]> jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/youtube.py
projects / yt-dlp.git / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
blame | history | raw | HEAD
[Youtube] Rewrite comment extraction (#167)
[yt-dlp.git] / yt_dlp / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import hashlib
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..compat import (
16 compat_chr,
17 compat_HTTPError,
18 compat_kwargs,
19 compat_parse_qs,
20 compat_str,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 )
26 from ..jsinterp import JSInterpreter
27 from ..utils import (
28 clean_html,
29 dict_get,
30 ExtractorError,
31 format_field,
32 float_or_none,
33 int_or_none,
34 mimetype2ext,
35 parse_codecs,
36 parse_duration,
37 qualities,
38 remove_start,
39 smuggle_url,
40 str_or_none,
41 str_to_int,
42 try_get,
43 unescapeHTML,
44 unified_strdate,
45 unsmuggle_url,
46 update_url_query,
47 url_or_none,
48 urlencode_postdata,
49 urljoin,
50 )
51
52
53 class YoutubeBaseInfoExtractor(InfoExtractor):
54 """Provide base functions for Youtube extractors"""
55 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
56 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
57
58 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
59 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
60 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
61
62 _RESERVED_NAMES = (
63 r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|'
64 r'movies|results|shared|hashtag|trending|feed|feeds|'
65 r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout')
66
67 _NETRC_MACHINE = 'youtube'
68 # If True it will raise an error if no login info is provided
69 _LOGIN_REQUIRED = False
70
71 _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)'
72
73 def _ids_to_results(self, ids):
74 return [
75 self.url_result(vid_id, 'Youtube', video_id=vid_id)
76 for vid_id in ids]
77
78 def _login(self):
79 """
80 Attempt to log in to YouTube.
81 True is returned if successful or skipped.
82 False is returned if login failed.
83
84 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
85 """
86 username, password = self._get_login_info()
87 # No authentication to be performed
88 if username is None:
89 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
90 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
91 # if self._downloader.params.get('cookiefile'): # TODO remove 'and False' later - too many people using outdated cookies and open issues, remind them.
92 # self.to_screen('[Cookies] Reminder - Make sure to always use up to date cookies!')
93 return True
94
95 login_page = self._download_webpage(
96 self._LOGIN_URL, None,
97 note='Downloading login page',
98 errnote='unable to fetch login page', fatal=False)
99 if login_page is False:
100 return
101
102 login_form = self._hidden_inputs(login_page)
103
104 def req(url, f_req, note, errnote):
105 data = login_form.copy()
106 data.update({
107 'pstMsg': 1,
108 'checkConnection': 'youtube',
109 'checkedDomains': 'youtube',
110 'hl': 'en',
111 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
112 'f.req': json.dumps(f_req),
113 'flowName': 'GlifWebSignIn',
114 'flowEntry': 'ServiceLogin',
115 # TODO: reverse actual botguard identifier generation algo
116 'bgRequest': '["identifier",""]',
117 })
118 return self._download_json(
119 url, None, note=note, errnote=errnote,
120 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
121 fatal=False,
122 data=urlencode_postdata(data), headers={
123 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
124 'Google-Accounts-XSRF': 1,
125 })
126
127 def warn(message):
128 self._downloader.report_warning(message)
129
130 lookup_req = [
131 username,
132 None, [], None, 'US', None, None, 2, False, True,
133 [
134 None, None,
135 [2, 1, None, 1,
136 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
137 None, [], 4],
138 1, [None, None, []], None, None, None, True
139 ],
140 username,
141 ]
142
143 lookup_results = req(
144 self._LOOKUP_URL, lookup_req,
145 'Looking up account info', 'Unable to look up account info')
146
147 if lookup_results is False:
148 return False
149
150 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
151 if not user_hash:
152 warn('Unable to extract user hash')
153 return False
154
155 challenge_req = [
156 user_hash,
157 None, 1, None, [1, None, None, None, [password, None, True]],
158 [
159 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
160 1, [None, None, []], None, None, None, True
161 ]]
162
163 challenge_results = req(
164 self._CHALLENGE_URL, challenge_req,
165 'Logging in', 'Unable to log in')
166
167 if challenge_results is False:
168 return
169
170 login_res = try_get(challenge_results, lambda x: x[0][5], list)
171 if login_res:
172 login_msg = try_get(login_res, lambda x: x[5], compat_str)
173 warn(
174 'Unable to login: %s' % 'Invalid password'
175 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
176 return False
177
178 res = try_get(challenge_results, lambda x: x[0][-1], list)
179 if not res:
180 warn('Unable to extract result entry')
181 return False
182
183 login_challenge = try_get(res, lambda x: x[0][0], list)
184 if login_challenge:
185 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
186 if challenge_str == 'TWO_STEP_VERIFICATION':
187 # SEND_SUCCESS - TFA code has been successfully sent to phone
188 # QUOTA_EXCEEDED - reached the limit of TFA codes
189 status = try_get(login_challenge, lambda x: x[5], compat_str)
190 if status == 'QUOTA_EXCEEDED':
191 warn('Exceeded the limit of TFA codes, try later')
192 return False
193
194 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
195 if not tl:
196 warn('Unable to extract TL')
197 return False
198
199 tfa_code = self._get_tfa_info('2-step verification code')
200
201 if not tfa_code:
202 warn(
203 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
204 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
205 return False
206
207 tfa_code = remove_start(tfa_code, 'G-')
208
209 tfa_req = [
210 user_hash, None, 2, None,
211 [
212 9, None, None, None, None, None, None, None,
213 [None, tfa_code, True, 2]
214 ]]
215
216 tfa_results = req(
217 self._TFA_URL.format(tl), tfa_req,
218 'Submitting TFA code', 'Unable to submit TFA code')
219
220 if tfa_results is False:
221 return False
222
223 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
224 if tfa_res:
225 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
226 warn(
227 'Unable to finish TFA: %s' % 'Invalid TFA code'
228 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
229 return False
230
231 check_cookie_url = try_get(
232 tfa_results, lambda x: x[0][-1][2], compat_str)
233 else:
234 CHALLENGES = {
235 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
236 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
237 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
238 }
239 challenge = CHALLENGES.get(
240 challenge_str,
241 '%s returned error %s.' % (self.IE_NAME, challenge_str))
242 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
243 return False
244 else:
245 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
246
247 if not check_cookie_url:
248 warn('Unable to extract CheckCookie URL')
249 return False
250
251 check_cookie_results = self._download_webpage(
252 check_cookie_url, None, 'Checking cookie', fatal=False)
253
254 if check_cookie_results is False:
255 return False
256
257 if 'https://myaccount.google.com/' not in check_cookie_results:
258 warn('Unable to log in')
259 return False
260
261 return True
262
263 def _download_webpage_handle(self, *args, **kwargs):
264 query = kwargs.get('query', {}).copy()
265 kwargs['query'] = query
266 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
267 *args, **compat_kwargs(kwargs))
268
269 def _real_initialize(self):
270 if self._downloader is None:
271 return
272 if not self._login():
273 return
274
275 _YT_WEB_CLIENT_VERSION = '2.20210301.08.00'
276 _DEFAULT_API_DATA = {
277 'context': {
278 'client': {
279 'clientName': 'WEB',
280 'clientVersion': _YT_WEB_CLIENT_VERSION,
281 }
282 },
283 }
284
285 _DEFAULT_BASIC_API_HEADERS = {
286 'X-YouTube-Client-Name': '1',
287 'X-YouTube-Client-Version': _YT_WEB_CLIENT_VERSION
288 }
289
290 _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;'
291 _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;'
292 _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)'
293
294 def _generate_sapisidhash_header(self):
295 sapisid_cookie = self._get_cookies('https://www.youtube.com').get('SAPISID')
296 if sapisid_cookie is None:
297 return
298 time_now = round(time.time())
299 sapisidhash = hashlib.sha1((str(time_now) + " " + sapisid_cookie.value + " " + "https://www.youtube.com").encode("utf-8")).hexdigest()
300 return "SAPISIDHASH %s_%s" % (time_now, sapisidhash)
301
302 def _call_api(self, ep, query, video_id, fatal=True, headers=None,
303 note='Downloading API JSON', errnote='Unable to download API page'):
304 data = self._DEFAULT_API_DATA.copy()
305 data.update(query)
306 headers = headers or {}
307 headers.update({'content-type': 'application/json'})
308 auth = self._generate_sapisidhash_header()
309 if auth is not None:
310 headers.update({'Authorization': auth, 'X-Origin': 'https://www.youtube.com'})
311 return self._download_json(
312 'https://www.youtube.com/youtubei/v1/%s' % ep,
313 video_id=video_id, fatal=fatal, note=note, errnote=errnote,
314 data=json.dumps(data).encode('utf8'), headers=headers,
315 query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'})
316
317 def _extract_yt_initial_data(self, video_id, webpage):
318 return self._parse_json(
319 self._search_regex(
320 (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE),
321 self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'),
322 video_id)
323
324 def _extract_identity_token(self, webpage, item_id):
325 ytcfg = self._extract_ytcfg(item_id, webpage)
326 if ytcfg:
327 token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str)
328 if token:
329 return token
330 return self._search_regex(
331 r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage,
332 'identity token', default=None)
333
334 @staticmethod
335 def _extract_account_syncid(data):
336 """Extract syncId required to download private playlists of secondary channels"""
337 sync_ids = (
338 try_get(data, lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], compat_str)
339 or '').split("||")
340 if len(sync_ids) >= 2 and sync_ids[1]:
341 # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel
342 # and just "user_syncid||" for primary channel. We only want the channel_syncid
343 return sync_ids[0]
344
345 def _extract_ytcfg(self, video_id, webpage):
346 return self._parse_json(
347 self._search_regex(
348 r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
349 default='{}'), video_id, fatal=False)
350
351 def _extract_video(self, renderer):
352 video_id = renderer.get('videoId')
353 title = try_get(
354 renderer,
355 (lambda x: x['title']['runs'][0]['text'],
356 lambda x: x['title']['simpleText']), compat_str)
357 description = try_get(
358 renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'],
359 compat_str)
360 duration = parse_duration(try_get(
361 renderer, lambda x: x['lengthText']['simpleText'], compat_str))
362 view_count_text = try_get(
363 renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or ''
364 view_count = str_to_int(self._search_regex(
365 r'^([\d,]+)', re.sub(r'\s', '', view_count_text),
366 'view count', default=None))
367 uploader = try_get(
368 renderer,
369 (lambda x: x['ownerText']['runs'][0]['text'],
370 lambda x: x['shortBylineText']['runs'][0]['text']), compat_str)
371 return {
372 '_type': 'url_transparent',
373 'ie_key': YoutubeIE.ie_key(),
374 'id': video_id,
375 'url': video_id,
376 'title': title,
377 'description': description,
378 'duration': duration,
379 'view_count': view_count,
380 'uploader': uploader,
381 }
382
383
384 class YoutubeIE(YoutubeBaseInfoExtractor):
385 IE_DESC = 'YouTube.com'
386 _INVIDIOUS_SITES = (
387 # invidious-redirect websites
388 r'(?:www\.)?redirect\.invidious\.io',
389 r'(?:(?:www|dev)\.)?invidio\.us',
390 # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md
391 r'(?:www\.)?invidious\.pussthecat\.org',
392 r'(?:www\.)?invidious\.048596\.xyz',
393 r'(?:www\.)?invidious\.zee\.li',
394 r'(?:www\.)?vid\.puffyan\.us',
395 r'(?:(?:www|au)\.)?ytprivate\.com',
396 r'(?:www\.)?invidious\.namazso\.eu',
397 r'(?:www\.)?invidious\.ethibox\.fr',
398 r'(?:www\.)?inv\.skyn3t\.in',
399 r'(?:www\.)?invidious\.himiko\.cloud',
400 r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
401 r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
402 r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
403 r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
404 # youtube-dl invidious instances list
405 r'(?:(?:www|no)\.)?invidiou\.sh',
406 r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
407 r'(?:www\.)?invidious\.kabi\.tk',
408 r'(?:www\.)?invidious\.13ad\.de',
409 r'(?:www\.)?invidious\.mastodon\.host',
410 r'(?:www\.)?invidious\.zapashcanon\.fr',
411 r'(?:www\.)?invidious\.kavin\.rocks',
412 r'(?:www\.)?invidious\.tube',
413 r'(?:www\.)?invidiou\.site',
414 r'(?:www\.)?invidious\.site',
415 r'(?:www\.)?invidious\.xyz',
416 r'(?:www\.)?invidious\.nixnet\.xyz',
417 r'(?:www\.)?invidious\.drycat\.fr',
418 r'(?:www\.)?tube\.poal\.co',
419 r'(?:www\.)?tube\.connect\.cafe',
420 r'(?:www\.)?vid\.wxzm\.sx',
421 r'(?:www\.)?vid\.mint\.lgbt',
422 r'(?:www\.)?yewtu\.be',
423 r'(?:www\.)?yt\.elukerio\.org',
424 r'(?:www\.)?yt\.lelux\.fi',
425 r'(?:www\.)?invidious\.ggc-project\.de',
426 r'(?:www\.)?yt\.maisputain\.ovh',
427 r'(?:www\.)?invidious\.toot\.koeln',
428 r'(?:www\.)?invidious\.fdn\.fr',
429 r'(?:www\.)?watch\.nettohikari\.com',
430 r'(?:www\.)?kgg2m7yk5aybusll\.onion',
431 r'(?:www\.)?qklhadlycap4cnod\.onion',
432 r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
433 r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
434 r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
435 r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
436 r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
437 r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
438 )
439 _VALID_URL = r"""(?x)^
440 (
441 (?:https?://|//) # http(s):// or protocol-independent URL
442 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com|
443 (?:www\.)?deturl\.com/www\.youtube\.com|
444 (?:www\.)?pwnyoutube\.com|
445 (?:www\.)?hooktube\.com|
446 (?:www\.)?yourepeat\.com|
447 tube\.majestyc\.net|
448 %(invidious)s|
449 youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains
450 (?:.*?\#/)? # handle anchor (#/) redirect urls
451 (?: # the various things that can precede the ID:
452 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
453 |(?: # or the v= param in all its forms
454 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
455 (?:\?|\#!?) # the params delimiter ? or # or #!
456 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
457 v=
458 )
459 ))
460 |(?:
461 youtu\.be| # just youtu.be/xxxx
462 vid\.plus| # or vid.plus/xxxx
463 zwearz\.com/watch| # or zwearz.com/watch/xxxx
464 %(invidious)s
465 )/
466 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
467 )
468 )? # all until now is optional -> you can pass the naked ID
469 (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
470 (?!.*?\blist=
471 (?:
472 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
473 WL # WL are handled by the watch later IE
474 )
475 )
476 (?(1).+)? # if we found the ID, everything can follow
477 $""" % {
478 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
479 'invidious': '|'.join(_INVIDIOUS_SITES),
480 }
481 _PLAYER_INFO_RE = (
482 r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
483 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
484 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
485 )
486 _formats = {
487 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
488 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
489 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
490 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
491 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
492 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
493 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
494 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
495 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
496 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
497 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
498 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
499 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
500 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
501 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
502 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
503 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
504 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
505
506
507 # 3D videos
508 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
509 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
510 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
511 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
512 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
513 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
514 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
515
516 # Apple HTTP Live Streaming
517 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
518 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
519 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
520 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
521 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
522 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
523 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
524 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
525
526 # DASH mp4 video
527 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
528 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
529 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
530 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
531 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
532 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
533 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
534 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
535 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
536 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
537 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
538 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
539
540 # Dash mp4 audio
541 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
542 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
543 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
544 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
545 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
546 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
547 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
548
549 # Dash webm
550 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
551 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
552 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
553 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
554 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
555 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
556 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
557 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
558 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
559 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
560 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
561 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
562 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
563 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
564 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
565 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
566 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
567 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
568 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
569 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
570 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
571 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
572
573 # Dash webm audio
574 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
575 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
576
577 # Dash webm audio with opus inside
578 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
579 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
580 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
581
582 # RTMP (unnamed)
583 '_rtmp': {'protocol': 'rtmp'},
584
585 # av01 video only formats sometimes served with "unknown" codecs
586 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
587 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
588 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
589 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
590 }
591 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
592
593 _GEO_BYPASS = False
594
595 IE_NAME = 'youtube'
596 _TESTS = [
597 {
598 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
599 'info_dict': {
600 'id': 'BaW_jenozKc',
601 'ext': 'mp4',
602 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
603 'uploader': 'Philipp Hagemeister',
604 'uploader_id': 'phihag',
605 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
606 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
607 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
608 'upload_date': '20121002',
609 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
610 'categories': ['Science & Technology'],
611 'tags': ['youtube-dl'],
612 'duration': 10,
613 'view_count': int,
614 'like_count': int,
615 'dislike_count': int,
616 'start_time': 1,
617 'end_time': 9,
618 }
619 },
620 {
621 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
622 'note': 'Embed-only video (#1746)',
623 'info_dict': {
624 'id': 'yZIXLfi8CZQ',
625 'ext': 'mp4',
626 'upload_date': '20120608',
627 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
628 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
629 'uploader': 'SET India',
630 'uploader_id': 'setindia',
631 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
632 'age_limit': 18,
633 },
634 'skip': 'Private video',
635 },
636 {
637 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ',
638 'note': 'Use the first video ID in the URL',
639 'info_dict': {
640 'id': 'BaW_jenozKc',
641 'ext': 'mp4',
642 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
643 'uploader': 'Philipp Hagemeister',
644 'uploader_id': 'phihag',
645 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
646 'upload_date': '20121002',
647 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
648 'categories': ['Science & Technology'],
649 'tags': ['youtube-dl'],
650 'duration': 10,
651 'view_count': int,
652 'like_count': int,
653 'dislike_count': int,
654 },
655 'params': {
656 'skip_download': True,
657 },
658 },
659 {
660 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
661 'note': '256k DASH audio (format 141) via DASH manifest',
662 'info_dict': {
663 'id': 'a9LDPn-MO4I',
664 'ext': 'm4a',
665 'upload_date': '20121002',
666 'uploader_id': '8KVIDEO',
667 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
668 'description': '',
669 'uploader': '8KVIDEO',
670 'title': 'UHDTV TEST 8K VIDEO.mp4'
671 },
672 'params': {
673 'youtube_include_dash_manifest': True,
674 'format': '141',
675 },
676 'skip': 'format 141 not served anymore',
677 },
678 # DASH manifest with encrypted signature
679 {
680 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
681 'info_dict': {
682 'id': 'IB3lcPjvWLA',
683 'ext': 'm4a',
684 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
685 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
686 'duration': 244,
687 'uploader': 'AfrojackVEVO',
688 'uploader_id': 'AfrojackVEVO',
689 'upload_date': '20131011',
690 'abr': 129.495,
691 },
692 'params': {
693 'youtube_include_dash_manifest': True,
694 'format': '141/bestaudio[ext=m4a]',
695 },
696 },
697 # Controversy video
698 {
699 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
700 'info_dict': {
701 'id': 'T4XJQO3qol8',
702 'ext': 'mp4',
703 'duration': 219,
704 'upload_date': '20100909',
705 'uploader': 'Amazing Atheist',
706 'uploader_id': 'TheAmazingAtheist',
707 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
708 'title': 'Burning Everyone\'s Koran',
709 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
710 }
711 },
712 # Normal age-gate video (embed allowed)
713 {
714 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
715 'info_dict': {
716 'id': 'HtVdAasjOgU',
717 'ext': 'mp4',
718 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
719 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
720 'duration': 142,
721 'uploader': 'The Witcher',
722 'uploader_id': 'WitcherGame',
723 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
724 'upload_date': '20140605',
725 'age_limit': 18,
726 },
727 },
728 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
729 # YouTube Red ad is not captured for creator
730 {
731 'url': '__2ABJjxzNo',
732 'info_dict': {
733 'id': '__2ABJjxzNo',
734 'ext': 'mp4',
735 'duration': 266,
736 'upload_date': '20100430',
737 'uploader_id': 'deadmau5',
738 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
739 'creator': 'deadmau5',
740 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336',
741 'uploader': 'deadmau5',
742 'title': 'Deadmau5 - Some Chords (HD)',
743 'alt_title': 'Some Chords',
744 },
745 'expected_warnings': [
746 'DASH manifest missing',
747 ]
748 },
749 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
750 {
751 'url': 'lqQg6PlCWgI',
752 'info_dict': {
753 'id': 'lqQg6PlCWgI',
754 'ext': 'mp4',
755 'duration': 6085,
756 'upload_date': '20150827',
757 'uploader_id': 'olympic',
758 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
759 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
760 'uploader': 'Olympic',
761 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
762 },
763 'params': {
764 'skip_download': 'requires avconv',
765 }
766 },
767 # Non-square pixels
768 {
769 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
770 'info_dict': {
771 'id': '_b-2C3KPAM0',
772 'ext': 'mp4',
773 'stretched_ratio': 16 / 9.,
774 'duration': 85,
775 'upload_date': '20110310',
776 'uploader_id': 'AllenMeow',
777 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
778 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
779 'uploader': '孫ᄋᄅ',
780 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
781 },
782 },
783 # url_encoded_fmt_stream_map is empty string
784 {
785 'url': 'qEJwOuvDf7I',
786 'info_dict': {
787 'id': 'qEJwOuvDf7I',
788 'ext': 'webm',
789 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
790 'description': '',
791 'upload_date': '20150404',
792 'uploader_id': 'spbelect',
793 'uploader': 'Наблюдатели Петербурга',
794 },
795 'params': {
796 'skip_download': 'requires avconv',
797 },
798 'skip': 'This live event has ended.',
799 },
800 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
801 {
802 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
803 'info_dict': {
804 'id': 'FIl7x6_3R5Y',
805 'ext': 'webm',
806 'title': 'md5:7b81415841e02ecd4313668cde88737a',
807 'description': 'md5:116377fd2963b81ec4ce64b542173306',
808 'duration': 220,
809 'upload_date': '20150625',
810 'uploader_id': 'dorappi2000',
811 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
812 'uploader': 'dorappi2000',
813 'formats': 'mincount:31',
814 },
815 'skip': 'not actual anymore',
816 },
817 # DASH manifest with segment_list
818 {
819 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
820 'md5': '8ce563a1d667b599d21064e982ab9e31',
821 'info_dict': {
822 'id': 'CsmdDsKjzN8',
823 'ext': 'mp4',
824 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
825 'uploader': 'Airtek',
826 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
827 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
828 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
829 },
830 'params': {
831 'youtube_include_dash_manifest': True,
832 'format': '135', # bestvideo
833 },
834 'skip': 'This live event has ended.',
835 },
836 {
837 # Multifeed videos (multiple cameras), URL is for Main Camera
838 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg',
839 'info_dict': {
840 'id': 'jvGDaLqkpTg',
841 'title': 'Tom Clancy Free Weekend Rainbow Whatever',
842 'description': 'md5:e03b909557865076822aa169218d6a5d',
843 },
844 'playlist': [{
845 'info_dict': {
846 'id': 'jvGDaLqkpTg',
847 'ext': 'mp4',
848 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)',
849 'description': 'md5:e03b909557865076822aa169218d6a5d',
850 'duration': 10643,
851 'upload_date': '20161111',
852 'uploader': 'Team PGP',
853 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
854 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
855 },
856 }, {
857 'info_dict': {
858 'id': '3AKt1R1aDnw',
859 'ext': 'mp4',
860 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)',
861 'description': 'md5:e03b909557865076822aa169218d6a5d',
862 'duration': 10991,
863 'upload_date': '20161111',
864 'uploader': 'Team PGP',
865 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
866 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
867 },
868 }, {
869 'info_dict': {
870 'id': 'RtAMM00gpVc',
871 'ext': 'mp4',
872 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)',
873 'description': 'md5:e03b909557865076822aa169218d6a5d',
874 'duration': 10995,
875 'upload_date': '20161111',
876 'uploader': 'Team PGP',
877 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
878 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
879 },
880 }, {
881 'info_dict': {
882 'id': '6N2fdlP3C5U',
883 'ext': 'mp4',
884 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)',
885 'description': 'md5:e03b909557865076822aa169218d6a5d',
886 'duration': 10990,
887 'upload_date': '20161111',
888 'uploader': 'Team PGP',
889 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg',
890 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg',
891 },
892 }],
893 'params': {
894 'skip_download': True,
895 },
896 },
897 {
898 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
899 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
900 'info_dict': {
901 'id': 'gVfLd0zydlo',
902 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
903 },
904 'playlist_count': 2,
905 'skip': 'Not multifeed anymore',
906 },
907 {
908 'url': 'https://vid.plus/FlRa-iH7PGw',
909 'only_matching': True,
910 },
911 {
912 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
913 'only_matching': True,
914 },
915 {
916 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
917 # Also tests cut-off URL expansion in video description (see
918 # https://github.com/ytdl-org/youtube-dl/issues/1892,
919 # https://github.com/ytdl-org/youtube-dl/issues/8164)
920 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
921 'info_dict': {
922 'id': 'lsguqyKfVQg',
923 'ext': 'mp4',
924 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
925 'alt_title': 'Dark Walk - Position Music',
926 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
927 'duration': 133,
928 'upload_date': '20151119',
929 'uploader_id': 'IronSoulElf',
930 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
931 'uploader': 'IronSoulElf',
932 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
933 'track': 'Dark Walk - Position Music',
934 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
935 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
936 },
937 'params': {
938 'skip_download': True,
939 },
940 },
941 {
942 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
943 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
944 'only_matching': True,
945 },
946 {
947 # Video with yt:stretch=17:0
948 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
949 'info_dict': {
950 'id': 'Q39EVAstoRM',
951 'ext': 'mp4',
952 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
953 'description': 'md5:ee18a25c350637c8faff806845bddee9',
954 'upload_date': '20151107',
955 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
956 'uploader': 'CH GAMER DROID',
957 },
958 'params': {
959 'skip_download': True,
960 },
961 'skip': 'This video does not exist.',
962 },
963 {
964 # Video licensed under Creative Commons
965 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
966 'info_dict': {
967 'id': 'M4gD1WSo5mA',
968 'ext': 'mp4',
969 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
970 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
971 'duration': 721,
972 'upload_date': '20150127',
973 'uploader_id': 'BerkmanCenter',
974 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
975 'uploader': 'The Berkman Klein Center for Internet & Society',
976 'license': 'Creative Commons Attribution license (reuse allowed)',
977 },
978 'params': {
979 'skip_download': True,
980 },
981 },
982 {
983 # Channel-like uploader_url
984 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
985 'info_dict': {
986 'id': 'eQcmzGIKrzg',
987 'ext': 'mp4',
988 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
989 'description': 'md5:13a2503d7b5904ef4b223aa101628f39',
990 'duration': 4060,
991 'upload_date': '20151119',
992 'uploader': 'Bernie Sanders',
993 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
994 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
995 'license': 'Creative Commons Attribution license (reuse allowed)',
996 },
997 'params': {
998 'skip_download': True,
999 },
1000 },
1001 {
1002 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1003 'only_matching': True,
1004 },
1005 {
1006 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1007 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1008 'only_matching': True,
1009 },
1010 {
1011 # Rental video preview
1012 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1013 'info_dict': {
1014 'id': 'uGpuVWrhIzE',
1015 'ext': 'mp4',
1016 'title': 'Piku - Trailer',
1017 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1018 'upload_date': '20150811',
1019 'uploader': 'FlixMatrix',
1020 'uploader_id': 'FlixMatrixKaravan',
1021 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1022 'license': 'Standard YouTube License',
1023 },
1024 'params': {
1025 'skip_download': True,
1026 },
1027 'skip': 'This video is not available.',
1028 },
1029 {
1030 # YouTube Red video with episode data
1031 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1032 'info_dict': {
1033 'id': 'iqKdEhx-dD4',
1034 'ext': 'mp4',
1035 'title': 'Isolation - Mind Field (Ep 1)',
1036 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd',
1037 'duration': 2085,
1038 'upload_date': '20170118',
1039 'uploader': 'Vsauce',
1040 'uploader_id': 'Vsauce',
1041 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1042 'series': 'Mind Field',
1043 'season_number': 1,
1044 'episode_number': 1,
1045 },
1046 'params': {
1047 'skip_download': True,
1048 },
1049 'expected_warnings': [
1050 'Skipping DASH manifest',
1051 ],
1052 },
1053 {
1054 # The following content has been identified by the YouTube community
1055 # as inappropriate or offensive to some audiences.
1056 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1057 'info_dict': {
1058 'id': '6SJNVb0GnPI',
1059 'ext': 'mp4',
1060 'title': 'Race Differences in Intelligence',
1061 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1062 'duration': 965,
1063 'upload_date': '20140124',
1064 'uploader': 'New Century Foundation',
1065 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1066 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1067 },
1068 'params': {
1069 'skip_download': True,
1070 },
1071 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.',
1072 },
1073 {
1074 # itag 212
1075 'url': '1t24XAntNCY',
1076 'only_matching': True,
1077 },
1078 {
1079 # geo restricted to JP
1080 'url': 'sJL6WA-aGkQ',
1081 'only_matching': True,
1082 },
1083 {
1084 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1085 'only_matching': True,
1086 },
1087 {
1088 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc',
1089 'only_matching': True,
1090 },
1091 {
1092 # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m
1093 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA',
1094 'only_matching': True,
1095 },
1096 {
1097 # DRM protected
1098 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1099 'only_matching': True,
1100 },
1101 {
1102 # Video with unsupported adaptive stream type formats
1103 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1104 'info_dict': {
1105 'id': 'Z4Vy8R84T1U',
1106 'ext': 'mp4',
1107 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1108 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1109 'duration': 433,
1110 'upload_date': '20130923',
1111 'uploader': 'Amelia Putri Harwita',
1112 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1113 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1114 'formats': 'maxcount:10',
1115 },
1116 'params': {
1117 'skip_download': True,
1118 'youtube_include_dash_manifest': False,
1119 },
1120 'skip': 'not actual anymore',
1121 },
1122 {
1123 # Youtube Music Auto-generated description
1124 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1125 'info_dict': {
1126 'id': 'MgNrAu2pzNs',
1127 'ext': 'mp4',
1128 'title': 'Voyeur Girl',
1129 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1130 'upload_date': '20190312',
1131 'uploader': 'Stephen - Topic',
1132 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1133 'artist': 'Stephen',
1134 'track': 'Voyeur Girl',
1135 'album': 'it\'s too much love to know my dear',
1136 'release_date': '20190313',
1137 'release_year': 2019,
1138 },
1139 'params': {
1140 'skip_download': True,
1141 },
1142 },
1143 {
1144 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1145 'only_matching': True,
1146 },
1147 {
1148 # invalid -> valid video id redirection
1149 'url': 'DJztXj2GPfl',
1150 'info_dict': {
1151 'id': 'DJztXj2GPfk',
1152 'ext': 'mp4',
1153 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1154 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1155 'upload_date': '20090125',
1156 'uploader': 'Prochorowka',
1157 'uploader_id': 'Prochorowka',
1158 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1159 'artist': 'Panjabi MC',
1160 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1161 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1162 },
1163 'params': {
1164 'skip_download': True,
1165 },
1166 'skip': 'Video unavailable',
1167 },
1168 {
1169 # empty description results in an empty string
1170 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1171 'info_dict': {
1172 'id': 'x41yOUIvK2k',
1173 'ext': 'mp4',
1174 'title': 'IMG 3456',
1175 'description': '',
1176 'upload_date': '20170613',
1177 'uploader_id': 'ElevageOrVert',
1178 'uploader': 'ElevageOrVert',
1179 },
1180 'params': {
1181 'skip_download': True,
1182 },
1183 },
1184 {
1185 # with '};' inside yt initial data (see [1])
1186 # see [2] for an example with '};' inside ytInitialPlayerResponse
1187 # 1. https://github.com/ytdl-org/youtube-dl/issues/27093
1188 # 2. https://github.com/ytdl-org/youtube-dl/issues/27216
1189 'url': 'https://www.youtube.com/watch?v=CHqg6qOn4no',
1190 'info_dict': {
1191 'id': 'CHqg6qOn4no',
1192 'ext': 'mp4',
1193 'title': 'Part 77 Sort a list of simple types in c#',
1194 'description': 'md5:b8746fa52e10cdbf47997903f13b20dc',
1195 'upload_date': '20130831',
1196 'uploader_id': 'kudvenkat',
1197 'uploader': 'kudvenkat',
1198 },
1199 'params': {
1200 'skip_download': True,
1201 },
1202 },
1203 {
1204 # another example of '};' in ytInitialData
1205 'url': 'https://www.youtube.com/watch?v=gVfgbahppCY',
1206 'only_matching': True,
1207 },
1208 {
1209 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ',
1210 'only_matching': True,
1211 },
1212 {
1213 # https://github.com/ytdl-org/youtube-dl/pull/28094
1214 'url': 'OtqTfy26tG0',
1215 'info_dict': {
1216 'id': 'OtqTfy26tG0',
1217 'ext': 'mp4',
1218 'title': 'Burn Out',
1219 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
1220 'upload_date': '20141120',
1221 'uploader': 'The Cinematic Orchestra - Topic',
1222 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
1223 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
1224 'artist': 'The Cinematic Orchestra',
1225 'track': 'Burn Out',
1226 'album': 'Every Day',
1227 'release_data': None,
1228 'release_year': None,
1229 },
1230 'params': {
1231 'skip_download': True,
1232 },
1233 },
1234 {
1235 # controversial video, only works with bpctr when authenticated with cookies
1236 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg',
1237 'only_matching': True,
1238 },
1239 ]
1240
1241 def __init__(self, *args, **kwargs):
1242 super(YoutubeIE, self).__init__(*args, **kwargs)
1243 self._code_cache = {}
1244 self._player_cache = {}
1245
1246 def _signature_cache_id(self, example_sig):
1247 """ Return a string representation of a signature """
1248 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1249
1250 @classmethod
1251 def _extract_player_info(cls, player_url):
1252 for player_re in cls._PLAYER_INFO_RE:
1253 id_m = re.search(player_re, player_url)
1254 if id_m:
1255 break
1256 else:
1257 raise ExtractorError('Cannot identify player %r' % player_url)
1258 return id_m.group('id')
1259
1260 def _extract_signature_function(self, video_id, player_url, example_sig):
1261 player_id = self._extract_player_info(player_url)
1262
1263 # Read from filesystem cache
1264 func_id = 'js_%s_%s' % (
1265 player_id, self._signature_cache_id(example_sig))
1266 assert os.path.basename(func_id) == func_id
1267
1268 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1269 if cache_spec is not None:
1270 return lambda s: ''.join(s[i] for i in cache_spec)
1271
1272 if player_id not in self._code_cache:
1273 self._code_cache[player_id] = self._download_webpage(
1274 player_url, video_id,
1275 note='Downloading player ' + player_id,
1276 errnote='Download of %s failed' % player_url)
1277 code = self._code_cache[player_id]
1278 res = self._parse_sig_js(code)
1279
1280 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1281 cache_res = res(test_string)
1282 cache_spec = [ord(c) for c in cache_res]
1283
1284 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1285 return res
1286
1287 def _print_sig_code(self, func, example_sig):
1288 def gen_sig_code(idxs):
1289 def _genslice(start, end, step):
1290 starts = '' if start == 0 else str(start)
1291 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1292 steps = '' if step == 1 else (':%d' % step)
1293 return 's[%s%s%s]' % (starts, ends, steps)
1294
1295 step = None
1296 # Quelch pyflakes warnings - start will be set when step is set
1297 start = '(Never used)'
1298 for i, prev in zip(idxs[1:], idxs[:-1]):
1299 if step is not None:
1300 if i - prev == step:
1301 continue
1302 yield _genslice(start, prev, step)
1303 step = None
1304 continue
1305 if i - prev in [-1, 1]:
1306 step = i - prev
1307 start = prev
1308 continue
1309 else:
1310 yield 's[%d]' % prev
1311 if step is None:
1312 yield 's[%d]' % i
1313 else:
1314 yield _genslice(start, i, step)
1315
1316 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1317 cache_res = func(test_string)
1318 cache_spec = [ord(c) for c in cache_res]
1319 expr_code = ' + '.join(gen_sig_code(cache_spec))
1320 signature_id_tuple = '(%s)' % (
1321 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1322 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1323 ' return %s\n') % (signature_id_tuple, expr_code)
1324 self.to_screen('Extracted signature function:\n' + code)
1325
1326 def _parse_sig_js(self, jscode):
1327 funcname = self._search_regex(
1328 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1329 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1330 r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
1331 r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
1332 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
1333 r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1334 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1335 # Obsolete patterns
1336 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1337 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1338 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1339 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1340 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1341 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1342 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1343 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1344 jscode, 'Initial JS player signature function name', group='sig')
1345
1346 jsi = JSInterpreter(jscode)
1347 initial_function = jsi.extract_function(funcname)
1348 return lambda s: initial_function([s])
1349
1350 def _decrypt_signature(self, s, video_id, player_url):
1351 """Turn the encrypted s field into a working signature"""
1352
1353 if player_url is None:
1354 raise ExtractorError('Cannot decrypt signature without player_url')
1355
1356 if player_url.startswith('//'):
1357 player_url = 'https:' + player_url
1358 elif not re.match(r'https?://', player_url):
1359 player_url = compat_urlparse.urljoin(
1360 'https://www.youtube.com', player_url)
1361 try:
1362 player_id = (player_url, self._signature_cache_id(s))
1363 if player_id not in self._player_cache:
1364 func = self._extract_signature_function(
1365 video_id, player_url, s
1366 )
1367 self._player_cache[player_id] = func
1368 func = self._player_cache[player_id]
1369 if self._downloader.params.get('youtube_print_sig_code'):
1370 self._print_sig_code(func, s)
1371 return func(s)
1372 except Exception as e:
1373 tb = traceback.format_exc()
1374 raise ExtractorError(
1375 'Signature extraction failed: ' + tb, cause=e)
1376
1377 def _mark_watched(self, video_id, player_response):
1378 playback_url = url_or_none(try_get(
1379 player_response,
1380 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']))
1381 if not playback_url:
1382 return
1383 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1384 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1385
1386 # cpn generation algorithm is reverse engineered from base.js.
1387 # In fact it works even with dummy cpn.
1388 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1389 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1390
1391 qs.update({
1392 'ver': ['2'],
1393 'cpn': [cpn],
1394 })
1395 playback_url = compat_urlparse.urlunparse(
1396 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1397
1398 self._download_webpage(
1399 playback_url, video_id, 'Marking watched',
1400 'Unable to mark watched', fatal=False)
1401
1402 @staticmethod
1403 def _extract_urls(webpage):
1404 # Embedded YouTube player
1405 entries = [
1406 unescapeHTML(mobj.group('url'))
1407 for mobj in re.finditer(r'''(?x)
1408 (?:
1409 <iframe[^>]+?src=|
1410 data-video-url=|
1411 <embed[^>]+?src=|
1412 embedSWF\(?:\s*|
1413 <object[^>]+data=|
1414 new\s+SWFObject\(
1415 )
1416 (["\'])
1417 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1418 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1419 \1''', webpage)]
1420
1421 # lazyYT YouTube embed
1422 entries.extend(list(map(
1423 unescapeHTML,
1424 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1425
1426 # Wordpress "YouTube Video Importer" plugin
1427 matches = re.findall(r'''(?x)<div[^>]+
1428 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1429 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1430 entries.extend(m[-1] for m in matches)
1431
1432 return entries
1433
1434 @staticmethod
1435 def _extract_url(webpage):
1436 urls = YoutubeIE._extract_urls(webpage)
1437 return urls[0] if urls else None
1438
1439 @classmethod
1440 def extract_id(cls, url):
1441 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1442 if mobj is None:
1443 raise ExtractorError('Invalid URL: %s' % url)
1444 video_id = mobj.group(2)
1445 return video_id
1446
1447 def _extract_chapters_from_json(self, data, video_id, duration):
1448 chapters_list = try_get(
1449 data,
1450 lambda x: x['playerOverlays']
1451 ['playerOverlayRenderer']
1452 ['decoratedPlayerBarRenderer']
1453 ['decoratedPlayerBarRenderer']
1454 ['playerBar']
1455 ['chapteredPlayerBarRenderer']
1456 ['chapters'],
1457 list)
1458 if not chapters_list:
1459 return
1460
1461 def chapter_time(chapter):
1462 return float_or_none(
1463 try_get(
1464 chapter,
1465 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1466 int),
1467 scale=1000)
1468 chapters = []
1469 for next_num, chapter in enumerate(chapters_list, start=1):
1470 start_time = chapter_time(chapter)
1471 if start_time is None:
1472 continue
1473 end_time = (chapter_time(chapters_list[next_num])
1474 if next_num < len(chapters_list) else duration)
1475 if end_time is None:
1476 continue
1477 title = try_get(
1478 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1479 compat_str)
1480 chapters.append({
1481 'start_time': start_time,
1482 'end_time': end_time,
1483 'title': title,
1484 })
1485 return chapters
1486
1487 def _extract_yt_initial_variable(self, webpage, regex, video_id, name):
1488 return self._parse_json(self._search_regex(
1489 (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE),
1490 regex), webpage, name, default='{}'), video_id, fatal=False)
1491
1492 @staticmethod
1493 def _join_text_entries(runs):
1494 text = None
1495 for run in runs:
1496 if not isinstance(run, dict):
1497 continue
1498 sub_text = try_get(run, lambda x: x['text'], compat_str)
1499 if sub_text:
1500 if not text:
1501 text = sub_text
1502 continue
1503 text += sub_text
1504 return text
1505
1506 def _extract_comment(self, comment_renderer, parent=None):
1507 comment_id = comment_renderer.get('commentId')
1508 if not comment_id:
1509 return
1510 comment_text_runs = try_get(comment_renderer, lambda x: x['contentText']['runs']) or []
1511 text = self._join_text_entries(comment_text_runs) or ''
1512 comment_time_text = try_get(comment_renderer, lambda x: x['publishedTimeText']['runs']) or []
1513 time_text = self._join_text_entries(comment_time_text)
1514
1515 author = try_get(comment_renderer, lambda x: x['authorText']['simpleText'], compat_str)
1516 author_id = try_get(comment_renderer,
1517 lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str)
1518 votes = str_to_int(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'],
1519 lambda x: x['likeCount']), compat_str)) or 0
1520 author_thumbnail = try_get(comment_renderer,
1521 lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str)
1522
1523 author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool)
1524 is_liked = try_get(comment_renderer, lambda x: x['isLiked'], bool)
1525
1526 return {
1527 'id': comment_id,
1528 'text': text,
1529 # TODO: This should be parsed to timestamp
1530 'time_text': time_text,
1531 'like_count': votes,
1532 'is_favorited': is_liked,
1533 'author': author,
1534 'author_id': author_id,
1535 'author_thumbnail': author_thumbnail,
1536 'author_is_uploader': author_is_uploader,
1537 'parent': parent or 'root'
1538 }
1539
1540 def _comment_entries(self, root_continuation_data, identity_token, account_syncid,
1541 session_token_list, parent=None, comment_counts=None):
1542
1543 def extract_thread(parent_renderer):
1544 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
1545 if not parent:
1546 comment_counts[2] = 0
1547 for content in contents:
1548 comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer'])
1549 comment_renderer = try_get(
1550 comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get(
1551 content, (lambda x: x['commentRenderer'], dict))
1552
1553 if not comment_renderer:
1554 continue
1555 comment = self._extract_comment(comment_renderer, parent)
1556 if not comment:
1557 continue
1558 comment_counts[0] += 1
1559 yield comment
1560 # Attempt to get the replies
1561 comment_replies_renderer = try_get(
1562 comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict)
1563
1564 if comment_replies_renderer:
1565 comment_counts[2] += 1
1566 comment_entries_iter = self._comment_entries(
1567 comment_replies_renderer, identity_token, account_syncid,
1568 parent=comment.get('id'), session_token_list=session_token_list,
1569 comment_counts=comment_counts)
1570
1571 for reply_comment in comment_entries_iter:
1572 yield reply_comment
1573
1574 if not comment_counts:
1575 # comment so far, est. total comments, current comment thread #
1576 comment_counts = [0, 0, 0]
1577 headers = self._DEFAULT_BASIC_API_HEADERS.copy()
1578
1579 # TODO: Generalize the download code with TabIE
1580 if identity_token:
1581 headers['x-youtube-identity-token'] = identity_token
1582
1583 if account_syncid:
1584 headers['X-Goog-PageId'] = account_syncid
1585 headers['X-Goog-AuthUser'] = 0
1586
1587 continuation = YoutubeTabIE._extract_continuation(root_continuation_data) # TODO
1588 first_continuation = False
1589 if parent is None:
1590 first_continuation = True
1591
1592 for page_num in itertools.count(0):
1593 if not continuation:
1594 break
1595 retries = self._downloader.params.get('extractor_retries', 3)
1596 count = -1
1597 last_error = None
1598
1599 while count < retries:
1600 count += 1
1601 if last_error:
1602 self.report_warning('%s. Retrying ...' % last_error)
1603 try:
1604 query = {
1605 'ctoken': continuation['ctoken'],
1606 'pbj': 1,
1607 'type': 'next',
1608 }
1609 if parent:
1610 query['action_get_comment_replies'] = 1
1611 else:
1612 query['action_get_comments'] = 1
1613
1614 comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1])
1615 if page_num == 0:
1616 if first_continuation:
1617 note_prefix = "Downloading initial comment continuation page"
1618 else:
1619 note_prefix = " Downloading comment reply thread %d %s" % (comment_counts[2], comment_prog_str)
1620 else:
1621 note_prefix = "%sDownloading comment%s page %d %s" % (
1622 " " if parent else "",
1623 ' replies' if parent else '',
1624 page_num,
1625 comment_prog_str)
1626
1627 browse = self._download_json(
1628 'https://www.youtube.com/comment_service_ajax', None,
1629 '%s %s' % (note_prefix, '(retry #%d)' % count if count else ''),
1630 headers=headers, query=query,
1631 data=urlencode_postdata({
1632 'session_token': session_token_list[0]
1633 }))
1634 except ExtractorError as e:
1635 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404, 413):
1636 if e.cause.code == 413:
1637 self.report_warning("Assumed end of comments (received HTTP Error 413)")
1638 return
1639 # Downloading page may result in intermittent 5xx HTTP error
1640 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
1641 last_error = 'HTTP Error %s' % e.cause.code
1642 if e.cause.code == 404:
1643 last_error = last_error + " (this API is probably deprecated)"
1644 if count < retries:
1645 continue
1646 raise
1647 else:
1648 session_token = try_get(browse, lambda x: x['xsrf_token'], compat_str)
1649 if session_token:
1650 session_token_list[0] = session_token
1651
1652 response = try_get(browse,
1653 (lambda x: x['response'],
1654 lambda x: x[1]['response'])) or {}
1655
1656 if response.get('continuationContents'):
1657 break
1658
1659 # YouTube sometimes gives reload: now json if something went wrong (e.g. bad auth)
1660 if browse.get('reload'):
1661 raise ExtractorError("Invalid or missing params in continuation request", expected=False)
1662
1663 # TODO: not tested, merged from old extractor
1664 err_msg = browse.get('externalErrorMessage')
1665 if err_msg:
1666 raise ExtractorError('YouTube said: %s' % err_msg, expected=False)
1667
1668 # Youtube sometimes sends incomplete data
1669 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
1670 last_error = 'Incomplete data received'
1671 if count >= retries:
1672 self._downloader.report_error(last_error)
1673
1674 if not response:
1675 break
1676
1677 known_continuation_renderers = {
1678 'itemSectionContinuation': extract_thread,
1679 'commentRepliesContinuation': extract_thread
1680 }
1681
1682 # extract next root continuation from the results
1683 continuation_contents = try_get(
1684 response, lambda x: x['continuationContents'], dict) or {}
1685
1686 for key, value in continuation_contents.items():
1687 if key not in known_continuation_renderers:
1688 continue
1689 continuation_renderer = value
1690
1691 if first_continuation:
1692 first_continuation = False
1693 expected_comment_count = try_get(
1694 continuation_renderer,
1695 (lambda x: x['header']['commentsHeaderRenderer']['countText']['runs'][0]['text'],
1696 lambda x: x['header']['commentsHeaderRenderer']['commentsCount']['runs'][0]['text']),
1697 compat_str)
1698
1699 if expected_comment_count:
1700 comment_counts[1] = str_to_int(expected_comment_count)
1701 self.to_screen("Downloading ~%d comments" % str_to_int(expected_comment_count))
1702 yield comment_counts[1]
1703
1704 # TODO: cli arg.
1705 # 1/True for newest, 0/False for popular (default)
1706 comment_sort_index = int(True)
1707 sort_continuation_renderer = try_get(
1708 continuation_renderer,
1709 lambda x: x['header']['commentsHeaderRenderer']['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems']
1710 [comment_sort_index]['continuation']['reloadContinuationData'], dict)
1711 # If this fails, the initial continuation page
1712 # starts off with popular anyways.
1713 if sort_continuation_renderer:
1714 continuation = YoutubeTabIE._build_continuation_query(
1715 continuation=sort_continuation_renderer.get('continuation'),
1716 ctp=sort_continuation_renderer.get('clickTrackingParams'))
1717 self.to_screen("Sorting comments by %s" % ('popular' if comment_sort_index == 0 else 'newest'))
1718 break
1719
1720 for entry in known_continuation_renderers[key](continuation_renderer):
1721 yield entry
1722
1723 continuation = YoutubeTabIE._extract_continuation(continuation_renderer) # TODO
1724 break
1725
1726 def _extract_comments(self, ytcfg, video_id, contents, webpage, xsrf_token):
1727 """Entry for comment extraction"""
1728 comments = []
1729 known_entry_comment_renderers = (
1730 'itemSectionRenderer',
1731 )
1732 estimated_total = 0
1733 for entry in contents:
1734 for key, renderer in entry.items():
1735 if key not in known_entry_comment_renderers:
1736 continue
1737
1738 comment_iter = self._comment_entries(
1739 renderer,
1740 identity_token=self._extract_identity_token(webpage, item_id=video_id),
1741 account_syncid=self._extract_account_syncid(ytcfg),
1742 session_token_list=[xsrf_token])
1743
1744 for comment in comment_iter:
1745 if isinstance(comment, int):
1746 estimated_total = comment
1747 continue
1748 comments.append(comment)
1749 break
1750 self.to_screen("Downloaded %d/%d comments" % (len(comments), estimated_total))
1751 return {
1752 'comments': comments,
1753 'comment_count': len(comments),
1754 }
1755
1756 def _real_extract(self, url):
1757 url, smuggled_data = unsmuggle_url(url, {})
1758 video_id = self._match_id(url)
1759 base_url = self.http_scheme() + '//www.youtube.com/'
1760 webpage_url = base_url + 'watch?v=' + video_id
1761 webpage = self._download_webpage(
1762 webpage_url + '&has_verified=1&bpctr=9999999999',
1763 video_id, fatal=False)
1764
1765 player_response = None
1766 if webpage:
1767 player_response = self._extract_yt_initial_variable(
1768 webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE,
1769 video_id, 'initial player response')
1770 if not player_response:
1771 player_response = self._call_api(
1772 'player', {'videoId': video_id}, video_id)
1773
1774 playability_status = player_response.get('playabilityStatus') or {}
1775 if playability_status.get('reason') == 'Sign in to confirm your age':
1776 pr = self._parse_json(try_get(compat_parse_qs(
1777 self._download_webpage(
1778 base_url + 'get_video_info', video_id,
1779 'Refetching age-gated info webpage',
1780 'unable to download video info webpage', query={
1781 'video_id': video_id,
1782 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1783 }, fatal=False)),
1784 lambda x: x['player_response'][0],
1785 compat_str) or '{}', video_id)
1786 if pr:
1787 player_response = pr
1788
1789 trailer_video_id = try_get(
1790 playability_status,
1791 lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'],
1792 compat_str)
1793 if trailer_video_id:
1794 return self.url_result(
1795 trailer_video_id, self.ie_key(), trailer_video_id)
1796
1797 def get_text(x):
1798 if not x:
1799 return
1800 return x.get('simpleText') or ''.join([r['text'] for r in x['runs']])
1801
1802 search_meta = (
1803 lambda x: self._html_search_meta(x, webpage, default=None)) \
1804 if webpage else lambda x: None
1805
1806 video_details = player_response.get('videoDetails') or {}
1807 microformat = try_get(
1808 player_response,
1809 lambda x: x['microformat']['playerMicroformatRenderer'],
1810 dict) or {}
1811 video_title = video_details.get('title') \
1812 or get_text(microformat.get('title')) \
1813 or search_meta(['og:title', 'twitter:title', 'title'])
1814 video_description = video_details.get('shortDescription')
1815
1816 if not smuggled_data.get('force_singlefeed', False):
1817 if not self._downloader.params.get('noplaylist'):
1818 multifeed_metadata_list = try_get(
1819 player_response,
1820 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1821 compat_str)
1822 if multifeed_metadata_list:
1823 entries = []
1824 feed_ids = []
1825 for feed in multifeed_metadata_list.split(','):
1826 # Unquote should take place before split on comma (,) since textual
1827 # fields may contain comma as well (see
1828 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1829 feed_data = compat_parse_qs(
1830 compat_urllib_parse_unquote_plus(feed))
1831
1832 def feed_entry(name):
1833 return try_get(
1834 feed_data, lambda x: x[name][0], compat_str)
1835
1836 feed_id = feed_entry('id')
1837 if not feed_id:
1838 continue
1839 feed_title = feed_entry('title')
1840 title = video_title
1841 if feed_title:
1842 title += ' (%s)' % feed_title
1843 entries.append({
1844 '_type': 'url_transparent',
1845 'ie_key': 'Youtube',
1846 'url': smuggle_url(
1847 base_url + 'watch?v=' + feed_data['id'][0],
1848 {'force_singlefeed': True}),
1849 'title': title,
1850 })
1851 feed_ids.append(feed_id)
1852 self.to_screen(
1853 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1854 % (', '.join(feed_ids), video_id))
1855 return self.playlist_result(
1856 entries, video_id, video_title, video_description)
1857 else:
1858 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1859
1860 formats = []
1861 itags = []
1862 itag_qualities = {}
1863 player_url = None
1864 q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
1865 streaming_data = player_response.get('streamingData') or {}
1866 streaming_formats = streaming_data.get('formats') or []
1867 streaming_formats.extend(streaming_data.get('adaptiveFormats') or [])
1868 for fmt in streaming_formats:
1869 if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
1870 continue
1871
1872 itag = str_or_none(fmt.get('itag'))
1873 quality = fmt.get('quality')
1874 if itag and quality:
1875 itag_qualities[itag] = quality
1876 # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
1877 # (adding `&sq=0` to the URL) and parsing emsg box to determine the
1878 # number of fragment that would subsequently requested with (`&sq=N`)
1879 if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
1880 continue
1881
1882 fmt_url = fmt.get('url')
1883 if not fmt_url:
1884 sc = compat_parse_qs(fmt.get('signatureCipher'))
1885 fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0]))
1886 encrypted_sig = try_get(sc, lambda x: x['s'][0])
1887 if not (sc and fmt_url and encrypted_sig):
1888 continue
1889 if not player_url:
1890 if not webpage:
1891 continue
1892 player_url = self._search_regex(
1893 r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"',
1894 webpage, 'player URL', fatal=False)
1895 if not player_url:
1896 continue
1897 signature = self._decrypt_signature(sc['s'][0], video_id, player_url)
1898 sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
1899 fmt_url += '&' + sp + '=' + signature
1900
1901 if itag:
1902 itags.append(itag)
1903 tbr = float_or_none(
1904 fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
1905 dct = {
1906 'asr': int_or_none(fmt.get('audioSampleRate')),
1907 'filesize': int_or_none(fmt.get('contentLength')),
1908 'format_id': itag,
1909 'format_note': fmt.get('qualityLabel') or quality,
1910 'fps': int_or_none(fmt.get('fps')),
1911 'height': int_or_none(fmt.get('height')),
1912 'quality': q(quality),
1913 'tbr': tbr,
1914 'url': fmt_url,
1915 'width': fmt.get('width'),
1916 }
1917 mimetype = fmt.get('mimeType')
1918 if mimetype:
1919 mobj = re.match(
1920 r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype)
1921 if mobj:
1922 dct['ext'] = mimetype2ext(mobj.group(1))
1923 dct.update(parse_codecs(mobj.group(2)))
1924 no_audio = dct.get('acodec') == 'none'
1925 no_video = dct.get('vcodec') == 'none'
1926 if no_audio:
1927 dct['vbr'] = tbr
1928 if no_video:
1929 dct['abr'] = tbr
1930 if no_audio or no_video:
1931 dct['downloader_options'] = {
1932 # Youtube throttles chunks >~10M
1933 'http_chunk_size': 10485760,
1934 }
1935 if dct.get('ext'):
1936 dct['container'] = dct['ext'] + '_dash'
1937 formats.append(dct)
1938
1939 hls_manifest_url = streaming_data.get('hlsManifestUrl')
1940 if hls_manifest_url:
1941 for f in self._extract_m3u8_formats(
1942 hls_manifest_url, video_id, 'mp4', fatal=False):
1943 itag = self._search_regex(
1944 r'/itag/(\d+)', f['url'], 'itag', default=None)
1945 if itag:
1946 f['format_id'] = itag
1947 formats.append(f)
1948
1949 if self._downloader.params.get('youtube_include_dash_manifest'):
1950 dash_manifest_url = streaming_data.get('dashManifestUrl')
1951 if dash_manifest_url:
1952 for f in self._extract_mpd_formats(
1953 dash_manifest_url, video_id, fatal=False):
1954 itag = f['format_id']
1955 if itag in itags:
1956 continue
1957 if itag in itag_qualities:
1958 # Not actually usefull since the sorting is already done with "quality,res,fps,codec"
1959 # but kept to maintain feature parity (and code similarity) with youtube-dl
1960 # Remove if this causes any issues with sorting in future
1961 f['quality'] = q(itag_qualities[itag])
1962 filesize = int_or_none(self._search_regex(
1963 r'/clen/(\d+)', f.get('fragment_base_url')
1964 or f['url'], 'file size', default=None))
1965 if filesize:
1966 f['filesize'] = filesize
1967 formats.append(f)
1968
1969 if not formats:
1970 if not self._downloader.params.get('allow_unplayable_formats') and streaming_data.get('licenseInfos'):
1971 raise ExtractorError(
1972 'This video is DRM protected.', expected=True)
1973 pemr = try_get(
1974 playability_status,
1975 lambda x: x['errorScreen']['playerErrorMessageRenderer'],
1976 dict) or {}
1977 reason = get_text(pemr.get('reason')) or playability_status.get('reason')
1978 subreason = pemr.get('subreason')
1979 if subreason:
1980 subreason = clean_html(get_text(subreason))
1981 if subreason == 'The uploader has not made this video available in your country.':
1982 countries = microformat.get('availableCountries')
1983 if not countries:
1984 regions_allowed = search_meta('regionsAllowed')
1985 countries = regions_allowed.split(',') if regions_allowed else None
1986 self.raise_geo_restricted(
1987 subreason, countries)
1988 reason += '\n' + subreason
1989 if reason:
1990 raise ExtractorError(reason, expected=True)
1991
1992 self._sort_formats(formats)
1993
1994 keywords = video_details.get('keywords') or []
1995 if not keywords and webpage:
1996 keywords = [
1997 unescapeHTML(m.group('content'))
1998 for m in re.finditer(self._meta_regex('og:video:tag'), webpage)]
1999 for keyword in keywords:
2000 if keyword.startswith('yt:stretch='):
2001 w, h = keyword.split('=')[1].split(':')
2002 w, h = int(w), int(h)
2003 if w > 0 and h > 0:
2004 ratio = w / h
2005 for f in formats:
2006 if f.get('vcodec') != 'none':
2007 f['stretched_ratio'] = ratio
2008
2009 thumbnails = []
2010 for container in (video_details, microformat):
2011 for thumbnail in (try_get(
2012 container,
2013 lambda x: x['thumbnail']['thumbnails'], list) or []):
2014 thumbnail_url = thumbnail.get('url')
2015 if not thumbnail_url:
2016 continue
2017 thumbnails.append({
2018 'height': int_or_none(thumbnail.get('height')),
2019 'url': thumbnail_url,
2020 'width': int_or_none(thumbnail.get('width')),
2021 })
2022 if thumbnails:
2023 break
2024 else:
2025 thumbnail = search_meta(['og:image', 'twitter:image'])
2026 if thumbnail:
2027 thumbnails = [{'url': thumbnail}]
2028
2029 category = microformat.get('category') or search_meta('genre')
2030 channel_id = video_details.get('channelId') \
2031 or microformat.get('externalChannelId') \
2032 or search_meta('channelId')
2033 duration = int_or_none(
2034 video_details.get('lengthSeconds')
2035 or microformat.get('lengthSeconds')) \
2036 or parse_duration(search_meta('duration'))
2037 is_live = video_details.get('isLive')
2038 owner_profile_url = microformat.get('ownerProfileUrl')
2039
2040 info = {
2041 'id': video_id,
2042 'title': self._live_title(video_title) if is_live else video_title,
2043 'formats': formats,
2044 'thumbnails': thumbnails,
2045 'description': video_description,
2046 'upload_date': unified_strdate(
2047 microformat.get('uploadDate')
2048 or search_meta('uploadDate')),
2049 'uploader': video_details['author'],
2050 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None,
2051 'uploader_url': owner_profile_url,
2052 'channel_id': channel_id,
2053 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None,
2054 'duration': duration,
2055 'view_count': int_or_none(
2056 video_details.get('viewCount')
2057 or microformat.get('viewCount')
2058 or search_meta('interactionCount')),
2059 'average_rating': float_or_none(video_details.get('averageRating')),
2060 'age_limit': 18 if (
2061 microformat.get('isFamilySafe') is False
2062 or search_meta('isFamilyFriendly') == 'false'
2063 or search_meta('og:restrictions:age') == '18+') else 0,
2064 'webpage_url': webpage_url,
2065 'categories': [category] if category else None,
2066 'tags': keywords,
2067 'is_live': is_live,
2068 'playable_in_embed': playability_status.get('playableInEmbed'),
2069 'was_live': video_details.get('isLiveContent')
2070 }
2071
2072 pctr = try_get(
2073 player_response,
2074 lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict)
2075 subtitles = {}
2076 if pctr:
2077 def process_language(container, base_url, lang_code, query):
2078 lang_subs = []
2079 for fmt in self._SUBTITLE_FORMATS:
2080 query.update({
2081 'fmt': fmt,
2082 })
2083 lang_subs.append({
2084 'ext': fmt,
2085 'url': update_url_query(base_url, query),
2086 })
2087 container[lang_code] = lang_subs
2088
2089 for caption_track in (pctr.get('captionTracks') or []):
2090 base_url = caption_track.get('baseUrl')
2091 if not base_url:
2092 continue
2093 if caption_track.get('kind') != 'asr':
2094 lang_code = caption_track.get('languageCode')
2095 if not lang_code:
2096 continue
2097 process_language(
2098 subtitles, base_url, lang_code, {})
2099 continue
2100 automatic_captions = {}
2101 for translation_language in (pctr.get('translationLanguages') or []):
2102 translation_language_code = translation_language.get('languageCode')
2103 if not translation_language_code:
2104 continue
2105 process_language(
2106 automatic_captions, base_url, translation_language_code,
2107 {'tlang': translation_language_code})
2108 info['automatic_captions'] = automatic_captions
2109 info['subtitles'] = subtitles
2110
2111 parsed_url = compat_urllib_parse_urlparse(url)
2112 for component in [parsed_url.fragment, parsed_url.query]:
2113 query = compat_parse_qs(component)
2114 for k, v in query.items():
2115 for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]:
2116 d_k += '_time'
2117 if d_k not in info and k in s_ks:
2118 info[d_k] = parse_duration(query[k][0])
2119
2120 # Youtube Music Auto-generated description
2121 if video_description:
2122 mobj = re.search(r'(?s)(?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?.+\nAuto-generated by YouTube\.\s*$', video_description)
2123 if mobj:
2124 release_year = mobj.group('release_year')
2125 release_date = mobj.group('release_date')
2126 if release_date:
2127 release_date = release_date.replace('-', '')
2128 if not release_year:
2129 release_year = release_date[:4]
2130 info.update({
2131 'album': mobj.group('album'.strip()),
2132 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
2133 'track': mobj.group('track').strip(),
2134 'release_date': release_date,
2135 'release_year': int_or_none(release_year),
2136 })
2137
2138 initial_data = None
2139 if webpage:
2140 initial_data = self._extract_yt_initial_variable(
2141 webpage, self._YT_INITIAL_DATA_RE, video_id,
2142 'yt initial data')
2143 if not initial_data:
2144 initial_data = self._call_api(
2145 'next', {'videoId': video_id}, video_id, fatal=False)
2146
2147 if not is_live:
2148 try:
2149 # This will error if there is no livechat
2150 initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2151 info['subtitles']['live_chat'] = [{
2152 'video_id': video_id,
2153 'ext': 'json',
2154 'protocol': 'youtube_live_chat_replay',
2155 }]
2156 except (KeyError, IndexError, TypeError):
2157 pass
2158
2159 if initial_data:
2160 chapters = self._extract_chapters_from_json(
2161 initial_data, video_id, duration)
2162 if not chapters:
2163 for engagment_pannel in (initial_data.get('engagementPanels') or []):
2164 contents = try_get(
2165 engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'],
2166 list)
2167 if not contents:
2168 continue
2169
2170 def chapter_time(mmlir):
2171 return parse_duration(
2172 get_text(mmlir.get('timeDescription')))
2173
2174 chapters = []
2175 for next_num, content in enumerate(contents, start=1):
2176 mmlir = content.get('macroMarkersListItemRenderer') or {}
2177 start_time = chapter_time(mmlir)
2178 end_time = chapter_time(try_get(
2179 contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \
2180 if next_num < len(contents) else duration
2181 if start_time is None or end_time is None:
2182 continue
2183 chapters.append({
2184 'start_time': start_time,
2185 'end_time': end_time,
2186 'title': get_text(mmlir.get('title')),
2187 })
2188 if chapters:
2189 break
2190 if chapters:
2191 info['chapters'] = chapters
2192
2193 contents = try_get(
2194 initial_data,
2195 lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'],
2196 list) or []
2197 for content in contents:
2198 vpir = content.get('videoPrimaryInfoRenderer')
2199 if vpir:
2200 stl = vpir.get('superTitleLink')
2201 if stl:
2202 stl = get_text(stl)
2203 if try_get(
2204 vpir,
2205 lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN':
2206 info['location'] = stl
2207 else:
2208 mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl)
2209 if mobj:
2210 info.update({
2211 'series': mobj.group(1),
2212 'season_number': int(mobj.group(2)),
2213 'episode_number': int(mobj.group(3)),
2214 })
2215 for tlb in (try_get(
2216 vpir,
2217 lambda x: x['videoActions']['menuRenderer']['topLevelButtons'],
2218 list) or []):
2219 tbr = tlb.get('toggleButtonRenderer') or {}
2220 for getter, regex in [(
2221 lambda x: x['defaultText']['accessibility']['accessibilityData'],
2222 r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([
2223 lambda x: x['accessibility'],
2224 lambda x: x['accessibilityData']['accessibilityData'],
2225 ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]:
2226 label = (try_get(tbr, getter, dict) or {}).get('label')
2227 if label:
2228 mobj = re.match(regex, label)
2229 if mobj:
2230 info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count'))
2231 break
2232 sbr_tooltip = try_get(
2233 vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip'])
2234 if sbr_tooltip:
2235 like_count, dislike_count = sbr_tooltip.split(' / ')
2236 info.update({
2237 'like_count': str_to_int(like_count),
2238 'dislike_count': str_to_int(dislike_count),
2239 })
2240 vsir = content.get('videoSecondaryInfoRenderer')
2241 if vsir:
2242 info['channel'] = get_text(try_get(
2243 vsir,
2244 lambda x: x['owner']['videoOwnerRenderer']['title'],
2245 compat_str))
2246 rows = try_get(
2247 vsir,
2248 lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'],
2249 list) or []
2250 multiple_songs = False
2251 for row in rows:
2252 if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True:
2253 multiple_songs = True
2254 break
2255 for row in rows:
2256 mrr = row.get('metadataRowRenderer') or {}
2257 mrr_title = mrr.get('title')
2258 if not mrr_title:
2259 continue
2260 mrr_title = get_text(mrr['title'])
2261 mrr_contents_text = get_text(mrr['contents'][0])
2262 if mrr_title == 'License':
2263 info['license'] = mrr_contents_text
2264 elif not multiple_songs:
2265 if mrr_title == 'Album':
2266 info['album'] = mrr_contents_text
2267 elif mrr_title == 'Artist':
2268 info['artist'] = mrr_contents_text
2269 elif mrr_title == 'Song':
2270 info['track'] = mrr_contents_text
2271
2272 fallbacks = {
2273 'channel': 'uploader',
2274 'channel_id': 'uploader_id',
2275 'channel_url': 'uploader_url',
2276 }
2277 for to, frm in fallbacks.items():
2278 if not info.get(to):
2279 info[to] = info.get(frm)
2280
2281 for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
2282 v = info.get(s_k)
2283 if v:
2284 info[d_k] = v
2285
2286 # get xsrf for annotations or comments
2287 get_annotations = self._downloader.params.get('writeannotations', False)
2288 get_comments = self._downloader.params.get('getcomments', False)
2289 if get_annotations or get_comments:
2290 xsrf_token = None
2291 ytcfg = self._extract_ytcfg(video_id, webpage)
2292 if ytcfg:
2293 xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str)
2294 if not xsrf_token:
2295 xsrf_token = self._search_regex(
2296 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2',
2297 webpage, 'xsrf token', group='xsrf_token', fatal=False)
2298
2299 # annotations
2300 if get_annotations:
2301 invideo_url = try_get(
2302 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2303 if xsrf_token and invideo_url:
2304 xsrf_field_name = None
2305 if ytcfg:
2306 xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str)
2307 if not xsrf_field_name:
2308 xsrf_field_name = self._search_regex(
2309 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2310 webpage, 'xsrf field name',
2311 group='xsrf_field_name', default='session_token')
2312 info['annotations'] = self._download_webpage(
2313 self._proto_relative_url(invideo_url),
2314 video_id, note='Downloading annotations',
2315 errnote='Unable to download video annotations', fatal=False,
2316 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2317
2318 if get_comments:
2319 info['__post_extractor'] = lambda: self._extract_comments(ytcfg, video_id, contents, webpage, xsrf_token)
2320
2321 self.mark_watched(video_id, player_response)
2322
2323 return info
2324
2325
2326 class YoutubeTabIE(YoutubeBaseInfoExtractor):
2327 IE_DESC = 'YouTube.com tab'
2328 _VALID_URL = r'''(?x)
2329 https?://
2330 (?:\w+\.)?
2331 (?:
2332 youtube(?:kids)?\.com|
2333 invidio\.us
2334 )/
2335 (?:
2336 (?:channel|c|user)/|
2337 (?P<not_channel>
2338 feed/|hashtag/|
2339 (?:playlist|watch)\?.*?\blist=
2340 )|
2341 (?!(?:%s)\b) # Direct URLs
2342 )
2343 (?P<id>[^/?\#&]+)
2344 ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
2345 IE_NAME = 'youtube:tab'
2346
2347 _TESTS = [{
2348 # playlists, multipage
2349 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid',
2350 'playlist_mincount': 94,
2351 'info_dict': {
2352 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2353 'title': 'Игорь Клейнер - Playlists',
2354 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2355 'uploader': 'Игорь Клейнер',
2356 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2357 },
2358 }, {
2359 # playlists, multipage, different order
2360 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2361 'playlist_mincount': 94,
2362 'info_dict': {
2363 'id': 'UCqj7Cz7revf5maW9g5pgNcg',
2364 'title': 'Игорь Клейнер - Playlists',
2365 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2',
2366 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg',
2367 'uploader': 'Игорь Клейнер',
2368 },
2369 }, {
2370 # playlists, singlepage
2371 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2372 'playlist_mincount': 4,
2373 'info_dict': {
2374 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2375 'title': 'ThirstForScience - Playlists',
2376 'description': 'md5:609399d937ea957b0f53cbffb747a14c',
2377 'uploader': 'ThirstForScience',
2378 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ',
2379 }
2380 }, {
2381 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
2382 'only_matching': True,
2383 }, {
2384 # basic, single video playlist
2385 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2386 'info_dict': {
2387 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2388 'uploader': 'Sergey M.',
2389 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2390 'title': 'youtube-dl public playlist',
2391 },
2392 'playlist_count': 1,
2393 }, {
2394 # empty playlist
2395 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2396 'info_dict': {
2397 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2398 'uploader': 'Sergey M.',
2399 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2400 'title': 'youtube-dl empty playlist',
2401 },
2402 'playlist_count': 0,
2403 }, {
2404 # Home tab
2405 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured',
2406 'info_dict': {
2407 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2408 'title': 'lex will - Home',
2409 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2410 'uploader': 'lex will',
2411 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2412 },
2413 'playlist_mincount': 2,
2414 }, {
2415 # Videos tab
2416 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos',
2417 'info_dict': {
2418 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2419 'title': 'lex will - Videos',
2420 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2421 'uploader': 'lex will',
2422 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2423 },
2424 'playlist_mincount': 975,
2425 }, {
2426 # Videos tab, sorted by popular
2427 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid',
2428 'info_dict': {
2429 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2430 'title': 'lex will - Videos',
2431 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2432 'uploader': 'lex will',
2433 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2434 },
2435 'playlist_mincount': 199,
2436 }, {
2437 # Playlists tab
2438 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists',
2439 'info_dict': {
2440 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2441 'title': 'lex will - Playlists',
2442 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2443 'uploader': 'lex will',
2444 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2445 },
2446 'playlist_mincount': 17,
2447 }, {
2448 # Community tab
2449 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community',
2450 'info_dict': {
2451 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2452 'title': 'lex will - Community',
2453 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2454 'uploader': 'lex will',
2455 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2456 },
2457 'playlist_mincount': 18,
2458 }, {
2459 # Channels tab
2460 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels',
2461 'info_dict': {
2462 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2463 'title': 'lex will - Channels',
2464 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488',
2465 'uploader': 'lex will',
2466 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2467 },
2468 'playlist_mincount': 12,
2469 }, {
2470 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2471 'only_matching': True,
2472 }, {
2473 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2474 'only_matching': True,
2475 }, {
2476 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA',
2477 'only_matching': True,
2478 }, {
2479 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2480 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2481 'info_dict': {
2482 'title': '29C3: Not my department',
2483 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2484 'uploader': 'Christiaan008',
2485 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg',
2486 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268',
2487 },
2488 'playlist_count': 96,
2489 }, {
2490 'note': 'Large playlist',
2491 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2492 'info_dict': {
2493 'title': 'Uploads from Cauchemar',
2494 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2495 'uploader': 'Cauchemar',
2496 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q',
2497 },
2498 'playlist_mincount': 1123,
2499 }, {
2500 # even larger playlist, 8832 videos
2501 'url': 'http://www.youtube.com/user/NASAgovVideo/videos',
2502 'only_matching': True,
2503 }, {
2504 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2505 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2506 'info_dict': {
2507 'title': 'Uploads from Interstellar Movie',
2508 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2509 'uploader': 'Interstellar Movie',
2510 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA',
2511 },
2512 'playlist_mincount': 21,
2513 }, {
2514 # https://github.com/ytdl-org/youtube-dl/issues/21844
2515 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2516 'info_dict': {
2517 'title': 'Data Analysis with Dr Mike Pound',
2518 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2519 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA',
2520 'uploader': 'Computerphile',
2521 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487',
2522 },
2523 'playlist_mincount': 11,
2524 }, {
2525 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2526 'only_matching': True,
2527 }, {
2528 # Playlist URL that does not actually serve a playlist
2529 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2530 'info_dict': {
2531 'id': 'FqZTN594JQw',
2532 'ext': 'webm',
2533 'title': "Smiley's People 01 detective, Adventure Series, Action",
2534 'uploader': 'STREEM',
2535 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2536 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2537 'upload_date': '20150526',
2538 'license': 'Standard YouTube License',
2539 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2540 'categories': ['People & Blogs'],
2541 'tags': list,
2542 'view_count': int,
2543 'like_count': int,
2544 'dislike_count': int,
2545 },
2546 'params': {
2547 'skip_download': True,
2548 },
2549 'skip': 'This video is not available.',
2550 'add_ie': [YoutubeIE.ie_key()],
2551 }, {
2552 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2553 'only_matching': True,
2554 }, {
2555 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
2556 'only_matching': True,
2557 }, {
2558 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
2559 'info_dict': {
2560 'id': '9Auq9mYxFEE',
2561 'ext': 'mp4',
2562 'title': compat_str,
2563 'uploader': 'Sky News',
2564 'uploader_id': 'skynews',
2565 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews',
2566 'upload_date': '20191102',
2567 'description': 'md5:85ddd75d888674631aaf9599a9a0b0ae',
2568 'categories': ['News & Politics'],
2569 'tags': list,
2570 'like_count': int,
2571 'dislike_count': int,
2572 },
2573 'params': {
2574 'skip_download': True,
2575 },
2576 }, {
2577 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2578 'info_dict': {
2579 'id': 'a48o2S1cPoo',
2580 'ext': 'mp4',
2581 'title': 'The Young Turks - Live Main Show',
2582 'uploader': 'The Young Turks',
2583 'uploader_id': 'TheYoungTurks',
2584 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2585 'upload_date': '20150715',
2586 'license': 'Standard YouTube License',
2587 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2588 'categories': ['News & Politics'],
2589 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2590 'like_count': int,
2591 'dislike_count': int,
2592 },
2593 'params': {
2594 'skip_download': True,
2595 },
2596 'only_matching': True,
2597 }, {
2598 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2599 'only_matching': True,
2600 }, {
2601 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2602 'only_matching': True,
2603 }, {
2604 'url': 'https://www.youtube.com/feed/trending',
2605 'only_matching': True,
2606 }, {
2607 # needs auth
2608 'url': 'https://www.youtube.com/feed/library',
2609 'only_matching': True,
2610 }, {
2611 # needs auth
2612 'url': 'https://www.youtube.com/feed/history',
2613 'only_matching': True,
2614 }, {
2615 # needs auth
2616 'url': 'https://www.youtube.com/feed/subscriptions',
2617 'only_matching': True,
2618 }, {
2619 # needs auth
2620 'url': 'https://www.youtube.com/feed/watch_later',
2621 'only_matching': True,
2622 }, {
2623 # no longer available?
2624 'url': 'https://www.youtube.com/feed/recommended',
2625 'only_matching': True,
2626 }, {
2627 # inline playlist with not always working continuations
2628 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C',
2629 'only_matching': True,
2630 }, {
2631 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8',
2632 'only_matching': True,
2633 }, {
2634 'url': 'https://www.youtube.com/course',
2635 'only_matching': True,
2636 }, {
2637 'url': 'https://www.youtube.com/zsecurity',
2638 'only_matching': True,
2639 }, {
2640 'url': 'http://www.youtube.com/NASAgovVideo/videos',
2641 'only_matching': True,
2642 }, {
2643 'url': 'https://www.youtube.com/TheYoungTurks/live',
2644 'only_matching': True,
2645 }]
2646
2647 @classmethod
2648 def suitable(cls, url):
2649 return False if YoutubeIE.suitable(url) else super(
2650 YoutubeTabIE, cls).suitable(url)
2651
2652 def _extract_channel_id(self, webpage):
2653 channel_id = self._html_search_meta(
2654 'channelId', webpage, 'channel id', default=None)
2655 if channel_id:
2656 return channel_id
2657 channel_url = self._html_search_meta(
2658 ('og:url', 'al:ios:url', 'al:android:url', 'al:web:url',
2659 'twitter:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad',
2660 'twitter:app:url:googleplay'), webpage, 'channel url')
2661 return self._search_regex(
2662 r'https?://(?:www\.)?youtube\.com/channel/([^/?#&])+',
2663 channel_url, 'channel id')
2664
2665 @staticmethod
2666 def _extract_basic_item_renderer(item):
2667 # Modified from _extract_grid_item_renderer
2668 known_renderers = (
2669 'playlistRenderer', 'videoRenderer', 'channelRenderer'
2670 'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer'
2671 )
2672 for key, renderer in item.items():
2673 if key not in known_renderers:
2674 continue
2675 return renderer
2676
2677 def _grid_entries(self, grid_renderer):
2678 for item in grid_renderer['items']:
2679 if not isinstance(item, dict):
2680 continue
2681 renderer = self._extract_basic_item_renderer(item)
2682 if not isinstance(renderer, dict):
2683 continue
2684 title = try_get(
2685 renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2686 # playlist
2687 playlist_id = renderer.get('playlistId')
2688 if playlist_id:
2689 yield self.url_result(
2690 'https://www.youtube.com/playlist?list=%s' % playlist_id,
2691 ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
2692 video_title=title)
2693 # video
2694 video_id = renderer.get('videoId')
2695 if video_id:
2696 yield self._extract_video(renderer)
2697 # channel
2698 channel_id = renderer.get('channelId')
2699 if channel_id:
2700 title = try_get(
2701 renderer, lambda x: x['title']['simpleText'], compat_str)
2702 yield self.url_result(
2703 'https://www.youtube.com/channel/%s' % channel_id,
2704 ie=YoutubeTabIE.ie_key(), video_title=title)
2705
2706 def _shelf_entries_from_content(self, shelf_renderer):
2707 content = shelf_renderer.get('content')
2708 if not isinstance(content, dict):
2709 return
2710 renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer')
2711 if renderer:
2712 # TODO: add support for nested playlists so each shelf is processed
2713 # as separate playlist
2714 # TODO: this includes only first N items
2715 for entry in self._grid_entries(renderer):
2716 yield entry
2717 renderer = content.get('horizontalListRenderer')
2718 if renderer:
2719 # TODO
2720 pass
2721
2722 def _shelf_entries(self, shelf_renderer, skip_channels=False):
2723 ep = try_get(
2724 shelf_renderer, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
2725 compat_str)
2726 shelf_url = urljoin('https://www.youtube.com', ep)
2727 if shelf_url:
2728 # Skipping links to another channels, note that checking for
2729 # endpoint.commandMetadata.webCommandMetadata.webPageTypwebPageType == WEB_PAGE_TYPE_CHANNEL
2730 # will not work
2731 if skip_channels and '/channels?' in shelf_url:
2732 return
2733 title = try_get(
2734 shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str)
2735 yield self.url_result(shelf_url, video_title=title)
2736 # Shelf may not contain shelf URL, fallback to extraction from content
2737 for entry in self._shelf_entries_from_content(shelf_renderer):
2738 yield entry
2739
2740 def _playlist_entries(self, video_list_renderer):
2741 for content in video_list_renderer['contents']:
2742 if not isinstance(content, dict):
2743 continue
2744 renderer = content.get('playlistVideoRenderer') or content.get('playlistPanelVideoRenderer')
2745 if not isinstance(renderer, dict):
2746 continue
2747 video_id = renderer.get('videoId')
2748 if not video_id:
2749 continue
2750 yield self._extract_video(renderer)
2751
2752 def _rich_entries(self, rich_grid_renderer):
2753 renderer = try_get(
2754 rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {}
2755 video_id = renderer.get('videoId')
2756 if not video_id:
2757 return
2758 yield self._extract_video(renderer)
2759
2760 def _video_entry(self, video_renderer):
2761 video_id = video_renderer.get('videoId')
2762 if video_id:
2763 return self._extract_video(video_renderer)
2764
2765 def _post_thread_entries(self, post_thread_renderer):
2766 post_renderer = try_get(
2767 post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict)
2768 if not post_renderer:
2769 return
2770 # video attachment
2771 video_renderer = try_get(
2772 post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict)
2773 video_id = None
2774 if video_renderer:
2775 entry = self._video_entry(video_renderer)
2776 if entry:
2777 yield entry
2778 # inline video links
2779 runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or []
2780 for run in runs:
2781 if not isinstance(run, dict):
2782 continue
2783 ep_url = try_get(
2784 run, lambda x: x['navigationEndpoint']['urlEndpoint']['url'], compat_str)
2785 if not ep_url:
2786 continue
2787 if not YoutubeIE.suitable(ep_url):
2788 continue
2789 ep_video_id = YoutubeIE._match_id(ep_url)
2790 if video_id == ep_video_id:
2791 continue
2792 yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id)
2793
2794 def _post_thread_continuation_entries(self, post_thread_continuation):
2795 contents = post_thread_continuation.get('contents')
2796 if not isinstance(contents, list):
2797 return
2798 for content in contents:
2799 renderer = content.get('backstagePostThreadRenderer')
2800 if not isinstance(renderer, dict):
2801 continue
2802 for entry in self._post_thread_entries(renderer):
2803 yield entry
2804
2805 @staticmethod
2806 def _build_continuation_query(continuation, ctp=None):
2807 query = {
2808 'ctoken': continuation,
2809 'continuation': continuation,
2810 }
2811 if ctp:
2812 query['itct'] = ctp
2813 return query
2814
2815 @staticmethod
2816 def _extract_next_continuation_data(renderer):
2817 next_continuation = try_get(
2818 renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict)
2819 if not next_continuation:
2820 return
2821 continuation = next_continuation.get('continuation')
2822 if not continuation:
2823 return
2824 ctp = next_continuation.get('clickTrackingParams')
2825 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2826
2827 @classmethod
2828 def _extract_continuation(cls, renderer):
2829 next_continuation = cls._extract_next_continuation_data(renderer)
2830 if next_continuation:
2831 return next_continuation
2832 contents = []
2833 for key in ('contents', 'items'):
2834 contents.extend(try_get(renderer, lambda x: x[key], list) or [])
2835 for content in contents:
2836 if not isinstance(content, dict):
2837 continue
2838 continuation_ep = try_get(
2839 content, lambda x: x['continuationItemRenderer']['continuationEndpoint'],
2840 dict)
2841 if not continuation_ep:
2842 continue
2843 continuation = try_get(
2844 continuation_ep, lambda x: x['continuationCommand']['token'], compat_str)
2845 if not continuation:
2846 continue
2847 ctp = continuation_ep.get('clickTrackingParams')
2848 return YoutubeTabIE._build_continuation_query(continuation, ctp)
2849
2850 def _entries(self, tab, item_id, identity_token, account_syncid):
2851
2852 def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds
2853 contents = try_get(parent_renderer, lambda x: x['contents'], list) or []
2854 for content in contents:
2855 if not isinstance(content, dict):
2856 continue
2857 is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict)
2858 if not is_renderer:
2859 renderer = content.get('richItemRenderer')
2860 if renderer:
2861 for entry in self._rich_entries(renderer):
2862 yield entry
2863 continuation_list[0] = self._extract_continuation(parent_renderer)
2864 continue
2865 isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or []
2866 for isr_content in isr_contents:
2867 if not isinstance(isr_content, dict):
2868 continue
2869
2870 known_renderers = {
2871 'playlistVideoListRenderer': self._playlist_entries,
2872 'gridRenderer': self._grid_entries,
2873 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'),
2874 'backstagePostThreadRenderer': self._post_thread_entries,
2875 'videoRenderer': lambda x: [self._video_entry(x)],
2876 }
2877 for key, renderer in isr_content.items():
2878 if key not in known_renderers:
2879 continue
2880 for entry in known_renderers[key](renderer):
2881 if entry:
2882 yield entry
2883 continuation_list[0] = self._extract_continuation(renderer)
2884 break
2885
2886 if not continuation_list[0]:
2887 continuation_list[0] = self._extract_continuation(is_renderer)
2888
2889 if not continuation_list[0]:
2890 continuation_list[0] = self._extract_continuation(parent_renderer)
2891
2892 continuation_list = [None] # Python 2 doesnot support nonlocal
2893 tab_content = try_get(tab, lambda x: x['content'], dict)
2894 if not tab_content:
2895 return
2896 parent_renderer = (
2897 try_get(tab_content, lambda x: x['sectionListRenderer'], dict)
2898 or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {})
2899 for entry in extract_entries(parent_renderer):
2900 yield entry
2901 continuation = continuation_list[0]
2902
2903 headers = {
2904 'x-youtube-client-name': '1',
2905 'x-youtube-client-version': '2.20201112.04.01',
2906 }
2907 if identity_token:
2908 headers['x-youtube-identity-token'] = identity_token
2909
2910 if account_syncid:
2911 headers['X-Goog-PageId'] = account_syncid
2912 headers['X-Goog-AuthUser'] = 0
2913
2914 for page_num in itertools.count(1):
2915 if not continuation:
2916 break
2917 retries = self._downloader.params.get('extractor_retries', 3)
2918 count = -1
2919 last_error = None
2920 while count < retries:
2921 count += 1
2922 if last_error:
2923 self.report_warning('%s. Retrying ...' % last_error)
2924 try:
2925 response = self._call_api(
2926 ep="browse", fatal=True, headers=headers,
2927 video_id='%s page %s' % (item_id, page_num),
2928 query={
2929 'continuation': continuation['continuation'],
2930 'clickTracking': {'clickTrackingParams': continuation['itct']},
2931 },
2932 note='Downloading API JSON%s' % (' (retry #%d)' % count if count else ''))
2933 except ExtractorError as e:
2934 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503, 404):
2935 # Downloading page may result in intermittent 5xx HTTP error
2936 # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289
2937 last_error = 'HTTP Error %s' % e.cause.code
2938 if count < retries:
2939 continue
2940 raise
2941 else:
2942 # Youtube sometimes sends incomplete data
2943 # See: https://github.com/ytdl-org/youtube-dl/issues/28194
2944 if dict_get(response,
2945 ('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')):
2946 break
2947
2948 # Youtube may send alerts if there was an issue with the continuation page
2949 self._extract_alerts(response, expected=False)
2950
2951 last_error = 'Incomplete data received'
2952 if count >= retries:
2953 self._downloader.report_error(last_error)
2954
2955 if not response:
2956 break
2957
2958 known_continuation_renderers = {
2959 'playlistVideoListContinuation': self._playlist_entries,
2960 'gridContinuation': self._grid_entries,
2961 'itemSectionContinuation': self._post_thread_continuation_entries,
2962 'sectionListContinuation': extract_entries, # for feeds
2963 }
2964 continuation_contents = try_get(
2965 response, lambda x: x['continuationContents'], dict) or {}
2966 continuation_renderer = None
2967 for key, value in continuation_contents.items():
2968 if key not in known_continuation_renderers:
2969 continue
2970 continuation_renderer = value
2971 continuation_list = [None]
2972 for entry in known_continuation_renderers[key](continuation_renderer):
2973 yield entry
2974 continuation = continuation_list[0] or self._extract_continuation(continuation_renderer)
2975 break
2976 if continuation_renderer:
2977 continue
2978
2979 known_renderers = {
2980 'gridPlaylistRenderer': (self._grid_entries, 'items'),
2981 'gridVideoRenderer': (self._grid_entries, 'items'),
2982 'playlistVideoRenderer': (self._playlist_entries, 'contents'),
2983 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds
2984 'richItemRenderer': (extract_entries, 'contents'), # for hashtag
2985 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents')
2986 }
2987 continuation_items = try_get(
2988 response,
2989 lambda x: dict_get(x, ('onResponseReceivedActions', 'onResponseReceivedEndpoints'))[0]['appendContinuationItemsAction']['continuationItems'], list)
2990 continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {}
2991 video_items_renderer = None
2992 for key, value in continuation_item.items():
2993 if key not in known_renderers:
2994 continue
2995 video_items_renderer = {known_renderers[key][1]: continuation_items}
2996 continuation_list = [None]
2997 for entry in known_renderers[key][0](video_items_renderer):
2998 yield entry
2999 continuation = continuation_list[0] or self._extract_continuation(video_items_renderer)
3000 break
3001 if video_items_renderer:
3002 continue
3003 break
3004
3005 @staticmethod
3006 def _extract_selected_tab(tabs):
3007 for tab in tabs:
3008 if try_get(tab, lambda x: x['tabRenderer']['selected'], bool):
3009 return tab['tabRenderer']
3010 else:
3011 raise ExtractorError('Unable to find selected tab')
3012
3013 @staticmethod
3014 def _extract_uploader(data):
3015 uploader = {}
3016 sidebar_renderer = try_get(
3017 data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list)
3018 if sidebar_renderer:
3019 for item in sidebar_renderer:
3020 if not isinstance(item, dict):
3021 continue
3022 renderer = item.get('playlistSidebarSecondaryInfoRenderer')
3023 if not isinstance(renderer, dict):
3024 continue
3025 owner = try_get(
3026 renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict)
3027 if owner:
3028 uploader['uploader'] = owner.get('text')
3029 uploader['uploader_id'] = try_get(
3030 owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str)
3031 uploader['uploader_url'] = urljoin(
3032 'https://www.youtube.com/',
3033 try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str))
3034 return {k: v for k, v in uploader.items() if v is not None}
3035
3036 def _extract_from_tabs(self, item_id, webpage, data, tabs):
3037 playlist_id = title = description = channel_url = channel_name = channel_id = None
3038 thumbnails_list = tags = []
3039
3040 selected_tab = self._extract_selected_tab(tabs)
3041 renderer = try_get(
3042 data, lambda x: x['metadata']['channelMetadataRenderer'], dict)
3043 if renderer:
3044 channel_name = renderer.get('title')
3045 channel_url = renderer.get('channelUrl')
3046 channel_id = renderer.get('externalId')
3047
3048 if not renderer:
3049 renderer = try_get(
3050 data, lambda x: x['metadata']['playlistMetadataRenderer'], dict)
3051 if renderer:
3052 title = renderer.get('title')
3053 description = renderer.get('description', '')
3054 playlist_id = channel_id
3055 tags = renderer.get('keywords', '').split()
3056 thumbnails_list = (
3057 try_get(renderer, lambda x: x['avatar']['thumbnails'], list)
3058 or try_get(
3059 data,
3060 lambda x: x['sidebar']['playlistSidebarRenderer']['items'][0]['playlistSidebarPrimaryInfoRenderer']['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'],
3061 list)
3062 or [])
3063
3064 thumbnails = []
3065 for t in thumbnails_list:
3066 if not isinstance(t, dict):
3067 continue
3068 thumbnail_url = url_or_none(t.get('url'))
3069 if not thumbnail_url:
3070 continue
3071 thumbnails.append({
3072 'url': thumbnail_url,
3073 'width': int_or_none(t.get('width')),
3074 'height': int_or_none(t.get('height')),
3075 })
3076
3077 if playlist_id is None:
3078 playlist_id = item_id
3079 if title is None:
3080 title = playlist_id
3081 title += format_field(selected_tab, 'title', ' - %s')
3082
3083 metadata = {
3084 'playlist_id': playlist_id,
3085 'playlist_title': title,
3086 'playlist_description': description,
3087 'uploader': channel_name,
3088 'uploader_id': channel_id,
3089 'uploader_url': channel_url,
3090 'thumbnails': thumbnails,
3091 'tags': tags,
3092 }
3093 if not channel_id:
3094 metadata.update(self._extract_uploader(data))
3095 metadata.update({
3096 'channel': metadata['uploader'],
3097 'channel_id': metadata['uploader_id'],
3098 'channel_url': metadata['uploader_url']})
3099 return self.playlist_result(
3100 self._entries(
3101 selected_tab, playlist_id,
3102 self._extract_identity_token(webpage, item_id),
3103 self._extract_account_syncid(data)),
3104 **metadata)
3105
3106 def _extract_mix_playlist(self, playlist, playlist_id):
3107 first_id = last_id = None
3108 for page_num in itertools.count(1):
3109 videos = list(self._playlist_entries(playlist))
3110 if not videos:
3111 return
3112 start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1
3113 if start >= len(videos):
3114 return
3115 for video in videos[start:]:
3116 if video['id'] == first_id:
3117 self.to_screen('First video %s found again; Assuming end of Mix' % first_id)
3118 return
3119 yield video
3120 first_id = first_id or videos[0]['id']
3121 last_id = videos[-1]['id']
3122
3123 _, data = self._extract_webpage(
3124 'https://www.youtube.com/watch?list=%s&v=%s' % (playlist_id, last_id),
3125 '%s page %d' % (playlist_id, page_num))
3126 playlist = try_get(
3127 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3128
3129 def _extract_from_playlist(self, item_id, url, data, playlist):
3130 title = playlist.get('title') or try_get(
3131 data, lambda x: x['titleText']['simpleText'], compat_str)
3132 playlist_id = playlist.get('playlistId') or item_id
3133
3134 # Delegating everything except mix playlists to regular tab-based playlist URL
3135 playlist_url = urljoin(url, try_get(
3136 playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'],
3137 compat_str))
3138 if playlist_url and playlist_url != url:
3139 return self.url_result(
3140 playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id,
3141 video_title=title)
3142
3143 return self.playlist_result(
3144 self._extract_mix_playlist(playlist, playlist_id),
3145 playlist_id=playlist_id, playlist_title=title)
3146
3147 def _extract_alerts(self, data, expected=False):
3148
3149 def _real_extract_alerts():
3150 for alert_dict in try_get(data, lambda x: x['alerts'], list) or []:
3151 if not isinstance(alert_dict, dict):
3152 continue
3153 for alert in alert_dict.values():
3154 alert_type = alert.get('type')
3155 if not alert_type:
3156 continue
3157 message = try_get(alert, lambda x: x['text']['simpleText'], compat_str)
3158 if message:
3159 yield alert_type, message
3160 for run in try_get(alert, lambda x: x['text']['runs'], list) or []:
3161 message = try_get(run, lambda x: x['text'], compat_str)
3162 if message:
3163 yield alert_type, message
3164
3165 err_msg = None
3166 for alert_type, alert_message in _real_extract_alerts():
3167 if alert_type.lower() == 'error':
3168 if err_msg:
3169 self._downloader.report_warning('YouTube said: %s - %s' % ('ERROR', err_msg))
3170 err_msg = alert_message
3171 else:
3172 self._downloader.report_warning('YouTube said: %s - %s' % (alert_type, alert_message))
3173
3174 if err_msg:
3175 raise ExtractorError('YouTube said: %s' % err_msg, expected=expected)
3176
3177 def _extract_webpage(self, url, item_id):
3178 retries = self._downloader.params.get('extractor_retries', 3)
3179 count = -1
3180 last_error = 'Incomplete yt initial data recieved'
3181 while count < retries:
3182 count += 1
3183 # Sometimes youtube returns a webpage with incomplete ytInitialData
3184 # See: https://github.com/yt-dlp/yt-dlp/issues/116
3185 if count:
3186 self.report_warning('%s. Retrying ...' % last_error)
3187 webpage = self._download_webpage(
3188 url, item_id,
3189 'Downloading webpage%s' % (' (retry #%d)' % count if count else ''))
3190 data = self._extract_yt_initial_data(item_id, webpage)
3191 self._extract_alerts(data, expected=True)
3192 if data.get('contents') or data.get('currentVideoEndpoint'):
3193 break
3194 if count >= retries:
3195 self._downloader.report_error(last_error)
3196 return webpage, data
3197
3198 def _real_extract(self, url):
3199 item_id = self._match_id(url)
3200 url = compat_urlparse.urlunparse(
3201 compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com'))
3202
3203 # This is not matched in a channel page with a tab selected
3204 mobj = re.match(r'(?P<pre>%s)(?P<post>/?(?![^#?]).*$)' % self._VALID_URL, url)
3205 mobj = mobj.groupdict() if mobj else {}
3206 if mobj and not mobj.get('not_channel'):
3207 self._downloader.report_warning(
3208 'A channel/user page was given. All the channel\'s videos will be downloaded. '
3209 'To download only the videos in the home page, add a "/featured" to the URL')
3210 url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '')
3211
3212 # Handle both video/playlist URLs
3213 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3214 video_id = qs.get('v', [None])[0]
3215 playlist_id = qs.get('list', [None])[0]
3216
3217 if not video_id and (mobj.get('not_channel') or '').startswith('watch'):
3218 if not playlist_id:
3219 # If there is neither video or playlist ids,
3220 # youtube redirects to home page, which is undesirable
3221 raise ExtractorError('Unable to recognize tab page')
3222 self._downloader.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id)
3223 url = 'https://www.youtube.com/playlist?list=%s' % playlist_id
3224
3225 if video_id and playlist_id:
3226 if self._downloader.params.get('noplaylist'):
3227 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
3228 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3229 self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id))
3230
3231 webpage, data = self._extract_webpage(url, item_id)
3232
3233 tabs = try_get(
3234 data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list)
3235 if tabs:
3236 return self._extract_from_tabs(item_id, webpage, data, tabs)
3237
3238 playlist = try_get(
3239 data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict)
3240 if playlist:
3241 return self._extract_from_playlist(item_id, url, data, playlist)
3242
3243 video_id = try_get(
3244 data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'],
3245 compat_str) or video_id
3246 if video_id:
3247 self._downloader.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id)
3248 return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id)
3249
3250 raise ExtractorError('Unable to recognize tab page')
3251
3252
3253 class YoutubePlaylistIE(InfoExtractor):
3254 IE_DESC = 'YouTube.com playlists'
3255 _VALID_URL = r'''(?x)(?:
3256 (?:https?://)?
3257 (?:\w+\.)?
3258 (?:
3259 (?:
3260 youtube(?:kids)?\.com|
3261 invidio\.us
3262 )
3263 /.*?\?.*?\blist=
3264 )?
3265 (?P<id>%(playlist_id)s)
3266 )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3267 IE_NAME = 'youtube:playlist'
3268 _TESTS = [{
3269 'note': 'issue #673',
3270 'url': 'PLBB231211A4F62143',
3271 'info_dict': {
3272 'title': '[OLD]Team Fortress 2 (Class-based LP)',
3273 'id': 'PLBB231211A4F62143',
3274 'uploader': 'Wickydoo',
3275 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q',
3276 },
3277 'playlist_mincount': 29,
3278 }, {
3279 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3280 'info_dict': {
3281 'title': 'YDL_safe_search',
3282 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
3283 },
3284 'playlist_count': 2,
3285 'skip': 'This playlist is private',
3286 }, {
3287 'note': 'embedded',
3288 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3289 'playlist_count': 4,
3290 'info_dict': {
3291 'title': 'JODA15',
3292 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
3293 'uploader': 'milan',
3294 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
3295 }
3296 }, {
3297 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3298 'playlist_mincount': 982,
3299 'info_dict': {
3300 'title': '2018 Chinese New Singles (11/6 updated)',
3301 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
3302 'uploader': 'LBK',
3303 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA',
3304 }
3305 }, {
3306 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
3307 'only_matching': True,
3308 }, {
3309 # music album playlist
3310 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
3311 'only_matching': True,
3312 }]
3313
3314 @classmethod
3315 def suitable(cls, url):
3316 return False if YoutubeTabIE.suitable(url) else super(
3317 YoutubePlaylistIE, cls).suitable(url)
3318
3319 def _real_extract(self, url):
3320 playlist_id = self._match_id(url)
3321 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
3322 if not qs:
3323 qs = {'list': playlist_id}
3324 return self.url_result(
3325 update_url_query('https://www.youtube.com/playlist', qs),
3326 ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3327
3328
3329 class YoutubeYtBeIE(InfoExtractor):
3330 IE_DESC = 'youtu.be'
3331 _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
3332 _TESTS = [{
3333 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
3334 'info_dict': {
3335 'id': 'yeWKywCrFtk',
3336 'ext': 'mp4',
3337 'title': 'Small Scale Baler and Braiding Rugs',
3338 'uploader': 'Backus-Page House Museum',
3339 'uploader_id': 'backuspagemuseum',
3340 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
3341 'upload_date': '20161008',
3342 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
3343 'categories': ['Nonprofits & Activism'],
3344 'tags': list,
3345 'like_count': int,
3346 'dislike_count': int,
3347 },
3348 'params': {
3349 'noplaylist': True,
3350 'skip_download': True,
3351 },
3352 }, {
3353 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
3354 'only_matching': True,
3355 }]
3356
3357 def _real_extract(self, url):
3358 mobj = re.match(self._VALID_URL, url)
3359 video_id = mobj.group('id')
3360 playlist_id = mobj.group('playlist_id')
3361 return self.url_result(
3362 update_url_query('https://www.youtube.com/watch', {
3363 'v': video_id,
3364 'list': playlist_id,
3365 'feature': 'youtu.be',
3366 }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id)
3367
3368
3369 class YoutubeYtUserIE(InfoExtractor):
3370 IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword'
3371 _VALID_URL = r'ytuser:(?P<id>.+)'
3372 _TESTS = [{
3373 'url': 'ytuser:phihag',
3374 'only_matching': True,
3375 }]
3376
3377 def _real_extract(self, url):
3378 user_id = self._match_id(url)
3379 return self.url_result(
3380 'https://www.youtube.com/user/%s' % user_id,
3381 ie=YoutubeTabIE.ie_key(), video_id=user_id)
3382
3383
3384 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3385 IE_NAME = 'youtube:favorites'
3386 IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)'
3387 _VALID_URL = r':ytfav(?:ou?rite)?s?'
3388 _LOGIN_REQUIRED = True
3389 _TESTS = [{
3390 'url': ':ytfav',
3391 'only_matching': True,
3392 }, {
3393 'url': ':ytfavorites',
3394 'only_matching': True,
3395 }]
3396
3397 def _real_extract(self, url):
3398 return self.url_result(
3399 'https://www.youtube.com/playlist?list=LL',
3400 ie=YoutubeTabIE.ie_key())
3401
3402
3403 class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor):
3404 IE_DESC = 'YouTube.com searches, "ytsearch" keyword'
3405 # there doesn't appear to be a real limit, for example if you search for
3406 # 'python' you get more than 8.000.000 results
3407 _MAX_RESULTS = float('inf')
3408 IE_NAME = 'youtube:search'
3409 _SEARCH_KEY = 'ytsearch'
3410 _SEARCH_PARAMS = None
3411 _TESTS = []
3412
3413 def _entries(self, query, n):
3414 data = {'query': query}
3415 if self._SEARCH_PARAMS:
3416 data['params'] = self._SEARCH_PARAMS
3417 total = 0
3418 for page_num in itertools.count(1):
3419 search = self._call_api(
3420 ep='search', video_id='query "%s"' % query, fatal=False,
3421 note='Downloading page %s' % page_num, query=data)
3422 if not search:
3423 break
3424 slr_contents = try_get(
3425 search,
3426 (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'],
3427 lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']),
3428 list)
3429 if not slr_contents:
3430 break
3431
3432 # Youtube sometimes adds promoted content to searches,
3433 # changing the index location of videos and token.
3434 # So we search through all entries till we find them.
3435 continuation_token = None
3436 for slr_content in slr_contents:
3437 if continuation_token is None:
3438 continuation_token = try_get(
3439 slr_content,
3440 lambda x: x['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'],
3441 compat_str)
3442
3443 isr_contents = try_get(
3444 slr_content,
3445 lambda x: x['itemSectionRenderer']['contents'],
3446 list)
3447 if not isr_contents:
3448 continue
3449 for content in isr_contents:
3450 if not isinstance(content, dict):
3451 continue
3452 video = content.get('videoRenderer')
3453 if not isinstance(video, dict):
3454 continue
3455 video_id = video.get('videoId')
3456 if not video_id:
3457 continue
3458
3459 yield self._extract_video(video)
3460 total += 1
3461 if total == n:
3462 return
3463
3464 if not continuation_token:
3465 break
3466 data['continuation'] = continuation_token
3467
3468 def _get_n_results(self, query, n):
3469 """Get a specified number of results for a query"""
3470 return self.playlist_result(self._entries(query, n), query)
3471
3472
3473 class YoutubeSearchDateIE(YoutubeSearchIE):
3474 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3475 _SEARCH_KEY = 'ytsearchdate'
3476 IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword'
3477 _SEARCH_PARAMS = 'CAI%3D'
3478
3479
3480 class YoutubeSearchURLIE(YoutubeSearchIE):
3481 IE_DESC = 'YouTube.com search URLs'
3482 IE_NAME = YoutubeSearchIE.IE_NAME + '_url'
3483 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)'
3484 # _MAX_RESULTS = 100
3485 _TESTS = [{
3486 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3487 'playlist_mincount': 5,
3488 'info_dict': {
3489 'title': 'youtube-dl test video',
3490 }
3491 }, {
3492 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3493 'only_matching': True,
3494 }]
3495
3496 @classmethod
3497 def _make_valid_url(cls):
3498 return cls._VALID_URL
3499
3500 def _real_extract(self, url):
3501 qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
3502 query = (qs.get('search_query') or qs.get('q'))[0]
3503 self._SEARCH_PARAMS = qs.get('sp', ('',))[0]
3504 return self._get_n_results(query, self._MAX_RESULTS)
3505
3506
3507 class YoutubeFeedsInfoExtractor(YoutubeTabIE):
3508 """
3509 Base class for feed extractors
3510 Subclasses must define the _FEED_NAME property.
3511 """
3512 _LOGIN_REQUIRED = True
3513 _TESTS = []
3514
3515 @property
3516 def IE_NAME(self):
3517 return 'youtube:%s' % self._FEED_NAME
3518
3519 def _real_initialize(self):
3520 self._login()
3521
3522 def _real_extract(self, url):
3523 return self.url_result(
3524 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3525 ie=YoutubeTabIE.ie_key())
3526
3527
3528 class YoutubeWatchLaterIE(InfoExtractor):
3529 IE_NAME = 'youtube:watchlater'
3530 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3531 _VALID_URL = r':ytwatchlater'
3532 _TESTS = [{
3533 'url': ':ytwatchlater',
3534 'only_matching': True,
3535 }]
3536
3537 def _real_extract(self, url):
3538 return self.url_result(
3539 'https://www.youtube.com/playlist?list=WL', ie=YoutubeTabIE.ie_key())
3540
3541
3542 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3543 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3544 _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?'
3545 _FEED_NAME = 'recommended'
3546 _TESTS = [{
3547 'url': ':ytrec',
3548 'only_matching': True,
3549 }, {
3550 'url': ':ytrecommended',
3551 'only_matching': True,
3552 }, {
3553 'url': 'https://youtube.com',
3554 'only_matching': True,
3555 }]
3556
3557
3558 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3559 IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)'
3560 _VALID_URL = r':ytsub(?:scription)?s?'
3561 _FEED_NAME = 'subscriptions'
3562 _TESTS = [{
3563 'url': ':ytsubs',
3564 'only_matching': True,
3565 }, {
3566 'url': ':ytsubscriptions',
3567 'only_matching': True,
3568 }]
3569
3570
3571 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3572 IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)'
3573 _VALID_URL = r':ythis(?:tory)?'
3574 _FEED_NAME = 'history'
3575 _TESTS = [{
3576 'url': ':ythistory',
3577 'only_matching': True,
3578 }]
3579
3580
3581 class YoutubeTruncatedURLIE(InfoExtractor):
3582 IE_NAME = 'youtube:truncated_url'
3583 IE_DESC = False # Do not list
3584 _VALID_URL = r'''(?x)
3585 (?:https?://)?
3586 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3587 (?:watch\?(?:
3588 feature=[a-z_]+|
3589 annotation_id=annotation_[^&]+|
3590 x-yt-cl=[0-9]+|
3591 hl=[^&]*|
3592 t=[0-9]+
3593 )?
3594 |
3595 attribution_link\?a=[^&]+
3596 )
3597 $
3598 '''
3599
3600 _TESTS = [{
3601 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3602 'only_matching': True,
3603 }, {
3604 'url': 'https://www.youtube.com/watch?',
3605 'only_matching': True,
3606 }, {
3607 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3608 'only_matching': True,
3609 }, {
3610 'url': 'https://www.youtube.com/watch?feature=foo',
3611 'only_matching': True,
3612 }, {
3613 'url': 'https://www.youtube.com/watch?hl=en-GB',
3614 'only_matching': True,
3615 }, {
3616 'url': 'https://www.youtube.com/watch?t=2372',
3617 'only_matching': True,
3618 }]
3619
3620 def _real_extract(self, url):
3621 raise ExtractorError(
3622 'Did you forget to quote the URL? Remember that & is a meta '
3623 'character in most shells, so you want to put the URL in quotes, '
3624 'like youtube-dl '
3625 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3626 ' or simply youtube-dl BaW_jenozKc .',
3627 expected=True)
3628
3629
3630 class YoutubeTruncatedIDIE(InfoExtractor):
3631 IE_NAME = 'youtube:truncated_id'
3632 IE_DESC = False # Do not list
3633 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3634
3635 _TESTS = [{
3636 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3637 'only_matching': True,
3638 }]
3639
3640 def _real_extract(self, url):
3641 video_id = self._match_id(url)
3642 raise ExtractorError(
3643 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3644 expected=True)
fork of https://github.com/yt-dlp/yt-dlp/