]> jfr.im git - yt-dlp.git/blob - yt_dlp/__init__.py
[ie/mlbtv] Fix extraction (#10296)
[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
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 shell_quote,
62 traverse_obj,
63 variadic,
64 write_string,
65 )
66 from .utils.networking import std_headers
67 from .YoutubeDL import YoutubeDL
68
69 _IN_CLI = False
70
71
72 def _exit(status=0, *args):
73 for msg in args:
74 sys.stderr.write(msg)
75 raise SystemExit(status)
76
77
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]
96
97
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
103
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
126
127
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
135
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
149
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 _video_multistreams_set = set_default_compat('multistreams', 'allow_multiple_video_streams', False, remove_compat=False)
161 _audio_multistreams_set = set_default_compat('multistreams', 'allow_multiple_audio_streams', False, remove_compat=False)
162 if _video_multistreams_set is False and _audio_multistreams_set is False:
163 _unused_compat_opt('multistreams')
164 if 'filename' in opts.compat_opts:
165 if opts.outtmpl.get('default') is None:
166 opts.outtmpl.update({'default': '%(title)s-%(id)s.%(ext)s'})
167 else:
168 _unused_compat_opt('filename')
169
170
171 def validate_options(opts):
172 def validate(cndn, name, value=None, msg=None):
173 if cndn:
174 return True
175 raise ValueError((msg or 'invalid {name} "{value}" given').format(name=name, value=value))
176
177 def validate_in(name, value, items, msg=None):
178 return validate(value is None or value in items, name, value, msg)
179
180 def validate_regex(name, value, regex):
181 return validate(value is None or re.match(regex, value), name, value)
182
183 def validate_positive(name, value, strict=False):
184 return validate(value is None or value > 0 or (not strict and value == 0),
185 name, value, '{name} "{value}" must be positive' + ('' if strict else ' or 0'))
186
187 def validate_minmax(min_val, max_val, min_name, max_name=None):
188 if max_val is None or min_val is None or max_val >= min_val:
189 return
190 if not max_name:
191 min_name, max_name = f'min {min_name}', f'max {min_name}'
192 raise ValueError(f'{max_name} "{max_val}" must be must be greater than or equal to {min_name} "{min_val}"')
193
194 # Usernames and passwords
195 validate(sum(map(bool, (opts.usenetrc, opts.netrc_cmd, opts.username))) <= 1, '.netrc',
196 msg='{name}, netrc command and username/password are mutually exclusive options')
197 validate(opts.password is None or opts.username is not None, 'account username', msg='{name} missing')
198 validate(opts.ap_password is None or opts.ap_username is not None,
199 'TV Provider account username', msg='{name} missing')
200 validate_in('TV Provider', opts.ap_mso, MSO_INFO,
201 'Unsupported {name} "{value}", use --ap-list-mso to get a list of supported TV Providers')
202
203 # Numbers
204 validate_positive('autonumber start', opts.autonumber_start)
205 validate_positive('autonumber size', opts.autonumber_size, True)
206 validate_positive('concurrent fragments', opts.concurrent_fragment_downloads, True)
207 validate_positive('playlist start', opts.playliststart, True)
208 if opts.playlistend != -1:
209 validate_minmax(opts.playliststart, opts.playlistend, 'playlist start', 'playlist end')
210
211 # Time ranges
212 validate_positive('subtitles sleep interval', opts.sleep_interval_subtitles)
213 validate_positive('requests sleep interval', opts.sleep_interval_requests)
214 validate_positive('sleep interval', opts.sleep_interval)
215 validate_positive('max sleep interval', opts.max_sleep_interval)
216 if opts.sleep_interval is None:
217 validate(
218 opts.max_sleep_interval is None, 'min sleep interval',
219 msg='{name} must be specified; use --min-sleep-interval')
220 elif opts.max_sleep_interval is None:
221 opts.max_sleep_interval = opts.sleep_interval
222 else:
223 validate_minmax(opts.sleep_interval, opts.max_sleep_interval, 'sleep interval')
224
225 if opts.wait_for_video is not None:
226 min_wait, max_wait, *_ = map(parse_duration, [*opts.wait_for_video.split('-', 1), None])
227 validate(min_wait is not None and not (max_wait is None and '-' in opts.wait_for_video),
228 'time range to wait for video', opts.wait_for_video)
229 validate_minmax(min_wait, max_wait, 'time range to wait for video')
230 opts.wait_for_video = (min_wait, max_wait)
231
232 # Format sort
233 for f in opts.format_sort:
234 validate_regex('format sorting', f, FormatSorter.regex)
235
236 # Postprocessor formats
237 validate_regex('merge output format', opts.merge_output_format,
238 r'({0})(/({0}))*'.format('|'.join(map(re.escape, FFmpegMergerPP.SUPPORTED_EXTS))))
239 validate_regex('audio format', opts.audioformat, FFmpegExtractAudioPP.FORMAT_RE)
240 validate_in('subtitle format', opts.convertsubtitles, FFmpegSubtitlesConvertorPP.SUPPORTED_EXTS)
241 validate_regex('thumbnail format', opts.convertthumbnails, FFmpegThumbnailsConvertorPP.FORMAT_RE)
242 validate_regex('recode video format', opts.recodevideo, FFmpegVideoConvertorPP.FORMAT_RE)
243 validate_regex('remux video format', opts.remuxvideo, FFmpegVideoRemuxerPP.FORMAT_RE)
244 if opts.audioquality:
245 opts.audioquality = opts.audioquality.strip('k').strip('K')
246 # int_or_none prevents inf, nan
247 validate_positive('audio quality', int_or_none(float_or_none(opts.audioquality), default=0))
248
249 # Retries
250 def parse_retries(name, value):
251 if value is None:
252 return None
253 elif value in ('inf', 'infinite'):
254 return float('inf')
255 try:
256 return int(value)
257 except (TypeError, ValueError):
258 validate(False, f'{name} retry count', value)
259
260 opts.retries = parse_retries('download', opts.retries)
261 opts.fragment_retries = parse_retries('fragment', opts.fragment_retries)
262 opts.extractor_retries = parse_retries('extractor', opts.extractor_retries)
263 opts.file_access_retries = parse_retries('file access', opts.file_access_retries)
264
265 # Retry sleep function
266 def parse_sleep_func(expr):
267 NUMBER_RE = r'\d+(?:\.\d+)?'
268 op, start, limit, step, *_ = (*tuple(re.fullmatch(
269 rf'(?:(linear|exp)=)?({NUMBER_RE})(?::({NUMBER_RE})?)?(?::({NUMBER_RE}))?',
270 expr.strip()).groups()), None, None)
271
272 if op == 'exp':
273 return lambda n: min(float(start) * (float(step or 2) ** n), float(limit or 'inf'))
274 else:
275 default_step = start if op or limit else 0
276 return lambda n: min(float(start) + float(step or default_step) * n, float(limit or 'inf'))
277
278 for key, expr in opts.retry_sleep.items():
279 if not expr:
280 del opts.retry_sleep[key]
281 continue
282 try:
283 opts.retry_sleep[key] = parse_sleep_func(expr)
284 except AttributeError:
285 raise ValueError(f'invalid {key} retry sleep expression {expr!r}')
286
287 # Bytes
288 def validate_bytes(name, value):
289 if value is None:
290 return None
291 numeric_limit = parse_bytes(value)
292 validate(numeric_limit is not None, 'rate limit', value)
293 return numeric_limit
294
295 opts.ratelimit = validate_bytes('rate limit', opts.ratelimit)
296 opts.throttledratelimit = validate_bytes('throttled rate limit', opts.throttledratelimit)
297 opts.min_filesize = validate_bytes('min filesize', opts.min_filesize)
298 opts.max_filesize = validate_bytes('max filesize', opts.max_filesize)
299 opts.buffersize = validate_bytes('buffer size', opts.buffersize)
300 opts.http_chunk_size = validate_bytes('http chunk size', opts.http_chunk_size)
301
302 # Output templates
303 def validate_outtmpl(tmpl, msg):
304 err = YoutubeDL.validate_outtmpl(tmpl)
305 if err:
306 raise ValueError(f'invalid {msg} "{tmpl}": {err}')
307
308 for k, tmpl in opts.outtmpl.items():
309 validate_outtmpl(tmpl, f'{k} output template')
310 for type_, tmpl_list in opts.forceprint.items():
311 for tmpl in tmpl_list:
312 validate_outtmpl(tmpl, f'{type_} print template')
313 for type_, tmpl_list in opts.print_to_file.items():
314 for tmpl, file in tmpl_list:
315 validate_outtmpl(tmpl, f'{type_} print to file template')
316 validate_outtmpl(file, f'{type_} print to file filename')
317 validate_outtmpl(opts.sponsorblock_chapter_title, 'SponsorBlock chapter title')
318 for k, tmpl in opts.progress_template.items():
319 k = f'{k[:-6]} console title' if '-title' in k else f'{k} progress'
320 validate_outtmpl(tmpl, f'{k} template')
321
322 outtmpl_default = opts.outtmpl.get('default')
323 if outtmpl_default == '':
324 opts.skip_download = None
325 del opts.outtmpl['default']
326
327 def parse_chapters(name, value, advanced=False):
328 parse_timestamp = lambda x: float('inf') if x in ('inf', 'infinite') else parse_duration(x)
329 TIMESTAMP_RE = r'''(?x)(?:
330 (?P<start_sign>-?)(?P<start>[^-]+)
331 )?\s*-\s*(?:
332 (?P<end_sign>-?)(?P<end>[^-]+)
333 )?'''
334
335 chapters, ranges, from_url = [], [], False
336 for regex in value or []:
337 if advanced and regex == '*from-url':
338 from_url = True
339 continue
340 elif not regex.startswith('*'):
341 try:
342 chapters.append(re.compile(regex))
343 except re.error as err:
344 raise ValueError(f'invalid {name} regex "{regex}" - {err}')
345 continue
346
347 for range_ in map(str.strip, regex[1:].split(',')):
348 mobj = range_ != '-' and re.fullmatch(TIMESTAMP_RE, range_)
349 dur = mobj and [parse_timestamp(mobj.group('start') or '0'), parse_timestamp(mobj.group('end') or 'inf')]
350 signs = mobj and (mobj.group('start_sign'), mobj.group('end_sign'))
351
352 err = None
353 if None in (dur or [None]):
354 err = 'Must be of the form "*start-end"'
355 elif not advanced and any(signs):
356 err = 'Negative timestamps are not allowed'
357 else:
358 dur[0] *= -1 if signs[0] else 1
359 dur[1] *= -1 if signs[1] else 1
360 if dur[1] == float('-inf'):
361 err = '"-inf" is not a valid end'
362 if err:
363 raise ValueError(f'invalid {name} time range "{regex}". {err}')
364 ranges.append(dur)
365
366 return chapters, ranges, from_url
367
368 opts.remove_chapters, opts.remove_ranges, _ = parse_chapters('--remove-chapters', opts.remove_chapters)
369 opts.download_ranges = download_range_func(*parse_chapters('--download-sections', opts.download_ranges, True))
370
371 # Cookies from browser
372 if opts.cookiesfrombrowser:
373 container = None
374 mobj = re.fullmatch(r'''(?x)
375 (?P<name>[^+:]+)
376 (?:\s*\+\s*(?P<keyring>[^:]+))?
377 (?:\s*:\s*(?!:)(?P<profile>.+?))?
378 (?:\s*::\s*(?P<container>.+))?
379 ''', opts.cookiesfrombrowser)
380 if mobj is None:
381 raise ValueError(f'invalid cookies from browser arguments: {opts.cookiesfrombrowser}')
382 browser_name, keyring, profile, container = mobj.group('name', 'keyring', 'profile', 'container')
383 browser_name = browser_name.lower()
384 if browser_name not in SUPPORTED_BROWSERS:
385 raise ValueError(f'unsupported browser specified for cookies: "{browser_name}". '
386 f'Supported browsers are: {", ".join(sorted(SUPPORTED_BROWSERS))}')
387 if keyring is not None:
388 keyring = keyring.upper()
389 if keyring not in SUPPORTED_KEYRINGS:
390 raise ValueError(f'unsupported keyring specified for cookies: "{keyring}". '
391 f'Supported keyrings are: {", ".join(sorted(SUPPORTED_KEYRINGS))}')
392 opts.cookiesfrombrowser = (browser_name, profile, keyring, container)
393
394 if opts.impersonate is not None:
395 opts.impersonate = ImpersonateTarget.from_str(opts.impersonate.lower())
396
397 # MetadataParser
398 def metadataparser_actions(f):
399 if isinstance(f, str):
400 cmd = f'--parse-metadata {shell_quote(f)}'
401 try:
402 actions = [MetadataFromFieldPP.to_action(f)]
403 except Exception as err:
404 raise ValueError(f'{cmd} is invalid; {err}')
405 else:
406 cmd = f'--replace-in-metadata {shell_quote(f)}'
407 actions = ((MetadataParserPP.Actions.REPLACE, x, *f[1:]) for x in f[0].split(','))
408
409 for action in actions:
410 try:
411 MetadataParserPP.validate_action(*action)
412 except Exception as err:
413 raise ValueError(f'{cmd} is invalid; {err}')
414 yield action
415
416 if opts.metafromtitle is not None:
417 opts.parse_metadata.setdefault('pre_process', []).append(f'title:{opts.metafromtitle}')
418 opts.parse_metadata = {
419 k: list(itertools.chain(*map(metadataparser_actions, v)))
420 for k, v in opts.parse_metadata.items()
421 }
422
423 # Other options
424 if opts.playlist_items is not None:
425 try:
426 tuple(PlaylistEntries.parse_playlist_items(opts.playlist_items))
427 except Exception as err:
428 raise ValueError(f'Invalid playlist-items {opts.playlist_items!r}: {err}')
429
430 opts.geo_bypass_country, opts.geo_bypass_ip_block = None, None
431 if opts.geo_bypass.lower() not in ('default', 'never'):
432 try:
433 GeoUtils.random_ipv4(opts.geo_bypass)
434 except Exception:
435 raise ValueError(f'Unsupported --xff "{opts.geo_bypass}"')
436 if len(opts.geo_bypass) == 2:
437 opts.geo_bypass_country = opts.geo_bypass
438 else:
439 opts.geo_bypass_ip_block = opts.geo_bypass
440 opts.geo_bypass = opts.geo_bypass.lower() != 'never'
441
442 opts.match_filter = match_filter_func(opts.match_filter, opts.breaking_match_filter)
443
444 if opts.download_archive is not None:
445 opts.download_archive = expand_path(opts.download_archive)
446
447 if opts.ffmpeg_location is not None:
448 opts.ffmpeg_location = expand_path(opts.ffmpeg_location)
449
450 if opts.user_agent is not None:
451 opts.headers.setdefault('User-Agent', opts.user_agent)
452 if opts.referer is not None:
453 opts.headers.setdefault('Referer', opts.referer)
454
455 if opts.no_sponsorblock:
456 opts.sponsorblock_mark = opts.sponsorblock_remove = set()
457
458 default_downloader = None
459 for proto, path in opts.external_downloader.items():
460 if path == 'native':
461 continue
462 ed = get_external_downloader(path)
463 if ed is None:
464 raise ValueError(
465 f'No such {format_field(proto, None, "%s ", ignore="default")}external downloader "{path}"')
466 elif ed and proto == 'default':
467 default_downloader = ed.get_basename()
468
469 for policy in opts.color.values():
470 if policy not in ('always', 'auto', 'no_color', 'never'):
471 raise ValueError(f'"{policy}" is not a valid color policy')
472
473 warnings, deprecation_warnings = [], []
474
475 # Common mistake: -f best
476 if opts.format == 'best':
477 warnings.append('.\n '.join((
478 '"-f best" selects the best pre-merged format which is often not the best option',
479 'To let yt-dlp download and merge the best available formats, simply do not pass any format selection',
480 'If you know what you are doing and want only the best pre-merged format, use "-f b" instead to suppress this warning')))
481
482 # --(postprocessor/downloader)-args without name
483 def report_args_compat(name, value, key1, key2=None, where=None):
484 if key1 in value and key2 not in value:
485 warnings.append(f'{name.title()} arguments given without specifying name. '
486 f'The arguments will be given to {where or f"all {name}s"}')
487 return True
488 return False
489
490 if report_args_compat('external downloader', opts.external_downloader_args,
491 'default', where=default_downloader) and default_downloader:
492 # Compat with youtube-dl's behavior. See https://github.com/ytdl-org/youtube-dl/commit/49c5293014bc11ec8c009856cd63cffa6296c1e1
493 opts.external_downloader_args.setdefault(default_downloader, opts.external_downloader_args.pop('default'))
494
495 if report_args_compat('post-processor', opts.postprocessor_args, 'default-compat', 'default'):
496 opts.postprocessor_args['default'] = opts.postprocessor_args.pop('default-compat')
497 opts.postprocessor_args.setdefault('sponskrub', [])
498
499 def report_conflict(arg1, opt1, arg2='--allow-unplayable-formats', opt2='allow_unplayable_formats',
500 val1=NO_DEFAULT, val2=NO_DEFAULT, default=False):
501 if val2 is NO_DEFAULT:
502 val2 = getattr(opts, opt2)
503 if not val2:
504 return
505
506 if val1 is NO_DEFAULT:
507 val1 = getattr(opts, opt1)
508 if val1:
509 warnings.append(f'{arg1} is ignored since {arg2} was given')
510 setattr(opts, opt1, default)
511
512 # Conflicting options
513 report_conflict('--playlist-reverse', 'playlist_reverse', '--playlist-random', 'playlist_random')
514 report_conflict('--playlist-reverse', 'playlist_reverse', '--lazy-playlist', 'lazy_playlist')
515 report_conflict('--playlist-random', 'playlist_random', '--lazy-playlist', 'lazy_playlist')
516 report_conflict('--dateafter', 'dateafter', '--date', 'date', default=None)
517 report_conflict('--datebefore', 'datebefore', '--date', 'date', default=None)
518 report_conflict('--exec-before-download', 'exec_before_dl_cmd',
519 '"--exec before_dl:"', 'exec_cmd', val2=opts.exec_cmd.get('before_dl'))
520 report_conflict('--id', 'useid', '--output', 'outtmpl', val2=opts.outtmpl.get('default'))
521 report_conflict('--remux-video', 'remuxvideo', '--recode-video', 'recodevideo')
522 report_conflict('--sponskrub', 'sponskrub', '--remove-chapters', 'remove_chapters')
523 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-mark', 'sponsorblock_mark')
524 report_conflict('--sponskrub', 'sponskrub', '--sponsorblock-remove', 'sponsorblock_remove')
525 report_conflict('--sponskrub-cut', 'sponskrub_cut', '--split-chapter', 'split_chapters',
526 val1=opts.sponskrub and opts.sponskrub_cut)
527
528 # Conflicts with --allow-unplayable-formats
529 report_conflict('--embed-metadata', 'addmetadata')
530 report_conflict('--embed-chapters', 'addchapters')
531 report_conflict('--embed-info-json', 'embed_infojson')
532 report_conflict('--embed-subs', 'embedsubtitles')
533 report_conflict('--embed-thumbnail', 'embedthumbnail')
534 report_conflict('--extract-audio', 'extractaudio')
535 report_conflict('--fixup', 'fixup', val1=opts.fixup not in (None, 'never', 'ignore'), default='never')
536 report_conflict('--recode-video', 'recodevideo')
537 report_conflict('--remove-chapters', 'remove_chapters', default=[])
538 report_conflict('--remux-video', 'remuxvideo')
539 report_conflict('--sponskrub', 'sponskrub')
540 report_conflict('--sponsorblock-remove', 'sponsorblock_remove', default=set())
541 report_conflict('--xattrs', 'xattrs')
542
543 # Fully deprecated options
544 def report_deprecation(val, old, new=None):
545 if not val:
546 return
547 deprecation_warnings.append(
548 f'{old} is deprecated and may be removed in a future version. Use {new} instead' if new
549 else f'{old} is deprecated and may not work as expected')
550
551 report_deprecation(opts.sponskrub, '--sponskrub', '--sponsorblock-mark or --sponsorblock-remove')
552 report_deprecation(not opts.prefer_ffmpeg, '--prefer-avconv', 'ffmpeg')
553 # report_deprecation(opts.include_ads, '--include-ads') # We may re-implement this in future
554 # report_deprecation(opts.call_home, '--call-home') # We may re-implement this in future
555 # report_deprecation(opts.writeannotations, '--write-annotations') # It's just that no website has it
556
557 # Dependent options
558 opts.date = DateRange.day(opts.date) if opts.date else DateRange(opts.dateafter, opts.datebefore)
559
560 if opts.exec_before_dl_cmd:
561 opts.exec_cmd['before_dl'] = opts.exec_before_dl_cmd
562
563 if opts.useid: # --id is not deprecated in youtube-dl
564 opts.outtmpl['default'] = '%(id)s.%(ext)s'
565
566 if opts.overwrites: # --force-overwrites implies --no-continue
567 opts.continue_dl = False
568
569 if (opts.addmetadata or opts.sponsorblock_mark) and opts.addchapters is None:
570 # Add chapters when adding metadata or marking sponsors
571 opts.addchapters = True
572
573 if opts.extractaudio and not opts.keepvideo and opts.format is None:
574 # Do not unnecessarily download audio
575 opts.format = 'bestaudio/best'
576
577 if opts.getcomments and opts.writeinfojson is None and not opts.embed_infojson:
578 # If JSON is not printed anywhere, but comments are requested, save it to file
579 if not opts.dumpjson or opts.print_json or opts.dump_single_json:
580 opts.writeinfojson = True
581
582 if opts.allsubtitles and not (opts.embedsubtitles or opts.writeautomaticsub):
583 # --all-sub automatically sets --write-sub if --write-auto-sub is not given
584 opts.writesubtitles = True
585
586 if opts.addmetadata and opts.embed_infojson is None:
587 # If embedding metadata and infojson is present, embed it
588 opts.embed_infojson = 'if_exists'
589
590 # Ask for passwords
591 if opts.username is not None and opts.password is None:
592 opts.password = getpass.getpass('Type account password and press [Return]: ')
593 if opts.ap_username is not None and opts.ap_password is None:
594 opts.ap_password = getpass.getpass('Type TV provider account password and press [Return]: ')
595
596 return warnings, deprecation_warnings
597
598
599 def get_postprocessors(opts):
600 yield from opts.add_postprocessors
601
602 for when, actions in opts.parse_metadata.items():
603 yield {
604 'key': 'MetadataParser',
605 'actions': actions,
606 'when': when,
607 }
608 sponsorblock_query = opts.sponsorblock_mark | opts.sponsorblock_remove
609 if sponsorblock_query:
610 yield {
611 'key': 'SponsorBlock',
612 'categories': sponsorblock_query,
613 'api': opts.sponsorblock_api,
614 'when': 'after_filter',
615 }
616 if opts.convertsubtitles:
617 yield {
618 'key': 'FFmpegSubtitlesConvertor',
619 'format': opts.convertsubtitles,
620 'when': 'before_dl',
621 }
622 if opts.convertthumbnails:
623 yield {
624 'key': 'FFmpegThumbnailsConvertor',
625 'format': opts.convertthumbnails,
626 'when': 'before_dl',
627 }
628 if opts.extractaudio:
629 yield {
630 'key': 'FFmpegExtractAudio',
631 'preferredcodec': opts.audioformat,
632 'preferredquality': opts.audioquality,
633 'nopostoverwrites': opts.nopostoverwrites,
634 }
635 if opts.remuxvideo:
636 yield {
637 'key': 'FFmpegVideoRemuxer',
638 'preferedformat': opts.remuxvideo,
639 }
640 if opts.recodevideo:
641 yield {
642 'key': 'FFmpegVideoConvertor',
643 'preferedformat': opts.recodevideo,
644 }
645 # If ModifyChapters is going to remove chapters, subtitles must already be in the container.
646 if opts.embedsubtitles:
647 keep_subs = 'no-keep-subs' not in opts.compat_opts
648 yield {
649 'key': 'FFmpegEmbedSubtitle',
650 # already_have_subtitle = True prevents the file from being deleted after embedding
651 'already_have_subtitle': opts.writesubtitles and keep_subs,
652 }
653 if not opts.writeautomaticsub and keep_subs:
654 opts.writesubtitles = True
655
656 # ModifyChapters must run before FFmpegMetadataPP
657 if opts.remove_chapters or sponsorblock_query:
658 yield {
659 'key': 'ModifyChapters',
660 'remove_chapters_patterns': opts.remove_chapters,
661 'remove_sponsor_segments': opts.sponsorblock_remove,
662 'remove_ranges': opts.remove_ranges,
663 'sponsorblock_chapter_title': opts.sponsorblock_chapter_title,
664 'force_keyframes': opts.force_keyframes_at_cuts,
665 }
666 # FFmpegMetadataPP should be run after FFmpegVideoConvertorPP and
667 # FFmpegExtractAudioPP as containers before conversion may not support
668 # metadata (3gp, webm, etc.)
669 # By default ffmpeg preserves metadata applicable for both
670 # source and target containers. From this point the container won't change,
671 # so metadata can be added here.
672 if opts.addmetadata or opts.addchapters or opts.embed_infojson:
673 yield {
674 'key': 'FFmpegMetadata',
675 'add_chapters': opts.addchapters,
676 'add_metadata': opts.addmetadata,
677 'add_infojson': opts.embed_infojson,
678 }
679 # Deprecated
680 # This should be above EmbedThumbnail since sponskrub removes the thumbnail attachment
681 # but must be below EmbedSubtitle and FFmpegMetadata
682 # See https://github.com/yt-dlp/yt-dlp/issues/204 , https://github.com/faissaloo/SponSkrub/issues/29
683 # If opts.sponskrub is None, sponskrub is used, but it silently fails if the executable can't be found
684 if opts.sponskrub is not False:
685 yield {
686 'key': 'SponSkrub',
687 'path': opts.sponskrub_path,
688 'args': opts.sponskrub_args,
689 'cut': opts.sponskrub_cut,
690 'force': opts.sponskrub_force,
691 'ignoreerror': opts.sponskrub is None,
692 '_from_cli': True,
693 }
694 if opts.embedthumbnail:
695 yield {
696 'key': 'EmbedThumbnail',
697 # already_have_thumbnail = True prevents the file from being deleted after embedding
698 'already_have_thumbnail': opts.writethumbnail,
699 }
700 if not opts.writethumbnail:
701 opts.writethumbnail = True
702 opts.outtmpl['pl_thumbnail'] = ''
703 if opts.split_chapters:
704 yield {
705 'key': 'FFmpegSplitChapters',
706 'force_keyframes': opts.force_keyframes_at_cuts,
707 }
708 # XAttrMetadataPP should be run after post-processors that may change file contents
709 if opts.xattrs:
710 yield {'key': 'XAttrMetadata'}
711 if opts.concat_playlist != 'never':
712 yield {
713 'key': 'FFmpegConcat',
714 'only_multi_video': opts.concat_playlist != 'always',
715 'when': 'playlist',
716 }
717 # Exec must be the last PP of each category
718 for when, exec_cmd in opts.exec_cmd.items():
719 yield {
720 'key': 'Exec',
721 'exec_cmd': exec_cmd,
722 'when': when,
723 }
724
725
726 ParsedOptions = collections.namedtuple('ParsedOptions', ('parser', 'options', 'urls', 'ydl_opts'))
727
728
729 def parse_options(argv=None):
730 """@returns ParsedOptions(parser, opts, urls, ydl_opts)"""
731 parser, opts, urls = parseOpts(argv)
732 urls = get_urls(urls, opts.batchfile, -1 if opts.quiet and not opts.verbose else opts.verbose)
733
734 set_compat_opts(opts)
735 try:
736 warnings, deprecation_warnings = validate_options(opts)
737 except ValueError as err:
738 parser.error(f'{err}\n')
739
740 postprocessors = list(get_postprocessors(opts))
741
742 print_only = bool(opts.forceprint) and all(k not in opts.forceprint for k in POSTPROCESS_WHEN[3:])
743 any_getting = any(getattr(opts, k) for k in (
744 'dumpjson', 'dump_single_json', 'getdescription', 'getduration', 'getfilename',
745 'getformat', 'getid', 'getthumbnail', 'gettitle', 'geturl',
746 ))
747 if opts.quiet is None:
748 opts.quiet = any_getting or opts.print_json or bool(opts.forceprint)
749
750 playlist_pps = [pp for pp in postprocessors if pp.get('when') == 'playlist']
751 write_playlist_infojson = (opts.writeinfojson and not opts.clean_infojson
752 and opts.allow_playlist_files and opts.outtmpl.get('pl_infojson') != '')
753 if not any((
754 opts.extract_flat,
755 opts.dump_single_json,
756 opts.forceprint.get('playlist'),
757 opts.print_to_file.get('playlist'),
758 write_playlist_infojson,
759 )):
760 if not playlist_pps:
761 opts.extract_flat = 'discard'
762 elif playlist_pps == [{'key': 'FFmpegConcat', 'only_multi_video': True, 'when': 'playlist'}]:
763 opts.extract_flat = 'discard_in_playlist'
764
765 final_ext = (
766 opts.recodevideo if opts.recodevideo in FFmpegVideoConvertorPP.SUPPORTED_EXTS
767 else opts.remuxvideo if opts.remuxvideo in FFmpegVideoRemuxerPP.SUPPORTED_EXTS
768 else opts.audioformat if (opts.extractaudio and opts.audioformat in FFmpegExtractAudioPP.SUPPORTED_EXTS)
769 else None)
770
771 return ParsedOptions(parser, opts, urls, {
772 'usenetrc': opts.usenetrc,
773 'netrc_location': opts.netrc_location,
774 'netrc_cmd': opts.netrc_cmd,
775 'username': opts.username,
776 'password': opts.password,
777 'twofactor': opts.twofactor,
778 'videopassword': opts.videopassword,
779 'ap_mso': opts.ap_mso,
780 'ap_username': opts.ap_username,
781 'ap_password': opts.ap_password,
782 'client_certificate': opts.client_certificate,
783 'client_certificate_key': opts.client_certificate_key,
784 'client_certificate_password': opts.client_certificate_password,
785 'quiet': opts.quiet,
786 'no_warnings': opts.no_warnings,
787 'forceurl': opts.geturl,
788 'forcetitle': opts.gettitle,
789 'forceid': opts.getid,
790 'forcethumbnail': opts.getthumbnail,
791 'forcedescription': opts.getdescription,
792 'forceduration': opts.getduration,
793 'forcefilename': opts.getfilename,
794 'forceformat': opts.getformat,
795 'forceprint': opts.forceprint,
796 'print_to_file': opts.print_to_file,
797 'forcejson': opts.dumpjson or opts.print_json,
798 'dump_single_json': opts.dump_single_json,
799 'force_write_download_archive': opts.force_write_download_archive,
800 'simulate': (print_only or any_getting or None) if opts.simulate is None else opts.simulate,
801 'skip_download': opts.skip_download,
802 'format': opts.format,
803 'allow_unplayable_formats': opts.allow_unplayable_formats,
804 'ignore_no_formats_error': opts.ignore_no_formats_error,
805 'format_sort': opts.format_sort,
806 'format_sort_force': opts.format_sort_force,
807 'allow_multiple_video_streams': opts.allow_multiple_video_streams,
808 'allow_multiple_audio_streams': opts.allow_multiple_audio_streams,
809 'check_formats': opts.check_formats,
810 'listformats': opts.listformats,
811 'listformats_table': opts.listformats_table,
812 'outtmpl': opts.outtmpl,
813 'outtmpl_na_placeholder': opts.outtmpl_na_placeholder,
814 'paths': opts.paths,
815 'autonumber_size': opts.autonumber_size,
816 'autonumber_start': opts.autonumber_start,
817 'restrictfilenames': opts.restrictfilenames,
818 'windowsfilenames': opts.windowsfilenames,
819 'ignoreerrors': opts.ignoreerrors,
820 'force_generic_extractor': opts.force_generic_extractor,
821 'allowed_extractors': opts.allowed_extractors or ['default'],
822 'ratelimit': opts.ratelimit,
823 'throttledratelimit': opts.throttledratelimit,
824 'overwrites': opts.overwrites,
825 'retries': opts.retries,
826 'file_access_retries': opts.file_access_retries,
827 'fragment_retries': opts.fragment_retries,
828 'extractor_retries': opts.extractor_retries,
829 'retry_sleep_functions': opts.retry_sleep,
830 'skip_unavailable_fragments': opts.skip_unavailable_fragments,
831 'keep_fragments': opts.keep_fragments,
832 'concurrent_fragment_downloads': opts.concurrent_fragment_downloads,
833 'buffersize': opts.buffersize,
834 'noresizebuffer': opts.noresizebuffer,
835 'http_chunk_size': opts.http_chunk_size,
836 'continuedl': opts.continue_dl,
837 'noprogress': opts.quiet if opts.noprogress is None else opts.noprogress,
838 'progress_with_newline': opts.progress_with_newline,
839 'progress_template': opts.progress_template,
840 'progress_delta': opts.progress_delta,
841 'playliststart': opts.playliststart,
842 'playlistend': opts.playlistend,
843 'playlistreverse': opts.playlist_reverse,
844 'playlistrandom': opts.playlist_random,
845 'lazy_playlist': opts.lazy_playlist,
846 'noplaylist': opts.noplaylist,
847 'logtostderr': opts.outtmpl.get('default') == '-',
848 'consoletitle': opts.consoletitle,
849 'nopart': opts.nopart,
850 'updatetime': opts.updatetime,
851 'writedescription': opts.writedescription,
852 'writeannotations': opts.writeannotations,
853 'writeinfojson': opts.writeinfojson,
854 'allow_playlist_files': opts.allow_playlist_files,
855 'clean_infojson': opts.clean_infojson,
856 'getcomments': opts.getcomments,
857 'writethumbnail': opts.writethumbnail is True,
858 'write_all_thumbnails': opts.writethumbnail == 'all',
859 'writelink': opts.writelink,
860 'writeurllink': opts.writeurllink,
861 'writewebloclink': opts.writewebloclink,
862 'writedesktoplink': opts.writedesktoplink,
863 'writesubtitles': opts.writesubtitles,
864 'writeautomaticsub': opts.writeautomaticsub,
865 'allsubtitles': opts.allsubtitles,
866 'listsubtitles': opts.listsubtitles,
867 'subtitlesformat': opts.subtitlesformat,
868 'subtitleslangs': opts.subtitleslangs,
869 'matchtitle': decodeOption(opts.matchtitle),
870 'rejecttitle': decodeOption(opts.rejecttitle),
871 'max_downloads': opts.max_downloads,
872 'prefer_free_formats': opts.prefer_free_formats,
873 'trim_file_name': opts.trim_file_name,
874 'verbose': opts.verbose,
875 'dump_intermediate_pages': opts.dump_intermediate_pages,
876 'write_pages': opts.write_pages,
877 'load_pages': opts.load_pages,
878 'test': opts.test,
879 'keepvideo': opts.keepvideo,
880 'min_filesize': opts.min_filesize,
881 'max_filesize': opts.max_filesize,
882 'min_views': opts.min_views,
883 'max_views': opts.max_views,
884 'daterange': opts.date,
885 'cachedir': opts.cachedir,
886 'youtube_print_sig_code': opts.youtube_print_sig_code,
887 'age_limit': opts.age_limit,
888 'download_archive': opts.download_archive,
889 'break_on_existing': opts.break_on_existing,
890 'break_on_reject': opts.break_on_reject,
891 'break_per_url': opts.break_per_url,
892 'skip_playlist_after_errors': opts.skip_playlist_after_errors,
893 'cookiefile': opts.cookiefile,
894 'cookiesfrombrowser': opts.cookiesfrombrowser,
895 'legacyserverconnect': opts.legacy_server_connect,
896 'nocheckcertificate': opts.no_check_certificate,
897 'prefer_insecure': opts.prefer_insecure,
898 'enable_file_urls': opts.enable_file_urls,
899 'http_headers': opts.headers,
900 'proxy': opts.proxy,
901 'socket_timeout': opts.socket_timeout,
902 'bidi_workaround': opts.bidi_workaround,
903 'debug_printtraffic': opts.debug_printtraffic,
904 'prefer_ffmpeg': opts.prefer_ffmpeg,
905 'include_ads': opts.include_ads,
906 'default_search': opts.default_search,
907 'dynamic_mpd': opts.dynamic_mpd,
908 'extractor_args': opts.extractor_args,
909 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
910 'youtube_include_hls_manifest': opts.youtube_include_hls_manifest,
911 'encoding': opts.encoding,
912 'extract_flat': opts.extract_flat,
913 'live_from_start': opts.live_from_start,
914 'wait_for_video': opts.wait_for_video,
915 'mark_watched': opts.mark_watched,
916 'merge_output_format': opts.merge_output_format,
917 'final_ext': final_ext,
918 'postprocessors': postprocessors,
919 'fixup': opts.fixup,
920 'source_address': opts.source_address,
921 'impersonate': opts.impersonate,
922 'call_home': opts.call_home,
923 'sleep_interval_requests': opts.sleep_interval_requests,
924 'sleep_interval': opts.sleep_interval,
925 'max_sleep_interval': opts.max_sleep_interval,
926 'sleep_interval_subtitles': opts.sleep_interval_subtitles,
927 'external_downloader': opts.external_downloader,
928 'download_ranges': opts.download_ranges,
929 'force_keyframes_at_cuts': opts.force_keyframes_at_cuts,
930 'list_thumbnails': opts.list_thumbnails,
931 'playlist_items': opts.playlist_items,
932 'xattr_set_filesize': opts.xattr_set_filesize,
933 'match_filter': opts.match_filter,
934 'color': opts.color,
935 'ffmpeg_location': opts.ffmpeg_location,
936 'hls_prefer_native': opts.hls_prefer_native,
937 'hls_use_mpegts': opts.hls_use_mpegts,
938 'hls_split_discontinuity': opts.hls_split_discontinuity,
939 'external_downloader_args': opts.external_downloader_args,
940 'postprocessor_args': opts.postprocessor_args,
941 'cn_verification_proxy': opts.cn_verification_proxy,
942 'geo_verification_proxy': opts.geo_verification_proxy,
943 'geo_bypass': opts.geo_bypass,
944 'geo_bypass_country': opts.geo_bypass_country,
945 'geo_bypass_ip_block': opts.geo_bypass_ip_block,
946 '_warnings': warnings,
947 '_deprecation_warnings': deprecation_warnings,
948 'compat_opts': opts.compat_opts,
949 })
950
951
952 def _real_main(argv=None):
953 setproctitle('yt-dlp')
954
955 parser, opts, all_urls, ydl_opts = parse_options(argv)
956
957 # Dump user agent
958 if opts.dump_user_agent:
959 ua = traverse_obj(opts.headers, 'User-Agent', casesense=False, default=std_headers['User-Agent'])
960 write_string(f'{ua}\n', out=sys.stdout)
961 return
962
963 if print_extractor_information(opts, all_urls):
964 return
965
966 # We may need ffmpeg_location without having access to the YoutubeDL instance
967 # See https://github.com/yt-dlp/yt-dlp/issues/2191
968 if opts.ffmpeg_location:
969 FFmpegPostProcessor._ffmpeg_location.set(opts.ffmpeg_location)
970
971 with YoutubeDL(ydl_opts) as ydl:
972 pre_process = opts.update_self or opts.rm_cachedir
973 actual_use = all_urls or opts.load_info_filename
974
975 if opts.rm_cachedir:
976 ydl.cache.remove()
977
978 try:
979 updater = Updater(ydl, opts.update_self)
980 if opts.update_self and updater.update() and actual_use:
981 if updater.cmd:
982 return updater.restart()
983 # This code is reachable only for zip variant in py < 3.10
984 # It makes sense to exit here, but the old behavior is to continue
985 ydl.report_warning('Restart yt-dlp to use the updated version')
986 # return 100, 'ERROR: The program must exit for the update to complete'
987 except Exception:
988 traceback.print_exc()
989 ydl._download_retcode = 100
990
991 if opts.list_impersonate_targets:
992
993 known_targets = [
994 # List of simplified targets we know are supported,
995 # to help users know what dependencies may be required.
996 (ImpersonateTarget('chrome'), 'curl_cffi'),
997 (ImpersonateTarget('edge'), 'curl_cffi'),
998 (ImpersonateTarget('safari'), 'curl_cffi'),
999 ]
1000
1001 available_targets = ydl._get_available_impersonate_targets()
1002
1003 def make_row(target, handler):
1004 return [
1005 join_nonempty(target.client.title(), target.version, delim='-') or '-',
1006 join_nonempty((target.os or '').title(), target.os_version, delim='-') or '-',
1007 handler,
1008 ]
1009
1010 rows = [make_row(target, handler) for target, handler in available_targets]
1011
1012 for known_target, known_handler in known_targets:
1013 if not any(
1014 known_target in target and handler == known_handler
1015 for target, handler in available_targets
1016 ):
1017 rows.append([
1018 ydl._format_out(text, ydl.Styles.SUPPRESS)
1019 for text in make_row(known_target, f'{known_handler} (not available)')
1020 ])
1021
1022 ydl.to_screen('[info] Available impersonate targets')
1023 ydl.to_stdout(render_table(['Client', 'OS', 'Source'], rows, extra_gap=2, delim='-'))
1024 return
1025
1026 if not actual_use:
1027 if pre_process:
1028 return ydl._download_retcode
1029
1030 args = sys.argv[1:] if argv is None else argv
1031 ydl.warn_if_short_id(args)
1032
1033 # Show a useful error message and wait for keypress if not launched from shell on Windows
1034 if not args and compat_os_name == 'nt' and getattr(sys, 'frozen', False):
1035 import ctypes.wintypes
1036 import msvcrt
1037
1038 kernel32 = ctypes.WinDLL('Kernel32')
1039
1040 buffer = (1 * ctypes.wintypes.DWORD)()
1041 attached_processes = kernel32.GetConsoleProcessList(buffer, 1)
1042 # If we only have a single process attached, then the executable was double clicked
1043 # When using `pyinstaller` with `--onefile`, two processes get attached
1044 is_onefile = hasattr(sys, '_MEIPASS') and os.path.basename(sys._MEIPASS).startswith('_MEI')
1045 if attached_processes == 1 or is_onefile and attached_processes == 2:
1046 print(parser._generate_error_message(
1047 'Do not double-click the executable, instead call it from a command line.\n'
1048 'Please read the README for further information on how to use yt-dlp: '
1049 'https://github.com/yt-dlp/yt-dlp#readme'))
1050 msvcrt.getch()
1051 _exit(2)
1052 parser.error(
1053 'You must provide at least one URL.\n'
1054 'Type yt-dlp --help to see a list of all options.')
1055
1056 parser.destroy()
1057 try:
1058 if opts.load_info_filename is not None:
1059 if all_urls:
1060 ydl.report_warning('URLs are ignored due to --load-info-json')
1061 return ydl.download_with_info_file(expand_path(opts.load_info_filename))
1062 else:
1063 return ydl.download(all_urls)
1064 except DownloadCancelled:
1065 ydl.to_screen('Aborting remaining downloads')
1066 return 101
1067
1068
1069 def main(argv=None):
1070 global _IN_CLI
1071 _IN_CLI = True
1072 try:
1073 _exit(*variadic(_real_main(argv)))
1074 except DownloadError:
1075 _exit(1)
1076 except SameFileError as e:
1077 _exit(f'ERROR: {e}')
1078 except KeyboardInterrupt:
1079 _exit('\nERROR: Interrupted by user')
1080 except BrokenPipeError as e:
1081 # https://docs.python.org/3/library/signal.html#note-on-sigpipe
1082 devnull = os.open(os.devnull, os.O_WRONLY)
1083 os.dup2(devnull, sys.stdout.fileno())
1084 _exit(f'\nERROR: {e}')
1085 except optparse.OptParseError as e:
1086 _exit(2, f'\n{e}')
1087
1088
1089 from .extractor import gen_extractors, list_extractors
1090
1091 __all__ = [
1092 'main',
1093 'YoutubeDL',
1094 'parse_options',
1095 'gen_extractors',
1096 'list_extractors',
1097 ]