]> jfr.im git - yt-dlp.git/blob - yt_dlp/__init__.py
[ie/eplus] Handle URLs without videos (#9855)
[yt-dlp.git] / yt_dlp / __init__.py
1 import sys
2
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
6
7 __license__ = 'The Unlicense'
8
9 import collections
10 import getpass
11 import itertools
12 import optparse
13 import os
14 import re
15 import traceback
16
17 from .compat import compat_os_name, compat_shlex_quote
18 from .cookies import SUPPORTED_BROWSERS, SUPPORTED_KEYRINGS
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,
34 )
35 from .update import Updater
36 from .utils import (
37 NO_DEFAULT,
38 POSTPROCESS_WHEN,
39 DateRange,
40 DownloadCancelled,
41 DownloadError,
42 FormatSorter,
43 GeoUtils,
44 PlaylistEntries,
45 SameFileError,
46 decodeOption,
47 download_range_func,
48 expand_path,
49 float_or_none,
50 format_field,
51 int_or_none,
52 join_nonempty,
53 match_filter_func,
54 parse_bytes,
55 parse_duration,
56 preferredencoding,
57 read_batch_urls,
58 read_stdin,
59 render_table,
60 setproctitle,
61 traverse_obj,
62 variadic,
63 write_string,
64 )
65 from .utils.networking import std_headers
66 from .YoutubeDL import YoutubeDL
67
68 _IN_CLI = False
69
70
71 def _exit(status=0, *args):
72 for msg in args:
73 sys.stderr.write(msg)
74 raise SystemExit(status)
75
76
77 def get_urls(urls, batchfile, verbose):
78 """
79 @param verbose -1: quiet, 0: normal, 1: verbose
80 """
81 batch_urls = []
82 if batchfile is not None:
83 try:
84 batch_urls = read_batch_urls(
85 read_stdin(None if verbose == -1 else 'URLs') if batchfile == '-'
86 else open(expand_path(batchfile), encoding='utf-8', errors='ignore'))
87 if verbose == 1:
88 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
89 except OSError:
90 _exit(f'ERROR: batch file {batchfile} could not be read')
91 _enc = preferredencoding()
92 return [
93 url.strip().decode(_enc, 'ignore') if isinstance(url, bytes) else url.strip()
94 for url in batch_urls + urls]
95
96
97 def print_extractor_information(opts, urls):
98 out = ''
99 if opts.list_extractors:
100 # Importing GenericIE is currently slow since it imports YoutubeIE
101 from .extractor.generic import GenericIE
102
103 urls = dict.fromkeys(urls, False)
104 for ie in list_extractor_classes(opts.age_limit):
105 out += ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie.working() else '') + '\n'
106 if ie == GenericIE:
107 matched_urls = [url for url, matched in urls.items() if not matched]
108 else:
109 matched_urls = tuple(filter(ie.suitable, urls.keys()))
110 urls.update(dict.fromkeys(matched_urls, True))
111 out += ''.join(f' {url}\n' for url in matched_urls)
112 elif opts.list_extractor_descriptions:
113 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
114 out = '\n'.join(
115 ie.description(markdown=False, search_examples=_SEARCHES)
116 for ie in list_extractor_classes(opts.age_limit) if ie.working() and ie.IE_DESC is not False)
117 elif opts.ap_list_mso:
118 out = 'Supported TV Providers:\n%s\n' % render_table(
119 ['mso', 'mso name'],
120 [[mso_id, mso_info['name']] for mso_id, mso_info in MSO_INFO.items()])
121 else:
122 return False
123 write_string(out, out=sys.stdout)
124 return True
125
126
127 def set_compat_opts(opts):
128 def _unused_compat_opt(name):
129 if name not in opts.compat_opts:
130 return False
131 opts.compat_opts.discard(name)
132 opts.compat_opts.update(['*%s' % name])
133 return True
134
135 def set_default_compat(compat_name, opt_name, default=True, remove_compat=True):
136 attr = getattr(opts, opt_name)
137 if compat_name in opts.compat_opts:
138 if attr is None:
139 setattr(opts, opt_name, not default)
140 return True
141 else:
142 if remove_compat:
143 _unused_compat_opt(compat_name)
144 return False
145 elif attr is None:
146 setattr(opts, opt_name, default)
147 return None
148
149 set_default_compat('abort-on-error', 'ignoreerrors', 'only_download')
150 set_default_compat('no-playlist-metafiles', 'allow_playlist_files')
151 set_default_compat('no-clean-infojson', 'clean_infojson')
152 if 'no-attach-info-json' in opts.compat_opts:
153 if opts.embed_infojson:
154 _unused_compat_opt('no-attach-info-json')
155 else:
156 opts.embed_infojson = False
157 if 'format-sort' in opts.compat_opts:
158 opts.format_sort.extend(FormatSorter.ytdl_default)
159 _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
160 _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
161 if _video_multistreams_set is False and _audio_multistreams_set is False:
162 _unused_compat_opt('multistreams')
163 if 'filename' in opts.compat_opts:
164 if opts.outtmpl.get('default') is None:
165 opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
166 else:
167 _unused_compat_opt('filename')
168
169
170 def validate_options(opts):
171 def validate(cndn, name, value=None, msg=None):
172 if cndn:
173 return True
174 raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
175
176 def validate_in(name, value, items, msg=None):
177 return validate(value is None or value in items, name, value, msg)
178
179 def validate_regex(name, value, regex):
180 return validate(value is None or re.match(regex, value), name, value)
181
182 def validate_positive(name, value, strict=False):
183 return validate(value is None or value > 0 or (not strict and value == 0),
184 name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
185
186 def validate_minmax(min_val, max_val, min_name, max_name=None):
187 if max_val is None or min_val is None or max_val >= min_val:
188 return
189 if not max_name:
190 min_name, max_name = f'min {min_name}', f'max {min_name}'
191 raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
192
193 # Usernames and passwords
194 validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
195 msg='{name}, netrc command and username/password are mutually exclusive options')
196 validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
197 validate(opts.ap_password is None or opts.ap_username is not None,
198 'TV Provider account username', msg='{name} missing')
199 validate_in('TV Provider', opts.ap_mso, MSO_INFO,
200 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
201
202 # Numbers
203 validate_positive('autonumber start', opts.autonumber_start)
204 validate_positive('autonumber size', opts.autonumber_size, True)
205 validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
206 validate_positive('playlist start', opts.playliststart, True)
207 if opts.playlistend != -1:
208 validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
209
210 # Time ranges
211 validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
212 validate_positive('requests sleep interval', opts.sleep_interval_requests)
213 validate_positive('sleep interval', opts.sleep_interval)
214 validate_positive('max sleep interval', opts.max_sleep_interval)
215 if opts.sleep_interval is None:
216 validate(
217 opts.max_sleep_interval is None, 'min sleep interval',
218 msg='{name} must be specified; use --min-sleep-interval')
219 elif opts.max_sleep_interval is None:
220 opts.max_sleep_interval = opts.sleep_interval
221 else:
222 validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
223
224 if opts.wait_for_video is not None:
225 min_wait, max_wait, *_ = map(parse_duration, opts.wait_for_video.split('-', 1) + [None])
226 validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
227 'time range to wait for video', opts.wait_for_video)
228 validate_minmax(min_wait, max_wait, 'time range to wait for video')
229 opts.wait_for_video = (min_wait, max_wait)
230
231 # Format sort
232 for f in opts.format_sort:
233 validate_regex('format sorting', f, FormatSorter.regex)
234
235 # Postprocessor formats
236 validate_regex('merge output format', opts.merge_output_format,
237 r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
238 validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
239 validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
240 validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
241 validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
242 validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
243 if opts.audioquality:
244 opts.audioquality = opts.audioquality.strip('k').strip('K')
245 # int_or_none prevents inf, nan
246 validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
247
248 # Retries
249 def parse_retries(name, value):
250 if value is None:
251 return None
252 elif value in ('inf', 'infinite'):
253 return float('inf')
254 try:
255 return int(value)
256 except (TypeError, ValueError):
257 validate(False, f'{name} retry count', value)
258
259 opts.retries = parse_retries('download', opts.retries)
260 opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
261 opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
262 opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
263
264 # Retry sleep function
265 def parse_sleep_func(expr):
266 NUMBER_RE = r'\d+(?:\.\d+)?'
267 op, start, limit, step, *_ = tuple(re.fullmatch(
268 rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
269 expr.strip()).groups()) + (None, None)
270
271 if op == 'exp':
272 return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
273 else:
274 default_step = start if op or limit else 0
275 return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
276
277 for key, expr in opts.retry_sleep.items():
278 if not expr:
279 del opts.retry_sleep[key]
280 continue
281 try:
282 opts.retry_sleep[key] = parse_sleep_func(expr)
283 except AttributeError:
284 raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
285
286 # Bytes
287 def validate_bytes(name, value):
288 if value is None:
289 return None
290 numeric_limit = parse_bytes(value)
291 validate(numeric_limit is not None, 'rate limit', value)
292 return numeric_limit
293
294 opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
295 opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
296 opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
297 opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
298 opts.buffersize = validate_bytes('buffer size', opts.buffersize)
299 opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
300
301 # Output templates
302 def validate_outtmpl(tmpl, msg):
303 err = YoutubeDL.validate_outtmpl(tmpl)
304 if err:
305 raise ValueError(f'invalid {msg} "{tmpl}": {err}')
306
307 for k, tmpl in opts.outtmpl.items():
308 validate_outtmpl(tmpl, f'{k} output template')
309 for type_, tmpl_list in opts.forceprint.items():
310 for tmpl in tmpl_list:
311 validate_outtmpl(tmpl, f'{type_} print template')
312 for type_, tmpl_list in opts.print_to_file.items():
313 for tmpl, file in tmpl_list:
314 validate_outtmpl(tmpl, f'{type_} print to file template')
315 validate_outtmpl(file, f'{type_} print to file filename')
316 validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
317 for k, tmpl in opts.progress_template.items():
318 k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
319 validate_outtmpl(tmpl, f'{k} template')
320
321 outtmpl_default = opts.outtmpl.get('default')
322 if outtmpl_default == '':
323 opts.skip_download = None
324 del opts.outtmpl['default']
325
326 def parse_chapters(name, value, advanced=False):
327 parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
328 TIMESTAMP_RE = r'''(?x)(?:
329 (?P<start_sign>-?)(?P<start>[^-]+)
330 )?\s*-\s*(?:
331 (?P<end_sign>-?)(?P<end>[^-]+)
332 )?'''
333
334 chapters, ranges, from_url = [], [], False
335 for regex in value or []:
336 if advanced and regex == '*from-url':
337 from_url = True
338 continue
339 elif not regex.startswith('*'):
340 try:
341 chapters.append(re.compile(regex))
342 except re.error as err:
343 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
344 continue
345
346 for range_ in map(str.strip, regex[1:].split(',')):
347 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
348 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
349 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
350
351 err = None
352 if None in (dur or [None]):
353 err = 'Must be of the form "*start-end"'
354 elif not advanced and any(signs):
355 err = 'Negative timestamps are not allowed'
356 else:
357 dur[0] *= -1 if signs[0] else 1
358 dur[1] *= -1 if signs[1] else 1
359 if dur[1] == float('-inf'):
360 err = '"-inf" is not a valid end'
361 if err:
362 raise ValueError(f'invalid {name} time range "{regex}". {err}')
363 ranges.append(dur)
364
365 return chapters, ranges, from_url
366
367 opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
368 opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
369
370 # Cookies from browser
371 if opts.cookiesfrombrowser:
372 container = None
373 mobj = re.fullmatch(r'''(?x)
374 (?P<name>[^+:]+)
375 (?:\s*\+\s*(?P<keyring>[^:]+))?
376 (?:\s*:\s*(?!:)(?P<profile>.+?))?
377 (?:\s*::\s*(?P<container>.+))?
378 ''', opts.cookiesfrombrowser)
379 if mobj is None:
380 raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
381 browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
382 browser_name = browser_name.lower()
383 if browser_name not in SUPPORTED_BROWSERS:
384 raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
385 f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
386 if keyring is not None:
387 keyring = keyring.upper()
388 if keyring not in SUPPORTED_KEYRINGS:
389 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
390 f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
391 opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
392
393 if opts.impersonate is not None:
394 opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
395
396 # MetadataParser
397 def metadataparser_actions(f):
398 if isinstance(f, str):
399 cmd = '--parse-metadata %s' % compat_shlex_quote(f)
400 try:
401 actions = [MetadataFromFieldPP.to_action(f)]
402 except Exception as err:
403 raise ValueError(f'{cmd} is invalid; {err}')
404 else:
405 cmd = '--replace-in-metadata %s' % ' '.join(map(compat_shlex_quote, f))
406 actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
407
408 for action in actions:
409 try:
410 MetadataParserPP.validate_action(*action)
411 except Exception as err:
412 raise ValueError(f'{cmd} is invalid; {err}')
413 yield action
414
415 if opts.metafromtitle is not None:
416 opts.parse_metadata.setdefault('pre_process', []).append('title:%s' % opts.metafromtitle)
417 opts.parse_metadata = {
418 k: list(itertools.chain(*map(metadataparser_actions, v)))
419 for k, v in opts.parse_metadata.items()
420 }
421
422 # Other options
423 if opts.playlist_items is not None:
424 try:
425 tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
426 except Exception as err:
427 raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
428
429 opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
430 if opts.geo_bypass.lower() not in ('default', 'never'):
431 try:
432 GeoUtils.random_ipv4(opts.geo_bypass)
433 except Exception:
434 raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
435 if len(opts.geo_bypass) == 2:
436 opts.geo_bypass_country = opts.geo_bypass
437 else:
438 opts.geo_bypass_ip_block = opts.geo_bypass
439 opts.geo_bypass = opts.geo_bypass.lower() != 'never'
440
441 opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
442
443 if opts.download_archive is not None:
444 opts.download_archive = expand_path(opts.download_archive)
445
446 if opts.ffmpeg_location is not None:
447 opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
448
449 if opts.user_agent is not None:
450 opts.headers.setdefault('User-Agent', opts.user_agent)
451 if opts.referer is not None:
452 opts.headers.setdefault('Referer', opts.referer)
453
454 if opts.no_sponsorblock:
455 opts.sponsorblock_mark = opts.sponsorblock_remove = set()
456
457 default_downloader = None
458 for proto, path in opts.external_downloader.items():
459 if path == 'native':
460 continue
461 ed = get_external_downloader(path)
462 if ed is None:
463 raise ValueError(
464 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
465 elif ed and proto == 'default':
466 default_downloader = ed.get_basename()
467
468 for policy in opts.color.values():
469 if policy not in ('always', 'auto', 'no_color', 'never'):
470 raise ValueError(f'"{policy}" is not a valid color policy')
471
472 warnings, deprecation_warnings = [], []
473
474 # Common mistake: -f best
475 if opts.format == 'best':
476 warnings.append('.\n '.join((
477 '"-f best" selects the best pre-merged format which is often not the best option',
478 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
479 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
480
481 # --(postprocessor/downloader)-args without name
482 def report_args_compat(name, value, key1, key2=None, where=None):
483 if key1 in value and key2 not in value:
484 warnings.append(f'{name.title()} arguments given without specifying name. '
485 f'The arguments will be given to {where or f"all {name}s"}')
486 return True
487 return False
488
489 if report_args_compat('external downloader', opts.external_downloader_args,
490 'default', where=default_downloader) and default_downloader:
491 # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
492 opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
493
494 if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
495 opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
496 opts.postprocessor_args.setdefault('sponskrub', [])
497
498 def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
499 val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
500 if val2 is NO_DEFAULT:
501 val2 = getattr(opts, opt2)
502 if not val2:
503 return
504
505 if val1 is NO_DEFAULT:
506 val1 = getattr(opts, opt1)
507 if val1:
508 warnings.append(f'{arg1} is ignored since {arg2} was given')
509 setattr(opts, opt1, default)
510
511 # Conflicting options
512 report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
513 report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
514 report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
515 report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
516 report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
517 report_conflict('--exec-before-download', 'exec_before_dl_cmd',
518 '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
519 report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
520 report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
521 report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
522 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
523 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
524 report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
525 val1=opts.sponskrub and opts.sponskrub_cut)
526
527 # Conflicts with --allow-unplayable-formats
528 report_conflict('--embed-metadata', 'addmetadata')
529 report_conflict('--embed-chapters', 'addchapters')
530 report_conflict('--embed-info-json', 'embed_infojson')
531 report_conflict('--embed-subs', 'embedsubtitles')
532 report_conflict('--embed-thumbnail', 'embedthumbnail')
533 report_conflict('--extract-audio', 'extractaudio')
534 report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
535 report_conflict('--recode-video', 'recodevideo')
536 report_conflict('--remove-chapters', 'remove_chapters', default=[])
537 report_conflict('--remux-video', 'remuxvideo')
538 report_conflict('--sponskrub', 'sponskrub')
539 report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
540 report_conflict('--xattrs', 'xattrs')
541
542 # Fully deprecated options
543 def report_deprecation(val, old, new=None):
544 if not val:
545 return
546 deprecation_warnings.append(
547 f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
548 else f'{old} is deprecated and may not work as expected')
549
550 report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
551 report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
552 # report_deprecation(opts.include_ads, '--include-ads') # We may re-implement this in future
553 # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
554 # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
555
556 # Dependent options
557 opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
558
559 if opts.exec_before_dl_cmd:
560 opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
561
562 if opts.useid: # --id is not deprecated in youtube-dl
563 opts.outtmpl['default'] = '%(id)s.%(ext)s'
564
565 if opts.overwrites: # --force-overwrites implies --no-continue
566 opts.continue_dl = False
567
568 if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
569 # Add chapters when adding metadata or marking sponsors
570 opts.addchapters = True
571
572 if opts.extractaudio and not opts.keepvideo and opts.format is None:
573 # Do not unnecessarily download audio
574 opts.format = 'bestaudio/best'
575
576 if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
577 # If JSON is not printed anywhere, but comments are requested, save it to file
578 if not opts.dumpjson or opts.print_json or opts.dump_single_json:
579 opts.writeinfojson = True
580
581 if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
582 # --all-sub automatically sets --write-sub if --write-auto-sub is not given
583 opts.writesubtitles = True
584
585 if opts.addmetadata and opts.embed_infojson is None:
586 # If embedding metadata and infojson is present, embed it
587 opts.embed_infojson = 'if_exists'
588
589 # Ask for passwords
590 if opts.username is not None and opts.password is None:
591 opts.password = getpass.getpass('Type account password and press [Return]: ')
592 if opts.ap_username is not None and opts.ap_password is None:
593 opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
594
595 return warnings, deprecation_warnings
596
597
598 def get_postprocessors(opts):
599 yield from opts.add_postprocessors
600
601 for when, actions in opts.parse_metadata.items():
602 yield {
603 'key': 'MetadataParser',
604 'actions': actions,
605 'when': when
606 }
607 sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
608 if sponsorblock_query:
609 yield {
610 'key': 'SponsorBlock',
611 'categories': sponsorblock_query,
612 'api': opts.sponsorblock_api,
613 'when': 'after_filter'
614 }
615 if opts.convertsubtitles:
616 yield {
617 'key': 'FFmpegSubtitlesConvertor',
618 'format': opts.convertsubtitles,
619 'when': 'before_dl'
620 }
621 if opts.convertthumbnails:
622 yield {
623 'key': 'FFmpegThumbnailsConvertor',
624 'format': opts.convertthumbnails,
625 'when': 'before_dl'
626 }
627 if opts.extractaudio:
628 yield {
629 'key': 'FFmpegExtractAudio',
630 'preferredcodec': opts.audioformat,
631 'preferredquality': opts.audioquality,
632 'nopostoverwrites': opts.nopostoverwrites,
633 }
634 if opts.remuxvideo:
635 yield {
636 'key': 'FFmpegVideoRemuxer',
637 'preferedformat': opts.remuxvideo,
638 }
639 if opts.recodevideo:
640 yield {
641 'key': 'FFmpegVideoConvertor',
642 'preferedformat': opts.recodevideo,
643 }
644 # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
645 if opts.embedsubtitles:
646 keep_subs = 'no-keep-subs' not in opts.compat_opts
647 yield {
648 'key': 'FFmpegEmbedSubtitle',
649 # already_have_subtitle = True prevents the file from being deleted after embedding
650 'already_have_subtitle': opts.writesubtitles and keep_subs
651 }
652 if not opts.writeautomaticsub and keep_subs:
653 opts.writesubtitles = True
654
655 # ModifyChapters must run before FFmpegMetadataPP
656 if opts.remove_chapters or sponsorblock_query:
657 yield {
658 'key': 'ModifyChapters',
659 'remove_chapters_patterns': opts.remove_chapters,
660 'remove_sponsor_segments': opts.sponsorblock_remove,
661 'remove_ranges': opts.remove_ranges,
662 'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
663 'force_keyframes': opts.force_keyframes_at_cuts
664 }
665 # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
666 # FFmpegExtractAudioPP as containers before conversion may not support
667 # metadata (3gp, webm, etc.)
668 # By default ffmpeg preserves metadata applicable for both
669 # source and target containers. From this point the container won't change,
670 # so metadata can be added here.
671 if opts.addmetadata or opts.addchapters or opts.embed_infojson:
672 yield {
673 'key': 'FFmpegMetadata',
674 'add_chapters': opts.addchapters,
675 'add_metadata': opts.addmetadata,
676 'add_infojson': opts.embed_infojson,
677 }
678 # Deprecated
679 # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
680 # but must be below EmbedSubtitle and FFmpegMetadata
681 # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
682 # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
683 if opts.sponskrub is not False:
684 yield {
685 'key': 'SponSkrub',
686 'path': opts.sponskrub_path,
687 'args': opts.sponskrub_args,
688 'cut': opts.sponskrub_cut,
689 'force': opts.sponskrub_force,
690 'ignoreerror': opts.sponskrub is None,
691 '_from_cli': True,
692 }
693 if opts.embedthumbnail:
694 yield {
695 'key': 'EmbedThumbnail',
696 # already_have_thumbnail = True prevents the file from being deleted after embedding
697 'already_have_thumbnail': opts.writethumbnail
698 }
699 if not opts.writethumbnail:
700 opts.writethumbnail = True
701 opts.outtmpl['pl_thumbnail'] = ''
702 if opts.split_chapters:
703 yield {
704 'key': 'FFmpegSplitChapters',
705 'force_keyframes': opts.force_keyframes_at_cuts,
706 }
707 # XAttrMetadataPP should be run after post-processors that may change file contents
708 if opts.xattrs:
709 yield {'key': 'XAttrMetadata'}
710 if opts.concat_playlist != 'never':
711 yield {
712 'key': 'FFmpegConcat',
713 'only_multi_video': opts.concat_playlist != 'always',
714 'when': 'playlist',
715 }
716 # Exec must be the last PP of each category
717 for when, exec_cmd in opts.exec_cmd.items():
718 yield {
719 'key': 'Exec',
720 'exec_cmd': exec_cmd,
721 'when': when,
722 }
723
724
725 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
726
727
728 def parse_options(argv=None):
729 """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
730 parser, opts, urls = parseOpts(argv)
731 urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
732
733 set_compat_opts(opts)
734 try:
735 warnings, deprecation_warnings = validate_options(opts)
736 except ValueError as err:
737 parser.error(f'{err}\n')
738
739 postprocessors = list(get_postprocessors(opts))
740
741 print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
742 any_getting = any(getattr(opts, k) for k in (
743 'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
744 'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl'
745 ))
746 if opts.quiet is None:
747 opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
748
749 playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
750 write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
751 and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
752 if not any((
753 opts.extract_flat,
754 opts.dump_single_json,
755 opts.forceprint.get('playlist'),
756 opts.print_to_file.get('playlist'),
757 write_playlist_infojson,
758 )):
759 if not playlist_pps:
760 opts.extract_flat = 'discard'
761 elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
762 opts.extract_flat = 'discard_in_playlist'
763
764 final_ext = (
765 opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
766 else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
767 else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
768 else None)
769
770 return ParsedOptions(parser, opts, urls, {
771 'usenetrc': opts.usenetrc,
772 'netrc_location': opts.netrc_location,
773 'netrc_cmd': opts.netrc_cmd,
774 'username': opts.username,
775 'password': opts.password,
776 'twofactor': opts.twofactor,
777 'videopassword': opts.videopassword,
778 'ap_mso': opts.ap_mso,
779 'ap_username': opts.ap_username,
780 'ap_password': opts.ap_password,
781 'client_certificate': opts.client_certificate,
782 'client_certificate_key': opts.client_certificate_key,
783 'client_certificate_password': opts.client_certificate_password,
784 'quiet': opts.quiet,
785 'no_warnings': opts.no_warnings,
786 'forceurl': opts.geturl,
787 'forcetitle': opts.gettitle,
788 'forceid': opts.getid,
789 'forcethumbnail': opts.getthumbnail,
790 'forcedescription': opts.getdescription,
791 'forceduration': opts.getduration,
792 'forcefilename': opts.getfilename,
793 'forceformat': opts.getformat,
794 'forceprint': opts.forceprint,
795 'print_to_file': opts.print_to_file,
796 'forcejson': opts.dumpjson or opts.print_json,
797 'dump_single_json': opts.dump_single_json,
798 'force_write_download_archive': opts.force_write_download_archive,
799 'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
800 'skip_download': opts.skip_download,
801 'format': opts.format,
802 'allow_unplayable_formats': opts.allow_unplayable_formats,
803 'ignore_no_formats_error': opts.ignore_no_formats_error,
804 'format_sort': opts.format_sort,
805 'format_sort_force': opts.format_sort_force,
806 'allow_multiple_video_streams': opts.allow_multiple_video_streams,
807 'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
808 'check_formats': opts.check_formats,
809 'listformats': opts.listformats,
810 'listformats_table': opts.listformats_table,
811 'outtmpl': opts.outtmpl,
812 'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
813 'paths': opts.paths,
814 'autonumber_size': opts.autonumber_size,
815 'autonumber_start': opts.autonumber_start,
816 'restrictfilenames': opts.restrictfilenames,
817 'windowsfilenames': opts.windowsfilenames,
818 'ignoreerrors': opts.ignoreerrors,
819 'force_generic_extractor': opts.force_generic_extractor,
820 'allowed_extractors': opts.allowed_extractors or ['default'],
821 'ratelimit': opts.ratelimit,
822 'throttledratelimit': opts.throttledratelimit,
823 'overwrites': opts.overwrites,
824 'retries': opts.retries,
825 'file_access_retries': opts.file_access_retries,
826 'fragment_retries': opts.fragment_retries,
827 'extractor_retries': opts.extractor_retries,
828 'retry_sleep_functions': opts.retry_sleep,
829 'skip_unavailable_fragments': opts.skip_unavailable_fragments,
830 'keep_fragments': opts.keep_fragments,
831 'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
832 'buffersize': opts.buffersize,
833 'noresizebuffer': opts.noresizebuffer,
834 'http_chunk_size': opts.http_chunk_size,
835 'continuedl': opts.continue_dl,
836 'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
837 'progress_with_newline': opts.progress_with_newline,
838 'progress_template': opts.progress_template,
839 'progress_delta': opts.progress_delta,
840 'playliststart': opts.playliststart,
841 'playlistend': opts.playlistend,
842 'playlistreverse': opts.playlist_reverse,
843 'playlistrandom': opts.playlist_random,
844 'lazy_playlist': opts.lazy_playlist,
845 'noplaylist': opts.noplaylist,
846 'logtostderr': opts.outtmpl.get('default') == '-',
847 'consoletitle': opts.consoletitle,
848 'nopart': opts.nopart,
849 'updatetime': opts.updatetime,
850 'writedescription': opts.writedescription,
851 'writeannotations': opts.writeannotations,
852 'writeinfojson': opts.writeinfojson,
853 'allow_playlist_files': opts.allow_playlist_files,
854 'clean_infojson': opts.clean_infojson,
855 'getcomments': opts.getcomments,
856 'writethumbnail': opts.writethumbnail is True,
857 'write_all_thumbnails': opts.writethumbnail == 'all',
858 'writelink': opts.writelink,
859 'writeurllink': opts.writeurllink,
860 'writewebloclink': opts.writewebloclink,
861 'writedesktoplink': opts.writedesktoplink,
862 'writesubtitles': opts.writesubtitles,
863 'writeautomaticsub': opts.writeautomaticsub,
864 'allsubtitles': opts.allsubtitles,
865 'listsubtitles': opts.listsubtitles,
866 'subtitlesformat': opts.subtitlesformat,
867 'subtitleslangs': opts.subtitleslangs,
868 'matchtitle': decodeOption(opts.matchtitle),
869 'rejecttitle': decodeOption(opts.rejecttitle),
870 'max_downloads': opts.max_downloads,
871 'prefer_free_formats': opts.prefer_free_formats,
872 'trim_file_name': opts.trim_file_name,
873 'verbose': opts.verbose,
874 'dump_intermediate_pages': opts.dump_intermediate_pages,
875 'write_pages': opts.write_pages,
876 'load_pages': opts.load_pages,
877 'test': opts.test,
878 'keepvideo': opts.keepvideo,
879 'min_filesize': opts.min_filesize,
880 'max_filesize': opts.max_filesize,
881 'min_views': opts.min_views,
882 'max_views': opts.max_views,
883 'daterange': opts.date,
884 'cachedir': opts.cachedir,
885 'youtube_print_sig_code': opts.youtube_print_sig_code,
886 'age_limit': opts.age_limit,
887 'download_archive': opts.download_archive,
888 'break_on_existing': opts.break_on_existing,
889 'break_on_reject': opts.break_on_reject,
890 'break_per_url': opts.break_per_url,
891 'skip_playlist_after_errors': opts.skip_playlist_after_errors,
892 'cookiefile': opts.cookiefile,
893 'cookiesfrombrowser': opts.cookiesfrombrowser,
894 'legacyserverconnect': opts.legacy_server_connect,
895 'nocheckcertificate': opts.no_check_certificate,
896 'prefer_insecure': opts.prefer_insecure,
897 'enable_file_urls': opts.enable_file_urls,
898 'http_headers': opts.headers,
899 'proxy': opts.proxy,
900 'socket_timeout': opts.socket_timeout,
901 'bidi_workaround': opts.bidi_workaround,
902 'debug_printtraffic': opts.debug_printtraffic,
903 'prefer_ffmpeg': opts.prefer_ffmpeg,
904 'include_ads': opts.include_ads,
905 'default_search': opts.default_search,
906 'dynamic_mpd': opts.dynamic_mpd,
907 'extractor_args': opts.extractor_args,
908 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
909 'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
910 'encoding': opts.encoding,
911 'extract_flat': opts.extract_flat,
912 'live_from_start': opts.live_from_start,
913 'wait_for_video': opts.wait_for_video,
914 'mark_watched': opts.mark_watched,
915 'merge_output_format': opts.merge_output_format,
916 'final_ext': final_ext,
917 'postprocessors': postprocessors,
918 'fixup': opts.fixup,
919 'source_address': opts.source_address,
920 'impersonate': opts.impersonate,
921 'call_home': opts.call_home,
922 'sleep_interval_requests': opts.sleep_interval_requests,
923 'sleep_interval': opts.sleep_interval,
924 'max_sleep_interval': opts.max_sleep_interval,
925 'sleep_interval_subtitles': opts.sleep_interval_subtitles,
926 'external_downloader': opts.external_downloader,
927 'download_ranges': opts.download_ranges,
928 'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
929 'list_thumbnails': opts.list_thumbnails,
930 'playlist_items': opts.playlist_items,
931 'xattr_set_filesize': opts.xattr_set_filesize,
932 'match_filter': opts.match_filter,
933 'color': opts.color,
934 'ffmpeg_location': opts.ffmpeg_location,
935 'hls_prefer_native': opts.hls_prefer_native,
936 'hls_use_mpegts': opts.hls_use_mpegts,
937 'hls_split_discontinuity': opts.hls_split_discontinuity,
938 'external_downloader_args': opts.external_downloader_args,
939 'postprocessor_args': opts.postprocessor_args,
940 'cn_verification_proxy': opts.cn_verification_proxy,
941 'geo_verification_proxy': opts.geo_verification_proxy,
942 'geo_bypass': opts.geo_bypass,
943 'geo_bypass_country': opts.geo_bypass_country,
944 'geo_bypass_ip_block': opts.geo_bypass_ip_block,
945 '_warnings': warnings,
946 '_deprecation_warnings': deprecation_warnings,
947 'compat_opts': opts.compat_opts,
948 })
949
950
951 def _real_main(argv=None):
952 setproctitle('yt-dlp')
953
954 parser, opts, all_urls, ydl_opts = parse_options(argv)
955
956 # Dump user agent
957 if opts.dump_user_agent:
958 ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
959 write_string(f'{ua}\n', out=sys.stdout)
960 return
961
962 if print_extractor_information(opts, all_urls):
963 return
964
965 # We may need ffmpeg_location without having access to the YoutubeDL instance
966 # See https://github.com/yt-dlp/yt-dlp/issues/2191
967 if opts.ffmpeg_location:
968 FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
969
970 with YoutubeDL(ydl_opts) as ydl:
971 pre_process = opts.update_self or opts.rm_cachedir
972 actual_use = all_urls or opts.load_info_filename
973
974 if opts.rm_cachedir:
975 ydl.cache.remove()
976
977 try:
978 updater = Updater(ydl, opts.update_self)
979 if opts.update_self and updater.update() and actual_use:
980 if updater.cmd:
981 return updater.restart()
982 # This code is reachable only for zip variant in py < 3.10
983 # It makes sense to exit here, but the old behavior is to continue
984 ydl.report_warning('Restart yt-dlp to use the updated version')
985 # return 100, 'ERROR: The program must exit for the update to complete'
986 except Exception:
987 traceback.print_exc()
988 ydl._download_retcode = 100
989
990 if opts.list_impersonate_targets:
991
992 known_targets = [
993 # List of simplified targets we know are supported,
994 # to help users know what dependencies may be required.
995 (ImpersonateTarget('chrome'), 'curl_cffi'),
996 (ImpersonateTarget('edge'), 'curl_cffi'),
997 (ImpersonateTarget('safari'), 'curl_cffi'),
998 ]
999
1000 available_targets = ydl._get_available_impersonate_targets()
1001
1002 def make_row(target, handler):
1003 return [
1004 join_nonempty(target.client.title(), target.version, delim='-') or '-',
1005 join_nonempty((target.os or "").title(), target.os_version, delim='-') or '-',
1006 handler,
1007 ]
1008
1009 rows = [make_row(target, handler) for target, handler in available_targets]
1010
1011 for known_target, known_handler in known_targets:
1012 if not any(
1013 known_target in target and handler == known_handler
1014 for target, handler in available_targets
1015 ):
1016 rows.append([
1017 ydl._format_out(text, ydl.Styles.SUPPRESS)
1018 for text in make_row(known_target, f'{known_handler} (not available)')
1019 ])
1020
1021 ydl.to_screen('[info] Available impersonate targets')
1022 ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
1023 return
1024
1025 if not actual_use:
1026 if pre_process:
1027 return ydl._download_retcode
1028
1029 args = sys.argv[1:] if argv is None else argv
1030 ydl.warn_if_short_id(args)
1031
1032 # Show a useful error message and wait for keypress if not launched from shell on Windows
1033 if not args and compat_os_name == 'nt' and getattr(sys, 'frozen', False):
1034 import ctypes.wintypes
1035 import msvcrt
1036
1037 kernel32 = ctypes.WinDLL('Kernel32')
1038
1039 buffer = (1 * ctypes.wintypes.DWORD)()
1040 attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
1041 # If we only have a single process attached, then the executable was double clicked
1042 # When using `pyinstaller` with `--onefile`, two processes get attached
1043 is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
1044 if attached_processes == 1 or is_onefile and attached_processes == 2:
1045 print(parser._generate_error_message(
1046 'Do not double-click the executable, instead call it from a command line.\n'
1047 'Please read the README for further information on how to use yt-dlp: '
1048 'https://github.com/yt-dlp/yt-dlp#readme'))
1049 msvcrt.getch()
1050 _exit(2)
1051 parser.error(
1052 'You must provide at least one URL.\n'
1053 'Type yt-dlp --help to see a list of all options.')
1054
1055 parser.destroy()
1056 try:
1057 if opts.load_info_filename is not None:
1058 if all_urls:
1059 ydl.report_warning('URLs are ignored due to --load-info-json')
1060 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
1061 else:
1062 return ydl.download(all_urls)
1063 except DownloadCancelled:
1064 ydl.to_screen('Aborting remaining downloads')
1065 return 101
1066
1067
1068 def main(argv=None):
1069 global _IN_CLI
1070 _IN_CLI = True
1071 try:
1072 _exit(*variadic(_real_main(argv)))
1073 except DownloadError:
1074 _exit(1)
1075 except SameFileError as e:
1076 _exit(f'ERROR: {e}')
1077 except KeyboardInterrupt:
1078 _exit('\nERROR: Interrupted by user')
1079 except BrokenPipeError as e:
1080 # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1081 devnull = os.open(os.devnull, os.O_WRONLY)
1082 os.dup2(devnull, sys.stdout.fileno())
1083 _exit(f'\nERROR: {e}')
1084 except optparse.OptParseError as e:
1085 _exit(2, f'\n{e}')
1086
1087
1088 from .extractor import gen_extractors, list_extractors
1089
1090 __all__ = [
1091 'main',
1092 'YoutubeDL',
1093 'parse_options',
1094 'gen_extractors',
1095 'list_extractors',
1096 ]