try:
87 batch_urls = read_batch_urls(
88 read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
89 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
90 if verbose == 1:
91 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
92 except OSError:
93 _exit(f'ERROR: batch file {batchfile} could not be read')
94 _enc = preferredencoding()
95 return [
96 url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
97 for url in batch_urls + urls]
100 def print_extractor_information(opts, urls):
101 out = ''
102 if opts.list_extractors:
103 # Importing GenericIE is currently slow since it imports YoutubeIE
104 from .extractor.generic import GenericIE
106 urls = dict.fromkeys(urls, False)
107 for ie in list_extractor_classes(opts.age_limit):
108 out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
109 if ie == GenericIE:
110 matched_urls = [url for url, matched in urls.items() if not matched]
111 else:
112 matched_urls = tuple(filter(ie.suitable, urls.keys()))
113 urls.update(dict.fromkeys(matched_urls, True))
114 out += ''.join(f' {url}\n' for url in matched_urls)
115 elif opts.list_extractor_descriptions:
116 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
117 out = '\n'.join(
118 ie.description(markdown=False, search_examples=_SEARCHES)
119 for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
120 elif opts.ap_list_mso:
121 out = 'Supported TV Providers:\n{}\n'.format(render_table(
122 ['mso', 'mso name'],
123 [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()]))
124 else:
125 return False
126 write_string(out, out=sys.stdout)
127 return True
130 def set_compat_opts(opts):
131 def _unused_compat_opt(name):
132 if name not in opts.compat_opts:
133 return False
134 opts.compat_opts.discard(name)
135 opts.compat_opts.update([f'*{name}'])
136 return True
138 def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
139 attr = getattr(opts, opt_name)
140 if compat_name in opts.compat_opts:
141 if attr is None:
142 setattr(opts, opt_name, not default)
143 return True
144 else:
145 if remove_compat:
146 _unused_compat_opt(compat_name)
147 return False
148 elif attr is None:
149 setattr(opts, opt_name, default)
150 return None
152 set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
153 set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
154 set_default_compat('no-clean-infojson', 'clean_infojson')
155 if 'no-attach-info-json' in opts.compat_opts:
156 if opts.embed_infojson:
157 _unused_compat_opt('no-attach-info-json')
158 else:
159 opts.embed_infojson = False
160 if 'format-sort' in opts.compat_opts:
161 opts.format_sort.extend(FormatSorter.ytdl_default)
162 _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
163 _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
164 if _video_multistreams_set is False and _audio_multistreams_set is False:
165 _unused_compat_opt('multistreams')
166 if 'filename' in opts.compat_opts:
167 if opts.outtmpl.get('default') is None:
168 opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
169 else:
170 _unused_compat_opt('filename')
173 def validate_options(opts):
174 def validate(cndn, name, value=None, msg=None):
175 if cndn:
176 return True
177 raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
179 def validate_in(name, value, items, msg=None):
180 return validate(value is None or value in items, name, value, msg)
182 def validate_regex(name, value, regex):
183 return validate(value is None or re.match(regex, value), name, value)
185 def validate_positive(name, value, strict=False):
186 return validate(value is None or value > 0 or (not strict and value == 0),
187 name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
189 def validate_minmax(min_val, max_val, min_name, max_name=None):
190 if max_val is None or min_val is None or max_val >= min_val:
191 return
192 if not max_name:
193 min_name, max_name = f'min {min_name}', f'max {min_name}'
194 raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
196 # Usernames and passwords
197 validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
198 msg='{name}, netrc command and username/password are mutually exclusive options')
199 validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
200 validate(opts.ap_password is None or opts.ap_username is not None,
201 'TV Provider account username', msg='{name} missing')
202 validate_in('TV Provider', opts.ap_mso, MSO_INFO,
203 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
205 # Numbers
206 validate_positive('autonumber start', opts.autonumber_start)
207 validate_positive('autonumber size', opts.autonumber_size, True)
208 validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
209 validate_positive('playlist start', opts.playliststart, True)
210 if opts.playlistend != -1:
211 validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
213 # Time ranges
214 validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
215 validate_positive('requests sleep interval', opts.sleep_interval_requests)
216 validate_positive('sleep interval', opts.sleep_interval)
217 validate_positive('max sleep interval', opts.max_sleep_interval)
218 if opts.sleep_interval is None:
219 validate(
220 opts.max_sleep_interval is None, 'min sleep interval',
221 msg='{name} must be specified; use --min-sleep-interval')
222 elif opts.max_sleep_interval is None:
223 opts.max_sleep_interval = opts.sleep_interval
224 else:
225 validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
227 if opts.wait_for_video is not None:
228 min_wait, max_wait, *_ = map(parse_duration, [*opts.wait_for_video.split('-', 1), None])
229 validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
230 'time range to wait for video', opts.wait_for_video)
231 validate_minmax(min_wait, max_wait, 'time range to wait for video')
232 opts.wait_for_video = (min_wait, max_wait)
234 # Format sort
235 for f in opts.format_sort:
236 validate_regex('format sorting', f, FormatSorter.regex)
238 # Postprocessor formats
239 if opts.convertsubtitles == 'none':
240 opts.convertsubtitles = None
241 if opts.convertthumbnails == 'none':
242 opts.convertthumbnails = None
244 validate_regex('merge output format', opts.merge_output_format,
245 r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
246 validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
247 validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
248 validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
249 validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
250 validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
251 if opts.audioquality:
252 opts.audioquality = opts.audioquality.strip('k').strip('K')
253 # int_or_none prevents inf, nan
254 validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
256 # Retries
257 def parse_retries(name, value):
258 if value is None:
259 return None
260 elif value in ('inf', 'infinite'):
261 return float('inf')
262 try:
263 return int(value)
264 except (TypeError, ValueError):
265 validate(False, f'{name} retry count', value)
267 opts.retries = parse_retries('download', opts.retries)
268 opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
269 opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
270 opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
272 # Retry sleep function
273 def parse_sleep_func(expr):
274 NUMBER_RE = r'\d+(?:\.\d+)?'
275 op, start, limit, step, *_ = (*tuple(re.fullmatch(
276 rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
277 expr.strip()).groups()), None, None)
279 if op == 'exp':
280 return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
281 else:
282 default_step = start if op or limit else 0
283 return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
285 for key, expr in opts.retry_sleep.items():
286 if not expr:
287 del opts.retry_sleep[key]
288 continue
289 try:
290 opts.retry_sleep[key] = parse_sleep_func(expr)
291 except AttributeError:
292 raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
294 # Bytes
295 def validate_bytes(name, value):
296 if value is None:
297 return None
298 numeric_limit = parse_bytes(value)
299 validate(numeric_limit is not None, 'rate limit', value)
300 return numeric_limit
302 opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
303 opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
304 opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
305 opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
306 opts.buffersize = validate_bytes('buffer size', opts.buffersize)
307 opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
309 # Output templates
310 def validate_outtmpl(tmpl, msg):
311 err = YoutubeDL.validate_outtmpl(tmpl)
312 if err:
313 raise ValueError(f'invalid {msg} "{tmpl}": {err}')
315 for k, tmpl in opts.outtmpl.items():
316 validate_outtmpl(tmpl, f'{k} output template')
317 for type_, tmpl_list in opts.forceprint.items():
318 for tmpl in tmpl_list:
319 validate_outtmpl(tmpl, f'{type_} print template')
320 for type_, tmpl_list in opts.print_to_file.items():
321 for tmpl, file in tmpl_list:
322 validate_outtmpl(tmpl, f'{type_} print to file template')
323 validate_outtmpl(file, f'{type_} print to file filename')
324 validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
325 for k, tmpl in opts.progress_template.items():
326 k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
327 validate_outtmpl(tmpl, f'{k} template')
329 outtmpl_default = opts.outtmpl.get('default')
330 if outtmpl_default == '':
331 opts.skip_download = None
332 del opts.outtmpl['default']
334 def parse_chapters(name, value, advanced=False):
335 parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
336 TIMESTAMP_RE = r'''(?x)(?:
337 (?P<start_sign>-?)(?P<start>[^-]+)
338 )?\s*-\s*(?:
339 (?P<end_sign>-?)(?P<end>[^-]+)
340 )?'''
342 chapters, ranges, from_url = [], [], False
343 for regex in value or []:
344 if advanced and regex == '*from-url':
345 from_url = True
346 continue
347 elif not regex.startswith('*'):
348 try:
349 chapters.append(re.compile(regex))
350 except re.error as err:
351 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
352 continue
354 for range_ in map(str.strip, regex[1:].split(',')):
355 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
356 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
357 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
359 err = None
360 if None in (dur or [None]):
361 err = 'Must be of the form "*start-end"'
362 elif not advanced and any(signs):
363 err = 'Negative timestamps are not allowed'
364 else:
365 dur[0] *= -1 if signs[0] else 1
366 dur[1] *= -1 if signs[1] else 1
367 if dur[1] == float('-inf'):
368 err = '"-inf" is not a valid end'
369 if err:
370 raise ValueError(f'invalid {name} time range "{regex}". {err}')
371 ranges.append(dur)
373 return chapters, ranges, from_url
375 opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
376 opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
378 # Cookies from browser
379 if opts.cookiesfrombrowser:
380 container = None
381 mobj = re.fullmatch(r'''(?x)
382 (?P<name>[^+:]+)
383 (?:\s*\+\s*(?P<keyring>[^:]+))?
384 (?:\s*:\s*(?!:)(?P<profile>.+?))?
385 (?:\s*::\s*(?P<container>.+))?
386 ''', opts.cookiesfrombrowser)
387 if mobj is None:
388 raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
389 browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
390 browser_name = browser_name.lower()
391 if browser_name not in SUPPORTED_BROWSERS:
392 raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
393 f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
394 if keyring is not None:
395 keyring = keyring.upper()
396 if keyring not in SUPPORTED_KEYRINGS:
397 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
398 f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
399 opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
401 if opts.impersonate is not None:
402 opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
404 # MetadataParser
405 def metadataparser_actions(f):
406 if isinstance(f, str):
407 cmd = f'--parse-metadata {shell_quote(f)}'
408 try:
409 actions = [MetadataFromFieldPP.to_action(f)]
410 except Exception as err:
411 raise ValueError(f'{cmd} is invalid; {err}')
412 else:
413 cmd = f'--replace-in-metadata {shell_quote(f)}'
414 actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
416 for action in actions:
417 try:
418 MetadataParserPP.validate_action(*action)
419 except Exception as err:
420 raise ValueError(f'{cmd} is invalid; {err}')
421 yield action
423 if opts.metafromtitle is not None:
424 opts.parse_metadata.setdefault('pre_process', []).append(f'title:{opts.metafromtitle}')
425 opts.parse_metadata = {
426 k: list(itertools.chain(*map(metadataparser_actions, v)))
427 for k, v in opts.parse_metadata.items()
430 # Other options
431 if opts.playlist_items is not None:
432 try:
433 tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
434 except Exception as err:
435 raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
437 opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
438 if opts.geo_bypass.lower() not in ('default', 'never'):
439 try:
440 GeoUtils.random_ipv4(opts.geo_bypass)
441 except Exception:
442 raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
443 if len(opts.geo_bypass) == 2:
444 opts.geo_bypass_country = opts.geo_bypass
445 else:
446 opts.geo_bypass_ip_block = opts.geo_bypass
447 opts.geo_bypass = opts.geo_bypass.lower() != 'never'
449 opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
451 if opts.download_archive is not None:
452 opts.download_archive = expand_path(opts.download_archive)
454 if opts.ffmpeg_location is not None:
455 opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
457 if opts.user_agent is not None:
458 opts.headers.setdefault('User-Agent', opts.user_agent)
459 if opts.referer is not None:
460 opts.headers.setdefault('Referer', opts.referer)
462 if opts.no_sponsorblock:
463 opts.sponsorblock_mark = opts.sponsorblock_remove = set()
465 default_downloader = None
466 for proto, path in opts.external_downloader.items():
467 if path == 'native':
468 continue
469 ed = get_external_downloader(path)
470 if ed is None:
471 raise ValueError(
472 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
473 elif ed and proto == 'default':
474 default_downloader = ed.get_basename()
476 for policy in opts.color.values():
477 if policy not in ('always', 'auto', 'auto-tty', 'no_color', 'no_color-tty', 'never'):
478 raise ValueError(f'"{policy}" is not a valid color policy')
480 warnings, deprecation_warnings = [], []
482 # Common mistake: -f best
483 if opts.format == 'best':
484 warnings.append('.\n '.join((
485 '"-f best" selects the best pre-merged format which is often not the best option',
486 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
487 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
489 # --(postprocessor/downloader)-args without name
490 def report_args_compat(name, value, key1, key2=None, where=None):
491 if key1 in value and key2 not in value:
492 warnings.append(f'{name.title()} arguments given without specifying name. '
493 f'The arguments will be given to {where or f"all {name}s"}')
494 return True
495 return False
497 if report_args_compat('external downloader', opts.external_downloader_args,
498 'default', where=default_downloader) and default_downloader:
499 # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
500 opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
502 if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
503 opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
504 opts.postprocessor_args.setdefault('sponskrub', [])
506 def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
507 val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
508 if val2 is NO_DEFAULT:
509 val2 = getattr(opts, opt2)
510 if not val2:
511 return
513 if val1 is NO_DEFAULT:
514 val1 = getattr(opts, opt1)
515 if val1:
516 warnings.append(f'{arg1} is ignored since {arg2} was given')
517 setattr(opts, opt1, default)
519 # Conflicting options
520 report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
521 report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
522 report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
523 report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
524 report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
525 report_conflict('--exec-before-download', 'exec_before_dl_cmd',
526 '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
527 report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
528 report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
529 report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
530 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
531 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
532 report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
533 val1=opts.sponskrub and opts.sponskrub_cut)
535 # Conflicts with --allow-unplayable-formats
536 report_conflict('--embed-metadata', 'addmetadata')
537 report_conflict('--embed-chapters', 'addchapters')
538 report_conflict('--embed-info-json', 'embed_infojson')
539 report_conflict('--embed-subs', 'embedsubtitles')
540 report_conflict('--embed-thumbnail', 'embedthumbnail')
541 report_conflict('--extract-audio', 'extractaudio')
542 report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
543 report_conflict('--recode-video', 'recodevideo')
544 report_conflict('--remove-chapters', 'remove_chapters', default=[])
545 report_conflict('--remux-video', 'remuxvideo')
546 report_conflict('--sponskrub', 'sponskrub')
547 report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
548 report_conflict('--xattrs', 'xattrs')
550 # Fully deprecated options
551 def report_deprecation(val, old, new=None):
552 if not val:
553 return
554 deprecation_warnings.append(
555 f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
556 else f'{old} is deprecated and may not work as expected')
558 report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
559 report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
560 # report_deprecation(opts.include_ads, '--include-ads') # We may re-implement this in future
561 # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
562 # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
564 # Dependent options
565 opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
567 if opts.exec_before_dl_cmd:
568 opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
570 if opts.useid: # --id is not deprecated in youtube-dl
571 opts.outtmpl['default'] = '%(id)s.%(ext)s'
573 if opts.overwrites: # --force-overwrites implies --no-continue
574 opts.continue_dl = False
576 if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
577 # Add chapters when adding metadata or marking sponsors
578 opts.addchapters = True
580 if opts.extractaudio and not opts.keepvideo and opts.format is None:
581 # Do not unnecessarily download audio
582 opts.format = 'bestaudio/best'
584 if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
585 # If JSON is not printed anywhere, but comments are requested, save it to file
586 if not opts.dumpjson or opts.print_json or opts.dump_single_json:
587 opts.writeinfojson = True
589 if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
590 # --all-sub automatically sets --write-sub if --write-auto-sub is not given
591 opts.writesubtitles = True
593 if opts.addmetadata and opts.embed_infojson is None:
594 # If embedding metadata and infojson is present, embed it
595 opts.embed_infojson = 'if_exists'
597 # Ask for passwords
598 if opts.username is not None and opts.password is None:
599 opts.password = getpass.getpass('Type account password and press [Return]: ')
600 if opts.ap_username is not None and opts.ap_password is None:
601 opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
603 # compat option changes global state destructively; only allow from cli
604 if 'allow-unsafe-ext' in opts.compat_opts:
605 warnings.append(
606 'Using allow-unsafe-ext opens you up to potential attacks. '
607 'Use with great care!')
608 _UnsafeExtensionError.sanitize_extension = lambda x, prepend=False: x
610 return warnings, deprecation_warnings
613 def get_postprocessors(opts):
614 yield from opts.add_postprocessors
616 for when, actions in opts.parse_metadata.items():
617 yield {
618 'key': 'MetadataParser',
619 'actions': actions,
620 'when': when,
622 sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
623 if sponsorblock_query:
624 yield {
625 'key': 'SponsorBlock',
626 'categories': sponsorblock_query,
627 'api': opts.sponsorblock_api,
628 'when': 'after_filter',
630 if opts.convertsubtitles:
631 yield {
632 'key': 'FFmpegSubtitlesConvertor',
633 'format': opts.convertsubtitles,
634 'when': 'before_dl',
636 if opts.convertthumbnails:
637 yield {
638 'key': 'FFmpegThumbnailsConvertor',
639 'format': opts.convertthumbnails,
640 'when': 'before_dl',
642 if opts.extractaudio:
643 yield {
644 'key': 'FFmpegExtractAudio',
645 'preferredcodec': opts.audioformat,
646 'preferredquality': opts.audioquality,
647 'nopostoverwrites': opts.nopostoverwrites,
649 if opts.remuxvideo:
650 yield {
651 'key': 'FFmpegVideoRemuxer',
652 'preferedformat': opts.remuxvideo,
654 if opts.recodevideo:
655 yield {
656 'key': 'FFmpegVideoConvertor',
657 'preferedformat': opts.recodevideo,
659 # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
660 if opts.embedsubtitles:
661 keep_subs = 'no-keep-subs' not in opts.compat_opts
662 yield {
663 'key': 'FFmpegEmbedSubtitle',
664 # already_have_subtitle = True prevents the file from being deleted after embedding
665 'already_have_subtitle': opts.writesubtitles and keep_subs,
667 if not opts.writeautomaticsub and keep_subs:
668 opts.writesubtitles = True
670 # ModifyChapters must run before FFmpegMetadataPP
671 if opts.remove_chapters or sponsorblock_query:
672 yield {
673 'key': 'ModifyChapters',
674 'remove_chapters_patterns': opts.remove_chapters,
675 'remove_sponsor_segments': opts.sponsorblock_remove,
676 'remove_ranges': opts.remove_ranges,
677 'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
678 'force_keyframes': opts.force_keyframes_at_cuts,
680 # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
681 # FFmpegExtractAudioPP as containers before conversion may not support
682 # metadata (3gp, webm, etc.)
683 # By default ffmpeg preserves metadata applicable for both
684 # source and target containers. From this point the container won't change,
685 # so metadata can be added here.
686 if opts.addmetadata or opts.addchapters or opts.embed_infojson:
687 yield {
688 'key': 'FFmpegMetadata',
689 'add_chapters': opts.addchapters,
690 'add_metadata': opts.addmetadata,
691 'add_infojson': opts.embed_infojson,
693 # Deprecated
694 # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
695 # but must be below EmbedSubtitle and FFmpegMetadata
696 # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
697 # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
698 if opts.sponskrub is not False:
699 yield {
700 'key': 'SponSkrub',
701 'path': opts.sponskrub_path,
702 'args': opts.sponskrub_args,
703 'cut': opts.sponskrub_cut,
704 'force': opts.sponskrub_force,
705 'ignoreerror': opts.sponskrub is None,
706 '_from_cli': True,
708 if opts.embedthumbnail:
709 yield {
710 'key': 'EmbedThumbnail',
711 # already_have_thumbnail = True prevents the file from being deleted after embedding
712 'already_have_thumbnail': opts.writethumbnail,
714 if not opts.writethumbnail:
715 opts.writethumbnail = True
716 opts.outtmpl['pl_thumbnail'] = ''
717 if opts.split_chapters:
718 yield {
719 'key': 'FFmpegSplitChapters',
720 'force_keyframes': opts.force_keyframes_at_cuts,
722 # XAttrMetadataPP should be run after post-processors that may change file contents
723 if opts.xattrs:
724 yield {'key': 'XAttrMetadata'}
725 if opts.concat_playlist != 'never':
726 yield {
727 'key': 'FFmpegConcat',
728 'only_multi_video': opts.concat_playlist != 'always',
729 'when': 'playlist',
731 # Exec must be the last PP of each category
732 for when, exec_cmd in opts.exec_cmd.items():
733 yield {
734 'key': 'Exec',
735 'exec_cmd': exec_cmd,
736 'when': when,
740 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
743 def parse_options(argv=None):
744 """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
745 parser, opts, urls = parseOpts(argv)
746 urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
748 set_compat_opts(opts)
749 try:
750 warnings, deprecation_warnings = validate_options(opts)
751 except ValueError as err:
752 parser.error(f'{err}\n')
754 postprocessors = list(get_postprocessors(opts))
756 print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
757 any_getting = any(getattr(opts, k) for k in (
758 'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
759 'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl',
761 if opts.quiet is None:
762 opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
764 playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
765 write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
766 and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
767 if not any((
768 opts.extract_flat,
769 opts.dump_single_json,
770 opts.forceprint.get('playlist'),
771 opts.print_to_file.get('playlist'),
772 write_playlist_infojson,
774 if not playlist_pps:
775 opts.extract_flat = 'discard'
776 elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
777 opts.extract_flat = 'discard_in_playlist'
779 final_ext = (
780 opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
781 else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
782 else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
783 else None)
785 return ParsedOptions(parser, opts, urls, {
786 'usenetrc': opts.usenetrc,
787 'netrc_location': opts.netrc_location,
788 'netrc_cmd': opts.netrc_cmd,
789 'username': opts.username,
790 'password': opts.password,
791 'twofactor': opts.twofactor,
792 'videopassword': opts.videopassword,
793 'ap_mso': opts.ap_mso,
794 'ap_username': opts.ap_username,
795 'ap_password': opts.ap_password,
796 'client_certificate': opts.client_certificate,
797 'client_certificate_key': opts.client_certificate_key,
798 'client_certificate_password': opts.client_certificate_password,
799 'quiet': opts.quiet,
800 'no_warnings': opts.no_warnings,
801 'forceurl': opts.geturl,
802 'forcetitle': opts.gettitle,
803 'forceid': opts.getid,
804 'forcethumbnail': opts.getthumbnail,
805 'forcedescription': opts.getdescription,
806 'forceduration': opts.getduration,
807 'forcefilename': opts.getfilename,
808 'forceformat': opts.getformat,
809 'forceprint': opts.forceprint,
810 'print_to_file': opts.print_to_file,
811 'forcejson': opts.dumpjson or opts.print_json,
812 'dump_single_json': opts.dump_single_json,
813 'force_write_download_archive': opts.force_write_download_archive,
814 'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
815 'skip_download': opts.skip_download,
816 'format': opts.format,
817 'allow_unplayable_formats': opts.allow_unplayable_formats,
818 'ignore_no_formats_error': opts.ignore_no_formats_error,
819 'format_sort': opts.format_sort,
820 'format_sort_force': opts.format_sort_force,
821 'allow_multiple_video_streams': opts.allow_multiple_video_streams,
822 'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
823 'check_formats': opts.check_formats,
824 'listformats': opts.listformats,
825 'listformats_table': opts.listformats_table,
826 'outtmpl': opts.outtmpl,
827 'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
828 'paths': opts.paths,
829 'autonumber_size': opts.autonumber_size,
830 'autonumber_start': opts.autonumber_start,
831 'restrictfilenames': opts.restrictfilenames,
832 'windowsfilenames': opts.windowsfilenames,
833 'ignoreerrors': opts.ignoreerrors,
834 'force_generic_extractor': opts.force_generic_extractor,
835 'allowed_extractors': opts.allowed_extractors or ['default'],
836 'ratelimit': opts.ratelimit,
837 'throttledratelimit': opts.throttledratelimit,
838 'overwrites': opts.overwrites,
839 'retries': opts.retries,
840 'file_access_retries': opts.file_access_retries,
841 'fragment_retries': opts.fragment_retries,
842 'extractor_retries': opts.extractor_retries,
843 'retry_sleep_functions': opts.retry_sleep,
844 'skip_unavailable_fragments': opts.skip_unavailable_fragments,
845 'keep_fragments': opts.keep_fragments,
846 'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
847 'buffersize': opts.buffersize,
848 'noresizebuffer': opts.noresizebuffer,
849 'http_chunk_size': opts.http_chunk_size,
850 'continuedl': opts.continue_dl,
851 'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
852 'progress_with_newline': opts.progress_with_newline,
853 'progress_template': opts.progress_template,
854 'progress_delta': opts.progress_delta,
855 'playliststart': opts.playliststart,
856 'playlistend': opts.playlistend,
857 'playlistreverse': opts.playlist_reverse,
858 'playlistrandom': opts.playlist_random,
859 'lazy_playlist': opts.lazy_playlist,
860 'noplaylist': opts.noplaylist,
861 'logtostderr': opts.outtmpl.get('default') == '-',
862 'consoletitle': opts.consoletitle,
863 'nopart': opts.nopart,
864 'updatetime': opts.updatetime,
865 'writedescription': opts.writedescription,
866 'writeannotations': opts.writeannotations,
867 'writeinfojson': opts.writeinfojson,
868 'allow_playlist_files': opts.allow_playlist_files,
869 'clean_infojson': opts.clean_infojson,
870 'getcomments': opts.getcomments,
871 'writethumbnail': opts.writethumbnail is True,
872 'write_all_thumbnails': opts.writethumbnail == 'all',
873 'writelink': opts.writelink,
874 'writeurllink': opts.writeurllink,
875 'writewebloclink': opts.writewebloclink,
876 'writedesktoplink': opts.writedesktoplink,
877 'writesubtitles': opts.writesubtitles,
878 'writeautomaticsub': opts.writeautomaticsub,
879 'allsubtitles': opts.allsubtitles,
880 'listsubtitles': opts.listsubtitles,
881 'subtitlesformat': opts.subtitlesformat,
882 'subtitleslangs': opts.subtitleslangs,
883 'matchtitle': decodeOption(opts.matchtitle),
884 'rejecttitle': decodeOption(opts.rejecttitle),
885 'max_downloads': opts.max_downloads,
886 'prefer_free_formats': opts.prefer_free_formats,
887 'trim_file_name': opts.trim_file_name,
888 'verbose': opts.verbose,
889 'dump_intermediate_pages': opts.dump_intermediate_pages,
890 'write_pages': opts.write_pages,
891 'load_pages': opts.load_pages,
892 'test': opts.test,
893 'keepvideo': opts.keepvideo,
894 'min_filesize': opts.min_filesize,
895 'max_filesize': opts.max_filesize,
896 'min_views': opts.min_views,
897 'max_views': opts.max_views,
898 'daterange': opts.date,
899 'cachedir': opts.cachedir,
900 'youtube_print_sig_code': opts.youtube_print_sig_code,
901 'age_limit': opts.age_limit,
902 'download_archive': opts.download_archive,
903 'break_on_existing': opts.break_on_existing,
904 'break_on_reject': opts.break_on_reject,
905 'break_per_url': opts.break_per_url,
906 'skip_playlist_after_errors': opts.skip_playlist_after_errors,
907 'cookiefile': opts.cookiefile,
908 'cookiesfrombrowser': opts.cookiesfrombrowser,
909 'legacyserverconnect': opts.legacy_server_connect,
910 'nocheckcertificate': opts.no_check_certificate,
911 'prefer_insecure': opts.prefer_insecure,
912 'enable_file_urls': opts.enable_file_urls,
913 'http_headers': opts.headers,
914 'proxy': opts.proxy,
915 'socket_timeout': opts.socket_timeout,
916 'bidi_workaround': opts.bidi_workaround,
917 'debug_printtraffic': opts.debug_printtraffic,
918 'prefer_ffmpeg': opts.prefer_ffmpeg,
919 'include_ads': opts.include_ads,
920 'default_search': opts.default_search,
921 'dynamic_mpd': opts.dynamic_mpd,
922 'extractor_args': opts.extractor_args,
923 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
924 'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
925 'encoding': opts.encoding,
926 'extract_flat': opts.extract_flat,
927 'live_from_start': opts.live_from_start,
928 'wait_for_video': opts.wait_for_video,
929 'mark_watched': opts.mark_watched,
930 'merge_output_format': opts.merge_output_format,
931 'final_ext': final_ext,
932 'postprocessors': postprocessors,
933 'fixup': opts.fixup,
934 'source_address': opts.source_address,
935 'impersonate': opts.impersonate,
936 'call_home': opts.call_home,
937 'sleep_interval_requests': opts.sleep_interval_requests,
938 'sleep_interval': opts.sleep_interval,
939 'max_sleep_interval': opts.max_sleep_interval,
940 'sleep_interval_subtitles': opts.sleep_interval_subtitles,
941 'external_downloader': opts.external_downloader,
942 'download_ranges': opts.download_ranges,
943 'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
944 'list_thumbnails': opts.list_thumbnails,
945 'playlist_items': opts.playlist_items,
946 'xattr_set_filesize': opts.xattr_set_filesize,
947 'match_filter': opts.match_filter,
948 'color': opts.color,
949 'ffmpeg_location': opts.ffmpeg_location,
950 'hls_prefer_native': opts.hls_prefer_native,
951 'hls_use_mpegts': opts.hls_use_mpegts,
952 'hls_split_discontinuity': opts.hls_split_discontinuity,
953 'external_downloader_args': opts.external_downloader_args,
954 'postprocessor_args': opts.postprocessor_args,
955 'cn_verification_proxy': opts.cn_verification_proxy,
956 'geo_verification_proxy': opts.geo_verification_proxy,
957 'geo_bypass': opts.geo_bypass,
958 'geo_bypass_country': opts.geo_bypass_country,
959 'geo_bypass_ip_block': opts.geo_bypass_ip_block,
960 '_warnings': warnings,
961 '_deprecation_warnings': deprecation_warnings,
962 'compat_opts': opts.compat_opts,
966 def _real_main(argv=None):
967 setproctitle('yt-dlp')
969 parser, opts, all_urls, ydl_opts = parse_options(argv)
971 # HACK: Set the plugin dirs early on
972 # TODO(coletdjnz): remove when plugin globals system is implemented
973 Config._plugin_dirs = opts.plugin_dirs
975 # Dump user agent
976 if opts.dump_user_agent:
977 ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
978 write_string(f'{ua}\n', out=sys.stdout)
979 return
981 if print_extractor_information(opts, all_urls):
982 return
984 # We may need ffmpeg_location without having access to the YoutubeDL instance
985 # See https://github.com/yt-dlp/yt-dlp/issues/2191
986 if opts.ffmpeg_location:
987 FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
989 with YoutubeDL(ydl_opts) as ydl:
990 pre_process = opts.update_self or opts.rm_cachedir
991 actual_use = all_urls or opts.load_info_filename
993 if opts.rm_cachedir:
994 ydl.cache.remove()
996 try:
997 updater = Updater(ydl, opts.update_self)
998 if opts.update_self and updater.update() and actual_use:
999 if updater.cmd:
1000 return updater.restart()
1001 # This code is reachable only for zip variant in py < 3.10
1002 # It makes sense to exit here, but the old behavior is to continue
1003 ydl.report_warning('Restart yt-dlp to use the updated version')
1004 # return 100, 'ERROR: The program must exit for the update to complete'
1005 except Exception:
1006 traceback.print_exc()
1007 ydl._download_retcode = 100
1009 if opts.list_impersonate_targets:
1011 known_targets = [
1012 # List of simplified targets we know are supported,
1013 # to help users know what dependencies may be required.
1014 (ImpersonateTarget('chrome'), 'curl_cffi'),
1015 (ImpersonateTarget('edge'), 'curl_cffi'),
1016 (ImpersonateTarget('safari'), 'curl_cffi'),
1019 available_targets = ydl._get_available_impersonate_targets()
1021 def make_row(target, handler):
1022 return [
1023 join_nonempty(target.client.title(), target.version, delim='-') or '-',
1024 join_nonempty((target.os or '').title(), target.os_version, delim='-') or '-',
1025 handler,
1028 rows = [make_row(target, handler) for target, handler in available_targets]
1030 for known_target, known_handler in known_targets:
1031 if not any(
1032 known_target in target and handler == known_handler
1033 for target, handler in available_targets
1035 rows.append([
1036 ydl._format_out(text, ydl.Styles.SUPPRESS)
1037 for text in make_row(known_target, f'{known_handler} (not available)')
1040 ydl.to_screen('[info] Available impersonate targets')
1041 ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
1042 return
1044 if not actual_use:
1045 if pre_process:
1046 return ydl._download_retcode
1048 args = sys.argv[1:] if argv is None else argv
1049 ydl.warn_if_short_id(args)
1051 # Show a useful error message and wait for keypress if not launched from shell on Windows
1052 if not args and compat_os_name == 'nt' and getattr(sys, 'frozen', False):
1053 import ctypes.wintypes
1054 import msvcrt
1056 kernel32 = ctypes.WinDLL('Kernel32')
1058 buffer = (1 * ctypes.wintypes.DWORD)()
1059 attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
1060 # If we only have a single process attached, then the executable was double clicked
1061 # When using `pyinstaller` with `--onefile`, two processes get attached
1062 is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
1063 if attached_processes == 1 or is_onefile and attached_processes == 2:
1064 print(parser._generate_error_message(
1065 'Do not double-click the executable, instead call it from a command line.\n'
1066 'Please read the README for further information on how to use yt-dlp: '
1067 'https://github.com/yt-dlp/yt-dlp#readme'))
1068 msvcrt.getch()
1069 _exit(2)
1070 parser.error(
1071 'You must provide at least one URL.\n'
1072 'Type yt-dlp --help to see a list of all options.')
1074 parser.destroy()
1075 try:
1076 if opts.load_info_filename is not None:
1077 if all_urls:
1078 ydl.report_warning('URLs are ignored due to --load-info-json')
1079 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
1080 else:
1081 return ydl.download(all_urls)
1082 except DownloadCancelled:
1083 ydl.to_screen('Aborting remaining downloads')
1084 return 101
1087 def main(argv=None):
1088 global _IN_CLI
1089 _IN_CLI = True
1090 try:
1091 _exit(*variadic(_real_main(argv)))
1092 except (CookieLoadError, DownloadError):
1093 _exit(1)
1094 except SameFileError as e:
1095 _exit(f'ERROR: {e}')
1096 except KeyboardInterrupt:
1097 _exit('\nERROR: Interrupted by user')
1098 except BrokenPipeError as e:
1099 # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1100 devnull = os.open(os.devnull, os.O_WRONLY)
1101 os.dup2(devnull, sys.stdout.fileno())
1102 _exit(f'\nERROR: {e}')
1103 except optparse.OptParseError as e:
1104 _exit(2, f'\n{e}')
1107 from .extractor import gen_extractors, list_extractors
1109 __all__ = [
1110 'main',
1111 'YoutubeDL',
1112 'parse_options',
1113 'gen_extractors',
1114 'list_extractors',