]> jfr.im git - yt-dlp.git/blob - youtube_dlc/extractor/youtube.py
[youtube] only playable on yt and age gated
[yt-dlp.git] / youtube_dlc / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
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 ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_HTTPError,
20 compat_kwargs,
21 compat_parse_qs,
22 compat_urllib_parse_unquote,
23 compat_urllib_parse_unquote_plus,
24 compat_urllib_parse_urlencode,
25 compat_urllib_parse_urlparse,
26 compat_urlparse,
27 compat_str,
28 )
29 from ..utils import (
30 bool_or_none,
31 clean_html,
32 error_to_compat_str,
33 extract_attributes,
34 ExtractorError,
35 float_or_none,
36 get_element_by_attribute,
37 get_element_by_id,
38 int_or_none,
39 mimetype2ext,
40 orderedSet,
41 parse_codecs,
42 parse_duration,
43 remove_quotes,
44 remove_start,
45 smuggle_url,
46 str_or_none,
47 str_to_int,
48 try_get,
49 unescapeHTML,
50 unified_strdate,
51 unsmuggle_url,
52 uppercase_escape,
53 url_or_none,
54 urlencode_postdata,
55 )
56
57
58 class YoutubeBaseInfoExtractor(InfoExtractor):
59 """Provide base functions for Youtube extractors"""
60 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
61 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
62
63 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
64 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
65 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
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,}'
72
73 _YOUTUBE_CLIENT_HEADERS = {
74 'x-youtube-client-name': '1',
75 'x-youtube-client-version': '1.20200609.04.02',
76 }
77
78 def _set_language(self):
79 self._set_cookie(
80 '.youtube.com', 'PREF', 'f1=50000000&f6=8&hl=en',
81 # YouTube sets the expire time to about two months
82 expire_time=time.time() + 2 * 30 * 24 * 3600)
83
84 def _ids_to_results(self, ids):
85 return [
86 self.url_result(vid_id, 'Youtube', video_id=vid_id)
87 for vid_id in ids]
88
89 def _login(self):
90 """
91 Attempt to log in to YouTube.
92 True is returned if successful or skipped.
93 False is returned if login failed.
94
95 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
96 """
97 username, password = self._get_login_info()
98 # No authentication to be performed
99 if username is None:
100 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
101 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
102 return True
103
104 login_page = self._download_webpage(
105 self._LOGIN_URL, None,
106 note='Downloading login page',
107 errnote='unable to fetch login page', fatal=False)
108 if login_page is False:
109 return
110
111 login_form = self._hidden_inputs(login_page)
112
113 def req(url, f_req, note, errnote):
114 data = login_form.copy()
115 data.update({
116 'pstMsg': 1,
117 'checkConnection': 'youtube',
118 'checkedDomains': 'youtube',
119 'hl': 'en',
120 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
121 'f.req': json.dumps(f_req),
122 'flowName': 'GlifWebSignIn',
123 'flowEntry': 'ServiceLogin',
124 # TODO: reverse actual botguard identifier generation algo
125 'bgRequest': '["identifier",""]',
126 })
127 return self._download_json(
128 url, None, note=note, errnote=errnote,
129 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
130 fatal=False,
131 data=urlencode_postdata(data), headers={
132 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
133 'Google-Accounts-XSRF': 1,
134 })
135
136 def warn(message):
137 self._downloader.report_warning(message)
138
139 lookup_req = [
140 username,
141 None, [], None, 'US', None, None, 2, False, True,
142 [
143 None, None,
144 [2, 1, None, 1,
145 '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',
146 None, [], 4],
147 1, [None, None, []], None, None, None, True
148 ],
149 username,
150 ]
151
152 lookup_results = req(
153 self._LOOKUP_URL, lookup_req,
154 'Looking up account info', 'Unable to look up account info')
155
156 if lookup_results is False:
157 return False
158
159 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
160 if not user_hash:
161 warn('Unable to extract user hash')
162 return False
163
164 challenge_req = [
165 user_hash,
166 None, 1, None, [1, None, None, None, [password, None, True]],
167 [
168 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],
169 1, [None, None, []], None, None, None, True
170 ]]
171
172 challenge_results = req(
173 self._CHALLENGE_URL, challenge_req,
174 'Logging in', 'Unable to log in')
175
176 if challenge_results is False:
177 return
178
179 login_res = try_get(challenge_results, lambda x: x[0][5], list)
180 if login_res:
181 login_msg = try_get(login_res, lambda x: x[5], compat_str)
182 warn(
183 'Unable to login: %s' % 'Invalid password'
184 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
185 return False
186
187 res = try_get(challenge_results, lambda x: x[0][-1], list)
188 if not res:
189 warn('Unable to extract result entry')
190 return False
191
192 login_challenge = try_get(res, lambda x: x[0][0], list)
193 if login_challenge:
194 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
195 if challenge_str == 'TWO_STEP_VERIFICATION':
196 # SEND_SUCCESS - TFA code has been successfully sent to phone
197 # QUOTA_EXCEEDED - reached the limit of TFA codes
198 status = try_get(login_challenge, lambda x: x[5], compat_str)
199 if status == 'QUOTA_EXCEEDED':
200 warn('Exceeded the limit of TFA codes, try later')
201 return False
202
203 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
204 if not tl:
205 warn('Unable to extract TL')
206 return False
207
208 tfa_code = self._get_tfa_info('2-step verification code')
209
210 if not tfa_code:
211 warn(
212 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
213 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
214 return False
215
216 tfa_code = remove_start(tfa_code, 'G-')
217
218 tfa_req = [
219 user_hash, None, 2, None,
220 [
221 9, None, None, None, None, None, None, None,
222 [None, tfa_code, True, 2]
223 ]]
224
225 tfa_results = req(
226 self._TFA_URL.format(tl), tfa_req,
227 'Submitting TFA code', 'Unable to submit TFA code')
228
229 if tfa_results is False:
230 return False
231
232 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
233 if tfa_res:
234 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
235 warn(
236 'Unable to finish TFA: %s' % 'Invalid TFA code'
237 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
238 return False
239
240 check_cookie_url = try_get(
241 tfa_results, lambda x: x[0][-1][2], compat_str)
242 else:
243 CHALLENGES = {
244 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
245 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
246 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
247 }
248 challenge = CHALLENGES.get(
249 challenge_str,
250 '%s returned error %s.' % (self.IE_NAME, challenge_str))
251 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
252 return False
253 else:
254 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
255
256 if not check_cookie_url:
257 warn('Unable to extract CheckCookie URL')
258 return False
259
260 check_cookie_results = self._download_webpage(
261 check_cookie_url, None, 'Checking cookie', fatal=False)
262
263 if check_cookie_results is False:
264 return False
265
266 if 'https://myaccount.google.com/' not in check_cookie_results:
267 warn('Unable to log in')
268 return False
269
270 return True
271
272 def _download_webpage_handle(self, *args, **kwargs):
273 query = kwargs.get('query', {}).copy()
274 query['disable_polymer'] = 'true'
275 kwargs['query'] = query
276 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
277 *args, **compat_kwargs(kwargs))
278
279 def _real_initialize(self):
280 if self._downloader is None:
281 return
282 self._set_language()
283 if not self._login():
284 return
285
286
287 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
288 # Extract entries from page with "Load more" button
289 def _entries(self, page, playlist_id):
290 more_widget_html = content_html = page
291 for page_num in itertools.count(1):
292 for entry in self._process_page(content_html):
293 yield entry
294
295 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
296 if not mobj:
297 break
298
299 count = 0
300 retries = 3
301 while count <= retries:
302 try:
303 # Downloading page may result in intermittent 5xx HTTP error
304 # that is usually worked around with a retry
305 more = self._download_json(
306 'https://www.youtube.com/%s' % mobj.group('more'), playlist_id,
307 'Downloading page #%s%s'
308 % (page_num, ' (retry #%d)' % count if count else ''),
309 transform_source=uppercase_escape,
310 headers=self._YOUTUBE_CLIENT_HEADERS)
311 break
312 except ExtractorError as e:
313 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
314 count += 1
315 if count <= retries:
316 continue
317 raise
318
319 content_html = more['content_html']
320 if not content_html.strip():
321 # Some webpages show a "Load more" button but they don't
322 # have more videos
323 break
324 more_widget_html = more['load_more_widget_html']
325
326
327 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
328 def _process_page(self, content):
329 for video_id, video_title in self.extract_videos_from_page(content):
330 yield self.url_result(video_id, 'Youtube', video_id, video_title)
331
332 def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page):
333 for mobj in re.finditer(video_re, page):
334 # The link with index 0 is not the first video of the playlist (not sure if still actual)
335 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
336 continue
337 video_id = mobj.group('id')
338 video_title = unescapeHTML(
339 mobj.group('title')) if 'title' in mobj.groupdict() else None
340 if video_title:
341 video_title = video_title.strip()
342 if video_title == '► Play all':
343 video_title = None
344 try:
345 idx = ids_in_page.index(video_id)
346 if video_title and not titles_in_page[idx]:
347 titles_in_page[idx] = video_title
348 except ValueError:
349 ids_in_page.append(video_id)
350 titles_in_page.append(video_title)
351
352 def extract_videos_from_page(self, page):
353 ids_in_page = []
354 titles_in_page = []
355 self.extract_videos_from_page_impl(
356 self._VIDEO_RE, page, ids_in_page, titles_in_page)
357 return zip(ids_in_page, titles_in_page)
358
359
360 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
361 def _process_page(self, content):
362 for playlist_id in orderedSet(re.findall(
363 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
364 content)):
365 yield self.url_result(
366 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
367
368 def _real_extract(self, url):
369 playlist_id = self._match_id(url)
370 webpage = self._download_webpage(url, playlist_id)
371 title = self._og_search_title(webpage, fatal=False)
372 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
373
374
375 class YoutubeIE(YoutubeBaseInfoExtractor):
376 IE_DESC = 'YouTube.com'
377 _VALID_URL = r"""(?x)^
378 (
379 (?:https?://|//) # http(s):// or protocol-independent URL
380 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/|
381 (?:www\.)?deturl\.com/www\.youtube\.com/|
382 (?:www\.)?pwnyoutube\.com/|
383 (?:www\.)?hooktube\.com/|
384 (?:www\.)?yourepeat\.com/|
385 tube\.majestyc\.net/|
386 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
387 (?:(?:www|dev)\.)?invidio\.us/|
388 (?:(?:www|no)\.)?invidiou\.sh/|
389 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
390 (?:www\.)?invidious\.kabi\.tk/|
391 (?:www\.)?invidious\.13ad\.de/|
392 (?:www\.)?invidious\.mastodon\.host/|
393 (?:www\.)?invidious\.nixnet\.xyz/|
394 (?:www\.)?invidious\.drycat\.fr/|
395 (?:www\.)?tube\.poal\.co/|
396 (?:www\.)?vid\.wxzm\.sx/|
397 (?:www\.)?yewtu\.be/|
398 (?:www\.)?yt\.elukerio\.org/|
399 (?:www\.)?yt\.lelux\.fi/|
400 (?:www\.)?invidious\.ggc-project\.de/|
401 (?:www\.)?yt\.maisputain\.ovh/|
402 (?:www\.)?invidious\.13ad\.de/|
403 (?:www\.)?invidious\.toot\.koeln/|
404 (?:www\.)?invidious\.fdn\.fr/|
405 (?:www\.)?watch\.nettohikari\.com/|
406 (?:www\.)?kgg2m7yk5aybusll\.onion/|
407 (?:www\.)?qklhadlycap4cnod\.onion/|
408 (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/|
409 (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/|
410 (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/|
411 (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/|
412 (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/|
413 (?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion/|
414 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
415 (?:.*?\#/)? # handle anchor (#/) redirect urls
416 (?: # the various things that can precede the ID:
417 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
418 |(?: # or the v= param in all its forms
419 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
420 (?:\?|\#!?) # the params delimiter ? or # or #!
421 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
422 v=
423 )
424 ))
425 |(?:
426 youtu\.be| # just youtu.be/xxxx
427 vid\.plus| # or vid.plus/xxxx
428 zwearz\.com/watch| # or zwearz.com/watch/xxxx
429 )/
430 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
431 )
432 )? # all until now is optional -> you can pass the naked ID
433 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
434 (?!.*?\blist=
435 (?:
436 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
437 WL # WL are handled by the watch later IE
438 )
439 )
440 (?(1).+)? # if we found the ID, everything can follow
441 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
442 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
443 _PLAYER_INFO_RE = (
444 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.(?P<ext>[a-z]+)$',
445 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.(?P<ext>[a-z]+)$',
446 )
447 _formats = {
448 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
449 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
450 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
451 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
452 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
453 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
454 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
455 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
456 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
457 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
458 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
459 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
460 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
461 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
462 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
463 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
464 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
465 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
466
467
468 # 3D videos
469 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
470 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
471 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
472 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
473 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
474 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
475 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
476
477 # Apple HTTP Live Streaming
478 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
479 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
480 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
481 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
482 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
483 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
484 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
485 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
486
487 # DASH mp4 video
488 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
489 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
490 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
491 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
492 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
493 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
494 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
495 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
496 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
497 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
498 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
499 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
500
501 # Dash mp4 audio
502 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
503 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
504 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
505 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
506 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
507 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
508 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
509
510 # Dash webm
511 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
512 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
513 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
514 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
515 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
516 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
517 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
518 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
519 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
520 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
521 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
522 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
523 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
524 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
525 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
526 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
527 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
528 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
529 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
530 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
531 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
532 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
533
534 # Dash webm audio
535 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
536 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
537
538 # Dash webm audio with opus inside
539 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
540 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
541 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
542
543 # RTMP (unnamed)
544 '_rtmp': {'protocol': 'rtmp'},
545
546 # av01 video only formats sometimes served with "unknown" codecs
547 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
548 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
549 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
550 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
551 }
552 _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
553
554 _GEO_BYPASS = False
555
556 IE_NAME = 'youtube'
557 _TESTS = [
558 {
559 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
560 'info_dict': {
561 'id': 'BaW_jenozKc',
562 'ext': 'mp4',
563 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
564 'uploader': 'Philipp Hagemeister',
565 'uploader_id': 'phihag',
566 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
567 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
568 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
569 'upload_date': '20121002',
570 '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 .',
571 'categories': ['Science & Technology'],
572 'tags': ['youtube-dl'],
573 'duration': 10,
574 'view_count': int,
575 'like_count': int,
576 'dislike_count': int,
577 'start_time': 1,
578 'end_time': 9,
579 }
580 },
581 {
582 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
583 'note': 'Test generic use_cipher_signature video (#897)',
584 'info_dict': {
585 'id': 'UxxajLWwzqY',
586 'ext': 'mp4',
587 'upload_date': '20120506',
588 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
589 'alt_title': 'I Love It (feat. Charli XCX)',
590 'description': 'md5:19a2f98d9032b9311e686ed039564f63',
591 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
592 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
593 'iconic ep', 'iconic', 'love', 'it'],
594 'duration': 180,
595 'uploader': 'Icona Pop',
596 'uploader_id': 'IconaPop',
597 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
598 'creator': 'Icona Pop',
599 'track': 'I Love It (feat. Charli XCX)',
600 'artist': 'Icona Pop',
601 }
602 },
603 {
604 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
605 'note': 'Test VEVO video with age protection (#956)',
606 'info_dict': {
607 'id': '07FYdnEawAQ',
608 'ext': 'mp4',
609 'upload_date': '20130703',
610 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
611 'alt_title': 'Tunnel Vision',
612 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
613 'duration': 419,
614 'uploader': 'justintimberlakeVEVO',
615 'uploader_id': 'justintimberlakeVEVO',
616 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
617 'creator': 'Justin Timberlake',
618 'track': 'Tunnel Vision',
619 'artist': 'Justin Timberlake',
620 'age_limit': 18,
621 }
622 },
623 {
624 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
625 'note': 'Embed-only video (#1746)',
626 'info_dict': {
627 'id': 'yZIXLfi8CZQ',
628 'ext': 'mp4',
629 'upload_date': '20120608',
630 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
631 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
632 'uploader': 'SET India',
633 'uploader_id': 'setindia',
634 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
635 'age_limit': 18,
636 }
637 },
638 {
639 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
640 'note': 'Use the first video ID in the URL',
641 'info_dict': {
642 'id': 'BaW_jenozKc',
643 'ext': 'mp4',
644 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
645 'uploader': 'Philipp Hagemeister',
646 'uploader_id': 'phihag',
647 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
648 'upload_date': '20121002',
649 '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 .',
650 'categories': ['Science & Technology'],
651 'tags': ['youtube-dl'],
652 'duration': 10,
653 'view_count': int,
654 'like_count': int,
655 'dislike_count': int,
656 },
657 'params': {
658 'skip_download': True,
659 },
660 },
661 {
662 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
663 'note': '256k DASH audio (format 141) via DASH manifest',
664 'info_dict': {
665 'id': 'a9LDPn-MO4I',
666 'ext': 'm4a',
667 'upload_date': '20121002',
668 'uploader_id': '8KVIDEO',
669 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
670 'description': '',
671 'uploader': '8KVIDEO',
672 'title': 'UHDTV TEST 8K VIDEO.mp4'
673 },
674 'params': {
675 'youtube_include_dash_manifest': True,
676 'format': '141',
677 },
678 'skip': 'format 141 not served anymore',
679 },
680 # DASH manifest with encrypted signature
681 {
682 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
683 'info_dict': {
684 'id': 'IB3lcPjvWLA',
685 'ext': 'm4a',
686 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
687 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
688 'duration': 244,
689 'uploader': 'AfrojackVEVO',
690 'uploader_id': 'AfrojackVEVO',
691 'upload_date': '20131011',
692 },
693 'params': {
694 'youtube_include_dash_manifest': True,
695 'format': '141/bestaudio[ext=m4a]',
696 },
697 },
698 # JS player signature function name containing $
699 {
700 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
701 'info_dict': {
702 'id': 'nfWlot6h_JM',
703 'ext': 'm4a',
704 'title': 'Taylor Swift - Shake It Off',
705 'description': 'md5:307195cd21ff7fa352270fe884570ef0',
706 'duration': 242,
707 'uploader': 'TaylorSwiftVEVO',
708 'uploader_id': 'TaylorSwiftVEVO',
709 'upload_date': '20140818',
710 },
711 'params': {
712 'youtube_include_dash_manifest': True,
713 'format': '141/bestaudio[ext=m4a]',
714 },
715 },
716 # Controversy video
717 {
718 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
719 'info_dict': {
720 'id': 'T4XJQO3qol8',
721 'ext': 'mp4',
722 'duration': 219,
723 'upload_date': '20100909',
724 'uploader': 'Amazing Atheist',
725 'uploader_id': 'TheAmazingAtheist',
726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
727 'title': 'Burning Everyone\'s Koran',
728 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\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',
729 }
730 },
731 # Normal age-gate video (No vevo, embed allowed)
732 {
733 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
734 'info_dict': {
735 'id': 'HtVdAasjOgU',
736 'ext': 'mp4',
737 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
738 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
739 'duration': 142,
740 'uploader': 'The Witcher',
741 'uploader_id': 'WitcherGame',
742 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
743 'upload_date': '20140605',
744 'age_limit': 18,
745 },
746 },
747 # Age-gate video with encrypted signature
748 {
749 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
750 'info_dict': {
751 'id': '6kLq3WMV1nU',
752 'ext': 'mp4',
753 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
754 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
755 'duration': 246,
756 'uploader': 'LloydVEVO',
757 'uploader_id': 'LloydVEVO',
758 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
759 'upload_date': '20110629',
760 'age_limit': 18,
761 },
762 },
763 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
764 # YouTube Red ad is not captured for creator
765 {
766 'url': '__2ABJjxzNo',
767 'info_dict': {
768 'id': '__2ABJjxzNo',
769 'ext': 'mp4',
770 'duration': 266,
771 'upload_date': '20100430',
772 'uploader_id': 'deadmau5',
773 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
774 'creator': 'Dada Life, deadmau5',
775 'description': 'md5:12c56784b8032162bb936a5f76d55360',
776 'uploader': 'deadmau5',
777 'title': 'Deadmau5 - Some Chords (HD)',
778 'alt_title': 'This Machine Kills Some Chords',
779 },
780 'expected_warnings': [
781 'DASH manifest missing',
782 ]
783 },
784 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
785 {
786 'url': 'lqQg6PlCWgI',
787 'info_dict': {
788 'id': 'lqQg6PlCWgI',
789 'ext': 'mp4',
790 'duration': 6085,
791 'upload_date': '20150827',
792 'uploader_id': 'olympic',
793 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
794 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
795 'uploader': 'Olympic',
796 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
797 },
798 'params': {
799 'skip_download': 'requires avconv',
800 }
801 },
802 # Non-square pixels
803 {
804 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
805 'info_dict': {
806 'id': '_b-2C3KPAM0',
807 'ext': 'mp4',
808 'stretched_ratio': 16 / 9.,
809 'duration': 85,
810 'upload_date': '20110310',
811 'uploader_id': 'AllenMeow',
812 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
813 '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',
814 'uploader': '孫ᄋᄅ',
815 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
816 },
817 },
818 # url_encoded_fmt_stream_map is empty string
819 {
820 'url': 'qEJwOuvDf7I',
821 'info_dict': {
822 'id': 'qEJwOuvDf7I',
823 'ext': 'webm',
824 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
825 'description': '',
826 'upload_date': '20150404',
827 'uploader_id': 'spbelect',
828 'uploader': 'Наблюдатели Петербурга',
829 },
830 'params': {
831 'skip_download': 'requires avconv',
832 },
833 'skip': 'This live event has ended.',
834 },
835 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
836 {
837 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
838 'info_dict': {
839 'id': 'FIl7x6_3R5Y',
840 'ext': 'webm',
841 'title': 'md5:7b81415841e02ecd4313668cde88737a',
842 'description': 'md5:116377fd2963b81ec4ce64b542173306',
843 'duration': 220,
844 'upload_date': '20150625',
845 'uploader_id': 'dorappi2000',
846 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
847 'uploader': 'dorappi2000',
848 'formats': 'mincount:31',
849 },
850 'skip': 'not actual anymore',
851 },
852 # DASH manifest with segment_list
853 {
854 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
855 'md5': '8ce563a1d667b599d21064e982ab9e31',
856 'info_dict': {
857 'id': 'CsmdDsKjzN8',
858 'ext': 'mp4',
859 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
860 'uploader': 'Airtek',
861 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
862 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
863 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
864 },
865 'params': {
866 'youtube_include_dash_manifest': True,
867 'format': '135', # bestvideo
868 },
869 'skip': 'This live event has ended.',
870 },
871 {
872 # Multifeed videos (multiple cameras), URL is for Main Camera
873 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
874 'info_dict': {
875 'id': 'jqWvoWXjCVs',
876 'title': 'teamPGP: Rocket League Noob Stream',
877 'description': 'md5:dc7872fb300e143831327f1bae3af010',
878 },
879 'playlist': [{
880 'info_dict': {
881 'id': 'jqWvoWXjCVs',
882 'ext': 'mp4',
883 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
884 'description': 'md5:dc7872fb300e143831327f1bae3af010',
885 'duration': 7335,
886 'upload_date': '20150721',
887 'uploader': 'Beer Games Beer',
888 'uploader_id': 'beergamesbeer',
889 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
890 'license': 'Standard YouTube License',
891 },
892 }, {
893 'info_dict': {
894 'id': '6h8e8xoXJzg',
895 'ext': 'mp4',
896 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
897 'description': 'md5:dc7872fb300e143831327f1bae3af010',
898 'duration': 7337,
899 'upload_date': '20150721',
900 'uploader': 'Beer Games Beer',
901 'uploader_id': 'beergamesbeer',
902 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
903 'license': 'Standard YouTube License',
904 },
905 }, {
906 'info_dict': {
907 'id': 'PUOgX5z9xZw',
908 'ext': 'mp4',
909 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
910 'description': 'md5:dc7872fb300e143831327f1bae3af010',
911 'duration': 7337,
912 'upload_date': '20150721',
913 'uploader': 'Beer Games Beer',
914 'uploader_id': 'beergamesbeer',
915 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
916 'license': 'Standard YouTube License',
917 },
918 }, {
919 'info_dict': {
920 'id': 'teuwxikvS5k',
921 'ext': 'mp4',
922 'title': 'teamPGP: Rocket League Noob Stream (zim)',
923 'description': 'md5:dc7872fb300e143831327f1bae3af010',
924 'duration': 7334,
925 'upload_date': '20150721',
926 'uploader': 'Beer Games Beer',
927 'uploader_id': 'beergamesbeer',
928 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
929 'license': 'Standard YouTube License',
930 },
931 }],
932 'params': {
933 'skip_download': True,
934 },
935 'skip': 'This video is not available.',
936 },
937 {
938 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
939 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
940 'info_dict': {
941 'id': 'gVfLd0zydlo',
942 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
943 },
944 'playlist_count': 2,
945 'skip': 'Not multifeed anymore',
946 },
947 {
948 'url': 'https://vid.plus/FlRa-iH7PGw',
949 'only_matching': True,
950 },
951 {
952 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
953 'only_matching': True,
954 },
955 {
956 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
957 # Also tests cut-off URL expansion in video description (see
958 # https://github.com/ytdl-org/youtube-dl/issues/1892,
959 # https://github.com/ytdl-org/youtube-dl/issues/8164)
960 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
961 'info_dict': {
962 'id': 'lsguqyKfVQg',
963 'ext': 'mp4',
964 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
965 'alt_title': 'Dark Walk - Position Music',
966 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
967 'duration': 133,
968 'upload_date': '20151119',
969 'uploader_id': 'IronSoulElf',
970 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
971 'uploader': 'IronSoulElf',
972 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
973 'track': 'Dark Walk - Position Music',
974 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
975 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
976 },
977 'params': {
978 'skip_download': True,
979 },
980 },
981 {
982 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
983 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
984 'only_matching': True,
985 },
986 {
987 # Video with yt:stretch=17:0
988 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
989 'info_dict': {
990 'id': 'Q39EVAstoRM',
991 'ext': 'mp4',
992 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
993 'description': 'md5:ee18a25c350637c8faff806845bddee9',
994 'upload_date': '20151107',
995 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
996 'uploader': 'CH GAMER DROID',
997 },
998 'params': {
999 'skip_download': True,
1000 },
1001 'skip': 'This video does not exist.',
1002 },
1003 {
1004 # Video licensed under Creative Commons
1005 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1006 'info_dict': {
1007 'id': 'M4gD1WSo5mA',
1008 'ext': 'mp4',
1009 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1010 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1011 'duration': 721,
1012 'upload_date': '20150127',
1013 'uploader_id': 'BerkmanCenter',
1014 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1015 'uploader': 'The Berkman Klein Center for Internet & Society',
1016 'license': 'Creative Commons Attribution license (reuse allowed)',
1017 },
1018 'params': {
1019 'skip_download': True,
1020 },
1021 },
1022 {
1023 # Channel-like uploader_url
1024 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1025 'info_dict': {
1026 'id': 'eQcmzGIKrzg',
1027 'ext': 'mp4',
1028 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1029 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
1030 'duration': 4060,
1031 'upload_date': '20151119',
1032 'uploader': 'Bernie Sanders',
1033 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1034 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1035 'license': 'Creative Commons Attribution license (reuse allowed)',
1036 },
1037 'params': {
1038 'skip_download': True,
1039 },
1040 },
1041 {
1042 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1043 'only_matching': True,
1044 },
1045 {
1046 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1047 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1048 'only_matching': True,
1049 },
1050 {
1051 # Rental video preview
1052 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1053 'info_dict': {
1054 'id': 'uGpuVWrhIzE',
1055 'ext': 'mp4',
1056 'title': 'Piku - Trailer',
1057 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1058 'upload_date': '20150811',
1059 'uploader': 'FlixMatrix',
1060 'uploader_id': 'FlixMatrixKaravan',
1061 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1062 'license': 'Standard YouTube License',
1063 },
1064 'params': {
1065 'skip_download': True,
1066 },
1067 'skip': 'This video is not available.',
1068 },
1069 {
1070 # YouTube Red video with episode data
1071 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1072 'info_dict': {
1073 'id': 'iqKdEhx-dD4',
1074 'ext': 'mp4',
1075 'title': 'Isolation - Mind Field (Ep 1)',
1076 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1077 'duration': 2085,
1078 'upload_date': '20170118',
1079 'uploader': 'Vsauce',
1080 'uploader_id': 'Vsauce',
1081 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1082 'series': 'Mind Field',
1083 'season_number': 1,
1084 'episode_number': 1,
1085 },
1086 'params': {
1087 'skip_download': True,
1088 },
1089 'expected_warnings': [
1090 'Skipping DASH manifest',
1091 ],
1092 },
1093 {
1094 # The following content has been identified by the YouTube community
1095 # as inappropriate or offensive to some audiences.
1096 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1097 'info_dict': {
1098 'id': '6SJNVb0GnPI',
1099 'ext': 'mp4',
1100 'title': 'Race Differences in Intelligence',
1101 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1102 'duration': 965,
1103 'upload_date': '20140124',
1104 'uploader': 'New Century Foundation',
1105 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1106 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1107 },
1108 'params': {
1109 'skip_download': True,
1110 },
1111 },
1112 {
1113 # itag 212
1114 'url': '1t24XAntNCY',
1115 'only_matching': True,
1116 },
1117 {
1118 # geo restricted to JP
1119 'url': 'sJL6WA-aGkQ',
1120 'only_matching': True,
1121 },
1122 {
1123 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1124 'only_matching': True,
1125 },
1126 {
1127 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1128 'only_matching': True,
1129 },
1130 {
1131 # DRM protected
1132 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1133 'only_matching': True,
1134 },
1135 {
1136 # Video with unsupported adaptive stream type formats
1137 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1138 'info_dict': {
1139 'id': 'Z4Vy8R84T1U',
1140 'ext': 'mp4',
1141 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1142 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1143 'duration': 433,
1144 'upload_date': '20130923',
1145 'uploader': 'Amelia Putri Harwita',
1146 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1147 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1148 'formats': 'maxcount:10',
1149 },
1150 'params': {
1151 'skip_download': True,
1152 'youtube_include_dash_manifest': False,
1153 },
1154 'skip': 'not actual anymore',
1155 },
1156 {
1157 # Youtube Music Auto-generated description
1158 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1159 'info_dict': {
1160 'id': 'MgNrAu2pzNs',
1161 'ext': 'mp4',
1162 'title': 'Voyeur Girl',
1163 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1164 'upload_date': '20190312',
1165 'uploader': 'Stephen - Topic',
1166 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1167 'artist': 'Stephen',
1168 'track': 'Voyeur Girl',
1169 'album': 'it\'s too much love to know my dear',
1170 'release_date': '20190313',
1171 'release_year': 2019,
1172 },
1173 'params': {
1174 'skip_download': True,
1175 },
1176 },
1177 {
1178 # Youtube Music Auto-generated description
1179 # Retrieve 'artist' field from 'Artist:' in video description
1180 # when it is present on youtube music video
1181 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1182 'info_dict': {
1183 'id': 'k0jLE7tTwjY',
1184 'ext': 'mp4',
1185 'title': 'Latch Feat. Sam Smith',
1186 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1187 'upload_date': '20150110',
1188 'uploader': 'Various Artists - Topic',
1189 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1190 'artist': 'Disclosure',
1191 'track': 'Latch Feat. Sam Smith',
1192 'album': 'Latch Featuring Sam Smith',
1193 'release_date': '20121008',
1194 'release_year': 2012,
1195 },
1196 'params': {
1197 'skip_download': True,
1198 },
1199 },
1200 {
1201 # Youtube Music Auto-generated description
1202 # handle multiple artists on youtube music video
1203 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1204 'info_dict': {
1205 'id': '74qn0eJSjpA',
1206 'ext': 'mp4',
1207 'title': 'Eastside',
1208 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1209 'upload_date': '20180710',
1210 'uploader': 'Benny Blanco - Topic',
1211 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1212 'artist': 'benny blanco, Halsey, Khalid',
1213 'track': 'Eastside',
1214 'album': 'Eastside',
1215 'release_date': '20180713',
1216 'release_year': 2018,
1217 },
1218 'params': {
1219 'skip_download': True,
1220 },
1221 },
1222 {
1223 # Youtube Music Auto-generated description
1224 # handle youtube music video with release_year and no release_date
1225 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1226 'info_dict': {
1227 'id': '-hcAI0g-f5M',
1228 'ext': 'mp4',
1229 'title': 'Put It On Me',
1230 'description': 'md5:f6422397c07c4c907c6638e1fee380a5',
1231 'upload_date': '20180426',
1232 'uploader': 'Matt Maeson - Topic',
1233 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1234 'artist': 'Matt Maeson',
1235 'track': 'Put It On Me',
1236 'album': 'The Hearse',
1237 'release_date': None,
1238 'release_year': 2018,
1239 },
1240 'params': {
1241 'skip_download': True,
1242 },
1243 },
1244 {
1245 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1246 'only_matching': True,
1247 },
1248 {
1249 # invalid -> valid video id redirection
1250 'url': 'DJztXj2GPfl',
1251 'info_dict': {
1252 'id': 'DJztXj2GPfk',
1253 'ext': 'mp4',
1254 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1255 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1256 'upload_date': '20090125',
1257 'uploader': 'Prochorowka',
1258 'uploader_id': 'Prochorowka',
1259 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1260 'artist': 'Panjabi MC',
1261 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1262 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1263 },
1264 'params': {
1265 'skip_download': True,
1266 },
1267 },
1268 {
1269 # empty description results in an empty string
1270 'url': 'https://www.youtube.com/watch?v=x41yOUIvK2k',
1271 'info_dict': {
1272 'id': 'x41yOUIvK2k',
1273 'ext': 'mp4',
1274 'title': 'IMG 3456',
1275 'description': '',
1276 'upload_date': '20170613',
1277 'uploader_id': 'ElevageOrVert',
1278 'uploader': 'ElevageOrVert',
1279 },
1280 'params': {
1281 'skip_download': True,
1282 },
1283 },
1284 ]
1285
1286 def __init__(self, *args, **kwargs):
1287 super(YoutubeIE, self).__init__(*args, **kwargs)
1288 self._player_cache = {}
1289
1290 def report_video_info_webpage_download(self, video_id):
1291 """Report attempt to download video info webpage."""
1292 self.to_screen('%s: Downloading video info webpage' % video_id)
1293
1294 def report_information_extraction(self, video_id):
1295 """Report attempt to extract video information."""
1296 self.to_screen('%s: Extracting video information' % video_id)
1297
1298 def report_unavailable_format(self, video_id, format):
1299 """Report extracted video URL."""
1300 self.to_screen('%s: Format %s not available' % (video_id, format))
1301
1302 def report_rtmp_download(self):
1303 """Indicate the download will use the RTMP protocol."""
1304 self.to_screen('RTMP download detected')
1305
1306 def _signature_cache_id(self, example_sig):
1307 """ Return a string representation of a signature """
1308 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1309
1310 @classmethod
1311 def _extract_player_info(cls, player_url):
1312 for player_re in cls._PLAYER_INFO_RE:
1313 id_m = re.search(player_re, player_url)
1314 if id_m:
1315 break
1316 else:
1317 raise ExtractorError('Cannot identify player %r' % player_url)
1318 return id_m.group('ext'), id_m.group('id')
1319
1320 def _extract_signature_function(self, video_id, player_url, example_sig):
1321 player_type, player_id = self._extract_player_info(player_url)
1322
1323 # Read from filesystem cache
1324 func_id = '%s_%s_%s' % (
1325 player_type, player_id, self._signature_cache_id(example_sig))
1326 assert os.path.basename(func_id) == func_id
1327
1328 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1329 if cache_spec is not None:
1330 return lambda s: ''.join(s[i] for i in cache_spec)
1331
1332 download_note = (
1333 'Downloading player %s' % player_url
1334 if self._downloader.params.get('verbose') else
1335 'Downloading %s player %s' % (player_type, player_id)
1336 )
1337 if player_type == 'js':
1338 code = self._download_webpage(
1339 player_url, video_id,
1340 note=download_note,
1341 errnote='Download of %s failed' % player_url)
1342 res = self._parse_sig_js(code)
1343 elif player_type == 'swf':
1344 urlh = self._request_webpage(
1345 player_url, video_id,
1346 note=download_note,
1347 errnote='Download of %s failed' % player_url)
1348 code = urlh.read()
1349 res = self._parse_sig_swf(code)
1350 else:
1351 assert False, 'Invalid player type %r' % player_type
1352
1353 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1354 cache_res = res(test_string)
1355 cache_spec = [ord(c) for c in cache_res]
1356
1357 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1358 return res
1359
1360 def _print_sig_code(self, func, example_sig):
1361 def gen_sig_code(idxs):
1362 def _genslice(start, end, step):
1363 starts = '' if start == 0 else str(start)
1364 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1365 steps = '' if step == 1 else (':%d' % step)
1366 return 's[%s%s%s]' % (starts, ends, steps)
1367
1368 step = None
1369 # Quelch pyflakes warnings - start will be set when step is set
1370 start = '(Never used)'
1371 for i, prev in zip(idxs[1:], idxs[:-1]):
1372 if step is not None:
1373 if i - prev == step:
1374 continue
1375 yield _genslice(start, prev, step)
1376 step = None
1377 continue
1378 if i - prev in [-1, 1]:
1379 step = i - prev
1380 start = prev
1381 continue
1382 else:
1383 yield 's[%d]' % prev
1384 if step is None:
1385 yield 's[%d]' % i
1386 else:
1387 yield _genslice(start, i, step)
1388
1389 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1390 cache_res = func(test_string)
1391 cache_spec = [ord(c) for c in cache_res]
1392 expr_code = ' + '.join(gen_sig_code(cache_spec))
1393 signature_id_tuple = '(%s)' % (
1394 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1395 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1396 ' return %s\n') % (signature_id_tuple, expr_code)
1397 self.to_screen('Extracted signature function:\n' + code)
1398
1399 def _parse_sig_js(self, jscode):
1400 funcname = self._search_regex(
1401 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1402 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1403 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\(\s*""\s*\)',
1404 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1405 # Obsolete patterns
1406 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1407 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1408 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1409 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1410 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1411 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1412 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1413 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1414 jscode, 'Initial JS player signature function name', group='sig')
1415
1416 jsi = JSInterpreter(jscode)
1417 initial_function = jsi.extract_function(funcname)
1418 return lambda s: initial_function([s])
1419
1420 def _parse_sig_swf(self, file_contents):
1421 swfi = SWFInterpreter(file_contents)
1422 TARGET_CLASSNAME = 'SignatureDecipher'
1423 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1424 initial_function = swfi.extract_function(searched_class, 'decipher')
1425 return lambda s: initial_function([s])
1426
1427 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1428 """Turn the encrypted s field into a working signature"""
1429
1430 if player_url is None:
1431 raise ExtractorError('Cannot decrypt signature without player_url')
1432
1433 if player_url.startswith('//'):
1434 player_url = 'https:' + player_url
1435 elif not re.match(r'https?://', player_url):
1436 player_url = compat_urlparse.urljoin(
1437 'https://www.youtube.com', player_url)
1438 try:
1439 player_id = (player_url, self._signature_cache_id(s))
1440 if player_id not in self._player_cache:
1441 func = self._extract_signature_function(
1442 video_id, player_url, s
1443 )
1444 self._player_cache[player_id] = func
1445 func = self._player_cache[player_id]
1446 if self._downloader.params.get('youtube_print_sig_code'):
1447 self._print_sig_code(func, s)
1448 return func(s)
1449 except Exception as e:
1450 tb = traceback.format_exc()
1451 raise ExtractorError(
1452 'Signature extraction failed: ' + tb, cause=e)
1453
1454 def _get_subtitles(self, video_id, webpage, has_live_chat_replay):
1455 try:
1456 subs_doc = self._download_xml(
1457 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1458 video_id, note=False)
1459 except ExtractorError as err:
1460 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1461 return {}
1462
1463 sub_lang_list = {}
1464 for track in subs_doc.findall('track'):
1465 lang = track.attrib['lang_code']
1466 if lang in sub_lang_list:
1467 continue
1468 sub_formats = []
1469 for ext in self._SUBTITLE_FORMATS:
1470 params = compat_urllib_parse_urlencode({
1471 'lang': lang,
1472 'v': video_id,
1473 'fmt': ext,
1474 'name': track.attrib['name'].encode('utf-8'),
1475 })
1476 sub_formats.append({
1477 'url': 'https://www.youtube.com/api/timedtext?' + params,
1478 'ext': ext,
1479 })
1480 sub_lang_list[lang] = sub_formats
1481 if has_live_chat_replay:
1482 sub_lang_list['live_chat'] = [
1483 {
1484 'video_id': video_id,
1485 'ext': 'json',
1486 'protocol': 'youtube_live_chat_replay',
1487 },
1488 ]
1489 if not sub_lang_list:
1490 self._downloader.report_warning('video doesn\'t have subtitles')
1491 return {}
1492 return sub_lang_list
1493
1494 def _get_ytplayer_config(self, video_id, webpage):
1495 patterns = (
1496 # User data may contain arbitrary character sequences that may affect
1497 # JSON extraction with regex, e.g. when '};' is contained the second
1498 # regex won't capture the whole JSON. Yet working around by trying more
1499 # concrete regex first keeping in mind proper quoted string handling
1500 # to be implemented in future that will replace this workaround (see
1501 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1502 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1503 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1504 r';ytplayer\.config\s*=\s*({.+?});',
1505 )
1506 config = self._search_regex(
1507 patterns, webpage, 'ytplayer.config', default=None)
1508 if config:
1509 return self._parse_json(
1510 uppercase_escape(config), video_id, fatal=False)
1511
1512 def _get_yt_initial_data(self, video_id, webpage):
1513 config = self._search_regex(
1514 (r'window\["ytInitialData"\]\s*=\s*(.*?)(?<=});',
1515 r'var\s+ytInitialData\s*=\s*(.*?)(?<=});'),
1516 webpage, 'ytInitialData', default=None)
1517 if config:
1518 return self._parse_json(
1519 uppercase_escape(config), video_id, fatal=False)
1520
1521 def _get_automatic_captions(self, video_id, webpage):
1522 """We need the webpage for getting the captions url, pass it as an
1523 argument to speed up the process."""
1524 self.to_screen('%s: Looking for automatic captions' % video_id)
1525 player_config = self._get_ytplayer_config(video_id, webpage)
1526 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1527 if not player_config:
1528 self._downloader.report_warning(err_msg)
1529 return {}
1530 try:
1531 args = player_config['args']
1532 caption_url = args.get('ttsurl')
1533 if caption_url:
1534 timestamp = args['timestamp']
1535 # We get the available subtitles
1536 list_params = compat_urllib_parse_urlencode({
1537 'type': 'list',
1538 'tlangs': 1,
1539 'asrs': 1,
1540 })
1541 list_url = caption_url + '&' + list_params
1542 caption_list = self._download_xml(list_url, video_id)
1543 original_lang_node = caption_list.find('track')
1544 if original_lang_node is None:
1545 self._downloader.report_warning('Video doesn\'t have automatic captions')
1546 return {}
1547 original_lang = original_lang_node.attrib['lang_code']
1548 caption_kind = original_lang_node.attrib.get('kind', '')
1549
1550 sub_lang_list = {}
1551 for lang_node in caption_list.findall('target'):
1552 sub_lang = lang_node.attrib['lang_code']
1553 sub_formats = []
1554 for ext in self._SUBTITLE_FORMATS:
1555 params = compat_urllib_parse_urlencode({
1556 'lang': original_lang,
1557 'tlang': sub_lang,
1558 'fmt': ext,
1559 'ts': timestamp,
1560 'kind': caption_kind,
1561 })
1562 sub_formats.append({
1563 'url': caption_url + '&' + params,
1564 'ext': ext,
1565 })
1566 sub_lang_list[sub_lang] = sub_formats
1567 return sub_lang_list
1568
1569 def make_captions(sub_url, sub_langs):
1570 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1571 caption_qs = compat_parse_qs(parsed_sub_url.query)
1572 captions = {}
1573 for sub_lang in sub_langs:
1574 sub_formats = []
1575 for ext in self._SUBTITLE_FORMATS:
1576 caption_qs.update({
1577 'tlang': [sub_lang],
1578 'fmt': [ext],
1579 })
1580 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1581 query=compat_urllib_parse_urlencode(caption_qs, True)))
1582 sub_formats.append({
1583 'url': sub_url,
1584 'ext': ext,
1585 })
1586 captions[sub_lang] = sub_formats
1587 return captions
1588
1589 # New captions format as of 22.06.2017
1590 player_response = args.get('player_response')
1591 if player_response and isinstance(player_response, compat_str):
1592 player_response = self._parse_json(
1593 player_response, video_id, fatal=False)
1594 if player_response:
1595 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1596 caption_tracks = renderer['captionTracks']
1597 for caption_track in caption_tracks:
1598 if 'kind' not in caption_track:
1599 # not an automatic transcription
1600 continue
1601 base_url = caption_track['baseUrl']
1602 sub_lang_list = []
1603 for lang in renderer['translationLanguages']:
1604 lang_code = lang.get('languageCode')
1605 if lang_code:
1606 sub_lang_list.append(lang_code)
1607 return make_captions(base_url, sub_lang_list)
1608
1609 self._downloader.report_warning("Couldn't find automatic captions for %s" % video_id)
1610 return {}
1611 # Some videos don't provide ttsurl but rather caption_tracks and
1612 # caption_translation_languages (e.g. 20LmZk1hakA)
1613 # Does not used anymore as of 22.06.2017
1614 caption_tracks = args['caption_tracks']
1615 caption_translation_languages = args['caption_translation_languages']
1616 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1617 sub_lang_list = []
1618 for lang in caption_translation_languages.split(','):
1619 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1620 sub_lang = lang_qs.get('lc', [None])[0]
1621 if sub_lang:
1622 sub_lang_list.append(sub_lang)
1623 return make_captions(caption_url, sub_lang_list)
1624 # An extractor error can be raise by the download process if there are
1625 # no automatic captions but there are subtitles
1626 except (KeyError, IndexError, ExtractorError):
1627 self._downloader.report_warning(err_msg)
1628 return {}
1629
1630 def _mark_watched(self, video_id, video_info, player_response):
1631 playback_url = url_or_none(try_get(
1632 player_response,
1633 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1634 video_info, lambda x: x['videostats_playback_base_url'][0]))
1635 if not playback_url:
1636 return
1637 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1638 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1639
1640 # cpn generation algorithm is reverse engineered from base.js.
1641 # In fact it works even with dummy cpn.
1642 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1643 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1644
1645 qs.update({
1646 'ver': ['2'],
1647 'cpn': [cpn],
1648 })
1649 playback_url = compat_urlparse.urlunparse(
1650 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1651
1652 self._download_webpage(
1653 playback_url, video_id, 'Marking watched',
1654 'Unable to mark watched', fatal=False)
1655
1656 @staticmethod
1657 def _extract_urls(webpage):
1658 # Embedded YouTube player
1659 entries = [
1660 unescapeHTML(mobj.group('url'))
1661 for mobj in re.finditer(r'''(?x)
1662 (?:
1663 <iframe[^>]+?src=|
1664 data-video-url=|
1665 <embed[^>]+?src=|
1666 embedSWF\(?:\s*|
1667 <object[^>]+data=|
1668 new\s+SWFObject\(
1669 )
1670 (["\'])
1671 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1672 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1673 \1''', webpage)]
1674
1675 # lazyYT YouTube embed
1676 entries.extend(list(map(
1677 unescapeHTML,
1678 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1679
1680 # Wordpress "YouTube Video Importer" plugin
1681 matches = re.findall(r'''(?x)<div[^>]+
1682 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1683 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1684 entries.extend(m[-1] for m in matches)
1685
1686 return entries
1687
1688 @staticmethod
1689 def _extract_url(webpage):
1690 urls = YoutubeIE._extract_urls(webpage)
1691 return urls[0] if urls else None
1692
1693 @classmethod
1694 def extract_id(cls, url):
1695 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1696 if mobj is None:
1697 raise ExtractorError('Invalid URL: %s' % url)
1698 video_id = mobj.group(2)
1699 return video_id
1700
1701 def _extract_chapters_from_json(self, webpage, video_id, duration):
1702 if not webpage:
1703 return
1704 initial_data = self._parse_json(
1705 self._search_regex(
1706 r'window\["ytInitialData"\] = (.+);\n', webpage,
1707 'player args', default='{}'),
1708 video_id, fatal=False)
1709 if not initial_data or not isinstance(initial_data, dict):
1710 return
1711 chapters_list = try_get(
1712 initial_data,
1713 lambda x: x['playerOverlays']
1714 ['playerOverlayRenderer']
1715 ['decoratedPlayerBarRenderer']
1716 ['decoratedPlayerBarRenderer']
1717 ['playerBar']
1718 ['chapteredPlayerBarRenderer']
1719 ['chapters'],
1720 list)
1721 if not chapters_list:
1722 return
1723
1724 def chapter_time(chapter):
1725 return float_or_none(
1726 try_get(
1727 chapter,
1728 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1729 int),
1730 scale=1000)
1731 chapters = []
1732 for next_num, chapter in enumerate(chapters_list, start=1):
1733 start_time = chapter_time(chapter)
1734 if start_time is None:
1735 continue
1736 end_time = (chapter_time(chapters_list[next_num])
1737 if next_num < len(chapters_list) else duration)
1738 if end_time is None:
1739 continue
1740 title = try_get(
1741 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1742 compat_str)
1743 chapters.append({
1744 'start_time': start_time,
1745 'end_time': end_time,
1746 'title': title,
1747 })
1748 return chapters
1749
1750 @staticmethod
1751 def _extract_chapters_from_description(description, duration):
1752 if not description:
1753 return None
1754 chapter_lines = re.findall(
1755 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1756 description)
1757 if not chapter_lines:
1758 return None
1759 chapters = []
1760 for next_num, (chapter_line, time_point) in enumerate(
1761 chapter_lines, start=1):
1762 start_time = parse_duration(time_point)
1763 if start_time is None:
1764 continue
1765 if start_time > duration:
1766 break
1767 end_time = (duration if next_num == len(chapter_lines)
1768 else parse_duration(chapter_lines[next_num][1]))
1769 if end_time is None:
1770 continue
1771 if end_time > duration:
1772 end_time = duration
1773 if start_time > end_time:
1774 break
1775 chapter_title = re.sub(
1776 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1777 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1778 chapters.append({
1779 'start_time': start_time,
1780 'end_time': end_time,
1781 'title': chapter_title,
1782 })
1783 return chapters
1784
1785 def _extract_chapters(self, webpage, description, video_id, duration):
1786 return (self._extract_chapters_from_json(webpage, video_id, duration)
1787 or self._extract_chapters_from_description(description, duration))
1788
1789 def _real_extract(self, url):
1790 url, smuggled_data = unsmuggle_url(url, {})
1791
1792 proto = (
1793 'http' if self._downloader.params.get('prefer_insecure', False)
1794 else 'https')
1795
1796 start_time = None
1797 end_time = None
1798 parsed_url = compat_urllib_parse_urlparse(url)
1799 for component in [parsed_url.fragment, parsed_url.query]:
1800 query = compat_parse_qs(component)
1801 if start_time is None and 't' in query:
1802 start_time = parse_duration(query['t'][0])
1803 if start_time is None and 'start' in query:
1804 start_time = parse_duration(query['start'][0])
1805 if end_time is None and 'end' in query:
1806 end_time = parse_duration(query['end'][0])
1807
1808 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1809 mobj = re.search(self._NEXT_URL_RE, url)
1810 if mobj:
1811 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1812 video_id = self.extract_id(url)
1813
1814 # Get video webpage
1815 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1816 video_webpage, urlh = self._download_webpage_handle(url, video_id)
1817
1818 qs = compat_parse_qs(compat_urllib_parse_urlparse(urlh.geturl()).query)
1819 video_id = qs.get('v', [None])[0] or video_id
1820
1821 # Attempt to extract SWF player URL
1822 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1823 if mobj is not None:
1824 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1825 else:
1826 player_url = None
1827
1828 dash_mpds = []
1829
1830 def add_dash_mpd(video_info):
1831 dash_mpd = video_info.get('dashmpd')
1832 if dash_mpd and dash_mpd[0] not in dash_mpds:
1833 dash_mpds.append(dash_mpd[0])
1834
1835 def add_dash_mpd_pr(pl_response):
1836 dash_mpd = url_or_none(try_get(
1837 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1838 compat_str))
1839 if dash_mpd and dash_mpd not in dash_mpds:
1840 dash_mpds.append(dash_mpd)
1841
1842 is_live = None
1843 view_count = None
1844
1845 def extract_view_count(v_info):
1846 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1847
1848 def extract_player_response(player_response, video_id):
1849 pl_response = str_or_none(player_response)
1850 if not pl_response:
1851 return
1852 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1853 if isinstance(pl_response, dict):
1854 add_dash_mpd_pr(pl_response)
1855 return pl_response
1856
1857 player_response = {}
1858
1859 # Get video info
1860 video_info = {}
1861 embed_webpage = None
1862 if (self._og_search_property('restrictions:age', video_webpage, default=None) == '18+'
1863 or re.search(r'player-age-gate-content">', video_webpage) is not None):
1864 cookie_keys = self._get_cookies('https://www.youtube.com').keys()
1865 age_gate = True
1866 # We simulate the access to the video from www.youtube.com/v/{video_id}
1867 # this can be viewed without login into Youtube
1868 url = proto + '://www.youtube.com/embed/%s' % video_id
1869 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1870 # check if video is only playable on youtube - if so it requires auth (cookies)
1871 if re.search(r'player-unavailable">', embed_webpage) is not None:
1872 if ({'VISITOR_INFO1_LIVE', 'HSID', 'SSID', 'SID'} <= cookie_keys
1873 or {'VISITOR_INFO1_LIVE', '__Secure-3PSID', 'LOGIN_INFO'} <= cookie_keys):
1874 age_gate = False
1875 # Try looking directly into the video webpage
1876 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1877 if ytplayer_config:
1878 args = ytplayer_config['args']
1879 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1880 # Convert to the same format returned by compat_parse_qs
1881 video_info = dict((k, [v]) for k, v in args.items())
1882 add_dash_mpd(video_info)
1883 # Rental video is not rented but preview is available (e.g.
1884 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1885 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1886 if not video_info and args.get('ypc_vid'):
1887 return self.url_result(
1888 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1889 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1890 is_live = True
1891 if not player_response:
1892 player_response = extract_player_response(args.get('player_response'), video_id)
1893 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1894 add_dash_mpd_pr(player_response)
1895 else:
1896 raise ExtractorError('Video is age restricted and only playable on Youtube. Requires cookies!', expected=True)
1897 else:
1898 data = compat_urllib_parse_urlencode({
1899 'video_id': video_id,
1900 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1901 'sts': self._search_regex(
1902 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1903 })
1904 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1905 try:
1906 video_info_webpage = self._download_webpage(
1907 video_info_url, video_id,
1908 note='Refetching age-gated info webpage',
1909 errnote='unable to download video info webpage')
1910 except ExtractorError:
1911 video_info_webpage = None
1912 if video_info_webpage:
1913 video_info = compat_parse_qs(video_info_webpage)
1914 pl_response = video_info.get('player_response', [None])[0]
1915 player_response = extract_player_response(pl_response, video_id)
1916 add_dash_mpd(video_info)
1917 view_count = extract_view_count(video_info)
1918 else:
1919 age_gate = False
1920 # Try looking directly into the video webpage
1921 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1922 if ytplayer_config:
1923 args = ytplayer_config['args']
1924 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1925 # Convert to the same format returned by compat_parse_qs
1926 video_info = dict((k, [v]) for k, v in args.items())
1927 add_dash_mpd(video_info)
1928 # Rental video is not rented but preview is available (e.g.
1929 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1930 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1931 if not video_info and args.get('ypc_vid'):
1932 return self.url_result(
1933 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1934 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1935 is_live = True
1936 if not player_response:
1937 player_response = extract_player_response(args.get('player_response'), video_id)
1938 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1939 add_dash_mpd_pr(player_response)
1940
1941 def extract_unavailable_message():
1942 messages = []
1943 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1944 msg = self._html_search_regex(
1945 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1946 video_webpage, 'unavailable %s' % kind, default=None)
1947 if msg:
1948 messages.append(msg)
1949 if messages:
1950 return '\n'.join(messages)
1951
1952 if not video_info and not player_response:
1953 unavailable_message = extract_unavailable_message()
1954 if not unavailable_message:
1955 unavailable_message = 'Unable to extract video data'
1956 raise ExtractorError(
1957 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1958
1959 if not isinstance(video_info, dict):
1960 video_info = {}
1961
1962 video_details = try_get(
1963 player_response, lambda x: x['videoDetails'], dict) or {}
1964
1965 microformat = try_get(
1966 player_response, lambda x: x['microformat']['playerMicroformatRenderer'], dict) or {}
1967
1968 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1969 if not video_title:
1970 self._downloader.report_warning('Unable to extract video title')
1971 video_title = '_'
1972
1973 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1974 if video_description:
1975
1976 def replace_url(m):
1977 redir_url = compat_urlparse.urljoin(url, m.group(1))
1978 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1979 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1980 qs = compat_parse_qs(parsed_redir_url.query)
1981 q = qs.get('q')
1982 if q and q[0]:
1983 return q[0]
1984 return redir_url
1985
1986 description_original = video_description = re.sub(r'''(?x)
1987 <a\s+
1988 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1989 (?:title|href)="([^"]+)"\s+
1990 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1991 class="[^"]*"[^>]*>
1992 [^<]+\.{3}\s*
1993 </a>
1994 ''', replace_url, video_description)
1995 video_description = clean_html(video_description)
1996 else:
1997 video_description = video_details.get('shortDescription')
1998 if video_description is None:
1999 video_description = self._html_search_meta('description', video_webpage)
2000
2001 if not smuggled_data.get('force_singlefeed', False):
2002 if not self._downloader.params.get('noplaylist'):
2003 multifeed_metadata_list = try_get(
2004 player_response,
2005 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
2006 compat_str) or try_get(
2007 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
2008 if multifeed_metadata_list:
2009 entries = []
2010 feed_ids = []
2011 for feed in multifeed_metadata_list.split(','):
2012 # Unquote should take place before split on comma (,) since textual
2013 # fields may contain comma as well (see
2014 # https://github.com/ytdl-org/youtube-dl/issues/8536)
2015 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
2016
2017 def feed_entry(name):
2018 return try_get(feed_data, lambda x: x[name][0], compat_str)
2019
2020 feed_id = feed_entry('id')
2021 if not feed_id:
2022 continue
2023 feed_title = feed_entry('title')
2024 title = video_title
2025 if feed_title:
2026 title += ' (%s)' % feed_title
2027 entries.append({
2028 '_type': 'url_transparent',
2029 'ie_key': 'Youtube',
2030 'url': smuggle_url(
2031 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
2032 {'force_singlefeed': True}),
2033 'title': title,
2034 })
2035 feed_ids.append(feed_id)
2036 self.to_screen(
2037 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
2038 % (', '.join(feed_ids), video_id))
2039 return self.playlist_result(entries, video_id, video_title, video_description)
2040 else:
2041 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2042
2043 if view_count is None:
2044 view_count = extract_view_count(video_info)
2045 if view_count is None and video_details:
2046 view_count = int_or_none(video_details.get('viewCount'))
2047 if view_count is None and microformat:
2048 view_count = int_or_none(microformat.get('viewCount'))
2049
2050 if is_live is None:
2051 is_live = bool_or_none(video_details.get('isLive'))
2052
2053 has_live_chat_replay = False
2054 if not is_live:
2055 yt_initial_data = self._get_yt_initial_data(video_id, video_webpage)
2056 try:
2057 yt_initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2058 has_live_chat_replay = True
2059 except (KeyError, IndexError, TypeError):
2060 pass
2061
2062 # Check for "rental" videos
2063 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
2064 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
2065
2066 def _extract_filesize(media_url):
2067 return int_or_none(self._search_regex(
2068 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
2069
2070 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
2071 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
2072
2073 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
2074 self.report_rtmp_download()
2075 formats = [{
2076 'format_id': '_rtmp',
2077 'protocol': 'rtmp',
2078 'url': video_info['conn'][0],
2079 'player_url': player_url,
2080 }]
2081 elif not is_live and (streaming_formats or len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
2082 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
2083 if 'rtmpe%3Dyes' in encoded_url_map:
2084 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
2085 formats = []
2086 formats_spec = {}
2087 fmt_list = video_info.get('fmt_list', [''])[0]
2088 if fmt_list:
2089 for fmt in fmt_list.split(','):
2090 spec = fmt.split('/')
2091 if len(spec) > 1:
2092 width_height = spec[1].split('x')
2093 if len(width_height) == 2:
2094 formats_spec[spec[0]] = {
2095 'resolution': spec[1],
2096 'width': int_or_none(width_height[0]),
2097 'height': int_or_none(width_height[1]),
2098 }
2099 for fmt in streaming_formats:
2100 itag = str_or_none(fmt.get('itag'))
2101 if not itag:
2102 continue
2103 quality = fmt.get('quality')
2104 quality_label = fmt.get('qualityLabel') or quality
2105 formats_spec[itag] = {
2106 'asr': int_or_none(fmt.get('audioSampleRate')),
2107 'filesize': int_or_none(fmt.get('contentLength')),
2108 'format_note': quality_label,
2109 'fps': int_or_none(fmt.get('fps')),
2110 'height': int_or_none(fmt.get('height')),
2111 # bitrate for itag 43 is always 2147483647
2112 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
2113 'width': int_or_none(fmt.get('width')),
2114 }
2115
2116 for fmt in streaming_formats:
2117 if fmt.get('drmFamilies') or fmt.get('drm_families'):
2118 continue
2119 url = url_or_none(fmt.get('url'))
2120
2121 if not url:
2122 cipher = fmt.get('cipher') or fmt.get('signatureCipher')
2123 if not cipher:
2124 continue
2125 url_data = compat_parse_qs(cipher)
2126 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
2127 if not url:
2128 continue
2129 else:
2130 cipher = None
2131 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
2132
2133 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
2134 # Unsupported FORMAT_STREAM_TYPE_OTF
2135 if stream_type == 3:
2136 continue
2137
2138 format_id = fmt.get('itag') or url_data['itag'][0]
2139 if not format_id:
2140 continue
2141 format_id = compat_str(format_id)
2142
2143 if cipher:
2144 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
2145 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
2146 jsplayer_url_json = self._search_regex(
2147 ASSETS_RE,
2148 embed_webpage if age_gate else video_webpage,
2149 'JS player URL (1)', default=None)
2150 if not jsplayer_url_json and not age_gate:
2151 # We need the embed website after all
2152 if embed_webpage is None:
2153 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
2154 embed_webpage = self._download_webpage(
2155 embed_url, video_id, 'Downloading embed webpage')
2156 jsplayer_url_json = self._search_regex(
2157 ASSETS_RE, embed_webpage, 'JS player URL')
2158
2159 player_url = json.loads(jsplayer_url_json)
2160 if player_url is None:
2161 player_url_json = self._search_regex(
2162 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2163 video_webpage, 'age gate player URL')
2164 player_url = json.loads(player_url_json)
2165
2166 if 'sig' in url_data:
2167 url += '&signature=' + url_data['sig'][0]
2168 elif 's' in url_data:
2169 encrypted_sig = url_data['s'][0]
2170
2171 if self._downloader.params.get('verbose'):
2172 if player_url is None:
2173 player_desc = 'unknown'
2174 else:
2175 player_type, player_version = self._extract_player_info(player_url)
2176 player_desc = '%s player %s' % ('flash' if player_type == 'swf' else 'html5', player_version)
2177 parts_sizes = self._signature_cache_id(encrypted_sig)
2178 self.to_screen('{%s} signature length %s, %s' %
2179 (format_id, parts_sizes, player_desc))
2180
2181 signature = self._decrypt_signature(
2182 encrypted_sig, video_id, player_url, age_gate)
2183 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2184 url += '&%s=%s' % (sp, signature)
2185 if 'ratebypass' not in url:
2186 url += '&ratebypass=yes'
2187
2188 dct = {
2189 'format_id': format_id,
2190 'url': url,
2191 'player_url': player_url,
2192 }
2193 if format_id in self._formats:
2194 dct.update(self._formats[format_id])
2195 if format_id in formats_spec:
2196 dct.update(formats_spec[format_id])
2197
2198 # Some itags are not included in DASH manifest thus corresponding formats will
2199 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2200 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2201 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2202 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2203
2204 if width is None:
2205 width = int_or_none(fmt.get('width'))
2206 if height is None:
2207 height = int_or_none(fmt.get('height'))
2208
2209 filesize = int_or_none(url_data.get(
2210 'clen', [None])[0]) or _extract_filesize(url)
2211
2212 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2213 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2214
2215 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2216 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2217 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2218
2219 more_fields = {
2220 'filesize': filesize,
2221 'tbr': tbr,
2222 'width': width,
2223 'height': height,
2224 'fps': fps,
2225 'format_note': quality_label or quality,
2226 }
2227 for key, value in more_fields.items():
2228 if value:
2229 dct[key] = value
2230 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2231 if type_:
2232 type_split = type_.split(';')
2233 kind_ext = type_split[0].split('/')
2234 if len(kind_ext) == 2:
2235 kind, _ = kind_ext
2236 dct['ext'] = mimetype2ext(type_split[0])
2237 if kind in ('audio', 'video'):
2238 codecs = None
2239 for mobj in re.finditer(
2240 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2241 if mobj.group('key') == 'codecs':
2242 codecs = mobj.group('val')
2243 break
2244 if codecs:
2245 dct.update(parse_codecs(codecs))
2246 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2247 dct['downloader_options'] = {
2248 # Youtube throttles chunks >~10M
2249 'http_chunk_size': 10485760,
2250 }
2251 formats.append(dct)
2252 else:
2253 manifest_url = (
2254 url_or_none(try_get(
2255 player_response,
2256 lambda x: x['streamingData']['hlsManifestUrl'],
2257 compat_str))
2258 or url_or_none(try_get(
2259 video_info, lambda x: x['hlsvp'][0], compat_str)))
2260 if manifest_url:
2261 formats = []
2262 m3u8_formats = self._extract_m3u8_formats(
2263 manifest_url, video_id, 'mp4', fatal=False)
2264 for a_format in m3u8_formats:
2265 itag = self._search_regex(
2266 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2267 if itag:
2268 a_format['format_id'] = itag
2269 if itag in self._formats:
2270 dct = self._formats[itag].copy()
2271 dct.update(a_format)
2272 a_format = dct
2273 a_format['player_url'] = player_url
2274 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2275 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2276 if self._downloader.params.get('youtube_include_hls_manifest', True):
2277 formats.append(a_format)
2278 else:
2279 error_message = extract_unavailable_message()
2280 if not error_message:
2281 error_message = clean_html(try_get(
2282 player_response, lambda x: x['playabilityStatus']['reason'],
2283 compat_str))
2284 if not error_message:
2285 error_message = clean_html(
2286 try_get(video_info, lambda x: x['reason'][0], compat_str))
2287 if error_message:
2288 raise ExtractorError(error_message, expected=True)
2289 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2290
2291 # uploader
2292 video_uploader = try_get(
2293 video_info, lambda x: x['author'][0],
2294 compat_str) or str_or_none(video_details.get('author'))
2295 if video_uploader:
2296 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2297 else:
2298 self._downloader.report_warning('unable to extract uploader name')
2299
2300 # uploader_id
2301 video_uploader_id = None
2302 video_uploader_url = None
2303 mobj = re.search(
2304 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2305 video_webpage)
2306 if mobj is not None:
2307 video_uploader_id = mobj.group('uploader_id')
2308 video_uploader_url = mobj.group('uploader_url')
2309 else:
2310 owner_profile_url = url_or_none(microformat.get('ownerProfileUrl'))
2311 if owner_profile_url:
2312 video_uploader_id = self._search_regex(
2313 r'(?:user|channel)/([^/]+)', owner_profile_url, 'uploader id',
2314 default=None)
2315 video_uploader_url = owner_profile_url
2316
2317 channel_id = (
2318 str_or_none(video_details.get('channelId'))
2319 or self._html_search_meta(
2320 'channelId', video_webpage, 'channel id', default=None)
2321 or self._search_regex(
2322 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2323 video_webpage, 'channel id', default=None, group='id'))
2324 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2325
2326 thumbnails = []
2327 thumbnails_list = try_get(
2328 video_details, lambda x: x['thumbnail']['thumbnails'], list) or []
2329 for t in thumbnails_list:
2330 if not isinstance(t, dict):
2331 continue
2332 thumbnail_url = url_or_none(t.get('url'))
2333 if not thumbnail_url:
2334 continue
2335 thumbnails.append({
2336 'url': thumbnail_url,
2337 'width': int_or_none(t.get('width')),
2338 'height': int_or_none(t.get('height')),
2339 })
2340
2341 if not thumbnails:
2342 video_thumbnail = None
2343 # We try first to get a high quality image:
2344 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2345 video_webpage, re.DOTALL)
2346 if m_thumb is not None:
2347 video_thumbnail = m_thumb.group(1)
2348 thumbnail_url = try_get(video_info, lambda x: x['thumbnail_url'][0], compat_str)
2349 if thumbnail_url:
2350 video_thumbnail = compat_urllib_parse_unquote_plus(thumbnail_url)
2351 if video_thumbnail:
2352 thumbnails.append({'url': video_thumbnail})
2353
2354 # upload date
2355 upload_date = self._html_search_meta(
2356 'datePublished', video_webpage, 'upload date', default=None)
2357 if not upload_date:
2358 upload_date = self._search_regex(
2359 [r'(?s)id="eow-date.*?>(.*?)</span>',
2360 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2361 video_webpage, 'upload date', default=None)
2362 if not upload_date:
2363 upload_date = microformat.get('publishDate') or microformat.get('uploadDate')
2364 upload_date = unified_strdate(upload_date)
2365
2366 video_license = self._html_search_regex(
2367 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2368 video_webpage, 'license', default=None)
2369
2370 m_music = re.search(
2371 r'''(?x)
2372 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2373 <ul[^>]*>\s*
2374 <li>(?P<title>.+?)
2375 by (?P<creator>.+?)
2376 (?:
2377 \(.+?\)|
2378 <a[^>]*
2379 (?:
2380 \bhref=["\']/red[^>]*>| # drop possible
2381 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2382 )
2383 .*?
2384 )?</li
2385 ''',
2386 video_webpage)
2387 if m_music:
2388 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2389 video_creator = clean_html(m_music.group('creator'))
2390 else:
2391 video_alt_title = video_creator = None
2392
2393 def extract_meta(field):
2394 return self._html_search_regex(
2395 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2396 video_webpage, field, default=None)
2397
2398 track = extract_meta('Song')
2399 artist = extract_meta('Artist')
2400 album = extract_meta('Album')
2401
2402 # Youtube Music Auto-generated description
2403 release_date = release_year = None
2404 if video_description:
2405 mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[]+)·(?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]+))?', video_description)
2406 if mobj:
2407 if not track:
2408 track = mobj.group('track').strip()
2409 if not artist:
2410 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2411 if not album:
2412 album = mobj.group('album'.strip())
2413 release_year = mobj.group('release_year')
2414 release_date = mobj.group('release_date')
2415 if release_date:
2416 release_date = release_date.replace('-', '')
2417 if not release_year:
2418 release_year = int(release_date[:4])
2419 if release_year:
2420 release_year = int(release_year)
2421
2422 m_episode = re.search(
2423 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2424 video_webpage)
2425 if m_episode:
2426 series = unescapeHTML(m_episode.group('series'))
2427 season_number = int(m_episode.group('season'))
2428 episode_number = int(m_episode.group('episode'))
2429 else:
2430 series = season_number = episode_number = None
2431
2432 m_cat_container = self._search_regex(
2433 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2434 video_webpage, 'categories', default=None)
2435 category = None
2436 if m_cat_container:
2437 category = self._html_search_regex(
2438 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2439 default=None)
2440 if not category:
2441 category = try_get(
2442 microformat, lambda x: x['category'], compat_str)
2443 video_categories = None if category is None else [category]
2444
2445 video_tags = [
2446 unescapeHTML(m.group('content'))
2447 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2448 if not video_tags:
2449 video_tags = try_get(video_details, lambda x: x['keywords'], list)
2450
2451 def _extract_count(count_name):
2452 return str_to_int(self._search_regex(
2453 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2454 % re.escape(count_name),
2455 video_webpage, count_name, default=None))
2456
2457 like_count = _extract_count('like')
2458 dislike_count = _extract_count('dislike')
2459
2460 if view_count is None:
2461 view_count = str_to_int(self._search_regex(
2462 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2463 'view count', default=None))
2464
2465 average_rating = (
2466 float_or_none(video_details.get('averageRating'))
2467 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2468
2469 # subtitles
2470 video_subtitles = self.extract_subtitles(
2471 video_id, video_webpage, has_live_chat_replay)
2472 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2473
2474 video_duration = try_get(
2475 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2476 if not video_duration:
2477 video_duration = int_or_none(video_details.get('lengthSeconds'))
2478 if not video_duration:
2479 video_duration = parse_duration(self._html_search_meta(
2480 'duration', video_webpage, 'video duration'))
2481
2482 # annotations
2483 video_annotations = None
2484 if self._downloader.params.get('writeannotations', False):
2485 xsrf_token = self._search_regex(
2486 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2487 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2488 invideo_url = try_get(
2489 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2490 if xsrf_token and invideo_url:
2491 xsrf_field_name = self._search_regex(
2492 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2493 video_webpage, 'xsrf field name',
2494 group='xsrf_field_name', default='session_token')
2495 video_annotations = self._download_webpage(
2496 self._proto_relative_url(invideo_url),
2497 video_id, note='Downloading annotations',
2498 errnote='Unable to download video annotations', fatal=False,
2499 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2500
2501 chapters = self._extract_chapters(video_webpage, description_original, video_id, video_duration)
2502
2503 # Look for the DASH manifest
2504 if self._downloader.params.get('youtube_include_dash_manifest', True):
2505 dash_mpd_fatal = True
2506 for mpd_url in dash_mpds:
2507 dash_formats = {}
2508 try:
2509 def decrypt_sig(mobj):
2510 s = mobj.group(1)
2511 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2512 return '/signature/%s' % dec_s
2513
2514 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2515
2516 for df in self._extract_mpd_formats(
2517 mpd_url, video_id, fatal=dash_mpd_fatal,
2518 formats_dict=self._formats):
2519 if not df.get('filesize'):
2520 df['filesize'] = _extract_filesize(df['url'])
2521 # Do not overwrite DASH format found in some previous DASH manifest
2522 if df['format_id'] not in dash_formats:
2523 dash_formats[df['format_id']] = df
2524 # Additional DASH manifests may end up in HTTP Error 403 therefore
2525 # allow them to fail without bug report message if we already have
2526 # some DASH manifest succeeded. This is temporary workaround to reduce
2527 # burst of bug reports until we figure out the reason and whether it
2528 # can be fixed at all.
2529 dash_mpd_fatal = False
2530 except (ExtractorError, KeyError) as e:
2531 self.report_warning(
2532 'Skipping DASH manifest: %r' % e, video_id)
2533 if dash_formats:
2534 # Remove the formats we found through non-DASH, they
2535 # contain less info and it can be wrong, because we use
2536 # fixed values (for example the resolution). See
2537 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2538 # example.
2539 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2540 formats.extend(dash_formats.values())
2541
2542 # Check for malformed aspect ratio
2543 stretched_m = re.search(
2544 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2545 video_webpage)
2546 if stretched_m:
2547 w = float(stretched_m.group('w'))
2548 h = float(stretched_m.group('h'))
2549 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2550 # We will only process correct ratios.
2551 if w > 0 and h > 0:
2552 ratio = w / h
2553 for f in formats:
2554 if f.get('vcodec') != 'none':
2555 f['stretched_ratio'] = ratio
2556
2557 if not formats:
2558 if 'reason' in video_info:
2559 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2560 regions_allowed = self._html_search_meta(
2561 'regionsAllowed', video_webpage, default=None)
2562 countries = regions_allowed.split(',') if regions_allowed else None
2563 self.raise_geo_restricted(
2564 msg=video_info['reason'][0], countries=countries)
2565 reason = video_info['reason'][0]
2566 if 'Invalid parameters' in reason:
2567 unavailable_message = extract_unavailable_message()
2568 if unavailable_message:
2569 reason = unavailable_message
2570 raise ExtractorError(
2571 'YouTube said: %s' % reason,
2572 expected=True, video_id=video_id)
2573 if video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos']):
2574 raise ExtractorError('This video is DRM protected.', expected=True)
2575
2576 self._sort_formats(formats)
2577
2578 self.mark_watched(video_id, video_info, player_response)
2579
2580 return {
2581 'id': video_id,
2582 'uploader': video_uploader,
2583 'uploader_id': video_uploader_id,
2584 'uploader_url': video_uploader_url,
2585 'channel_id': channel_id,
2586 'channel_url': channel_url,
2587 'upload_date': upload_date,
2588 'license': video_license,
2589 'creator': video_creator or artist,
2590 'title': video_title,
2591 'alt_title': video_alt_title or track,
2592 'thumbnails': thumbnails,
2593 'description': video_description,
2594 'categories': video_categories,
2595 'tags': video_tags,
2596 'subtitles': video_subtitles,
2597 'automatic_captions': automatic_captions,
2598 'duration': video_duration,
2599 'age_limit': 18 if age_gate else 0,
2600 'annotations': video_annotations,
2601 'chapters': chapters,
2602 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2603 'view_count': view_count,
2604 'like_count': like_count,
2605 'dislike_count': dislike_count,
2606 'average_rating': average_rating,
2607 'formats': formats,
2608 'is_live': is_live,
2609 'start_time': start_time,
2610 'end_time': end_time,
2611 'series': series,
2612 'season_number': season_number,
2613 'episode_number': episode_number,
2614 'track': track,
2615 'artist': artist,
2616 'album': album,
2617 'release_date': release_date,
2618 'release_year': release_year,
2619 }
2620
2621
2622 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2623 IE_DESC = 'YouTube.com playlists'
2624 _VALID_URL = r"""(?x)(?:
2625 (?:https?://)?
2626 (?:\w+\.)?
2627 (?:
2628 (?:
2629 youtube(?:kids)?\.com|
2630 invidio\.us
2631 )
2632 /
2633 (?:
2634 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2635 \? (?:.*?[&;])*? (?:p|a|list)=
2636 | p/
2637 )|
2638 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2639 )
2640 (
2641 (?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2642 # Top tracks, they can also include dots
2643 |(?:MC)[\w\.]*
2644 )
2645 .*
2646 |
2647 (%(playlist_id)s)
2648 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2649 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2650 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2651 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2652 IE_NAME = 'youtube:playlist'
2653 _TESTS = [{
2654 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2655 'info_dict': {
2656 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2657 'uploader': 'Sergey M.',
2658 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2659 'title': 'youtube-dl public playlist',
2660 },
2661 'playlist_count': 1,
2662 }, {
2663 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2664 'info_dict': {
2665 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2666 'uploader': 'Sergey M.',
2667 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2668 'title': 'youtube-dl empty playlist',
2669 },
2670 'playlist_count': 0,
2671 }, {
2672 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2673 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2674 'info_dict': {
2675 'title': '29C3: Not my department',
2676 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2677 'uploader': 'Christiaan008',
2678 'uploader_id': 'ChRiStIaAn008',
2679 },
2680 'playlist_count': 96,
2681 }, {
2682 'note': 'issue #673',
2683 'url': 'PLBB231211A4F62143',
2684 'info_dict': {
2685 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2686 'id': 'PLBB231211A4F62143',
2687 'uploader': 'Wickydoo',
2688 'uploader_id': 'Wickydoo',
2689 },
2690 'playlist_mincount': 26,
2691 }, {
2692 'note': 'Large playlist',
2693 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2694 'info_dict': {
2695 'title': 'Uploads from Cauchemar',
2696 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2697 'uploader': 'Cauchemar',
2698 'uploader_id': 'Cauchemar89',
2699 },
2700 'playlist_mincount': 799,
2701 }, {
2702 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2703 'info_dict': {
2704 'title': 'YDL_safe_search',
2705 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2706 },
2707 'playlist_count': 2,
2708 'skip': 'This playlist is private',
2709 }, {
2710 'note': 'embedded',
2711 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2712 'playlist_count': 4,
2713 'info_dict': {
2714 'title': 'JODA15',
2715 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2716 'uploader': 'milan',
2717 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2718 }
2719 }, {
2720 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2721 'playlist_mincount': 485,
2722 'info_dict': {
2723 'title': '2018 Chinese New Singles (11/6 updated)',
2724 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2725 'uploader': 'LBK',
2726 'uploader_id': 'sdragonfang',
2727 }
2728 }, {
2729 'note': 'Embedded SWF player',
2730 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2731 'playlist_count': 4,
2732 'info_dict': {
2733 'title': 'JODA7',
2734 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2735 },
2736 'skip': 'This playlist does not exist',
2737 }, {
2738 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2739 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2740 'info_dict': {
2741 'title': 'Uploads from Interstellar Movie',
2742 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2743 'uploader': 'Interstellar Movie',
2744 'uploader_id': 'InterstellarMovie1',
2745 },
2746 'playlist_mincount': 21,
2747 }, {
2748 # Playlist URL that does not actually serve a playlist
2749 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2750 'info_dict': {
2751 'id': 'FqZTN594JQw',
2752 'ext': 'webm',
2753 'title': "Smiley's People 01 detective, Adventure Series, Action",
2754 'uploader': 'STREEM',
2755 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2756 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2757 'upload_date': '20150526',
2758 'license': 'Standard YouTube License',
2759 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2760 'categories': ['People & Blogs'],
2761 'tags': list,
2762 'view_count': int,
2763 'like_count': int,
2764 'dislike_count': int,
2765 },
2766 'params': {
2767 'skip_download': True,
2768 },
2769 'skip': 'This video is not available.',
2770 'add_ie': [YoutubeIE.ie_key()],
2771 }, {
2772 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2773 'info_dict': {
2774 'id': 'yeWKywCrFtk',
2775 'ext': 'mp4',
2776 'title': 'Small Scale Baler and Braiding Rugs',
2777 'uploader': 'Backus-Page House Museum',
2778 'uploader_id': 'backuspagemuseum',
2779 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2780 'upload_date': '20161008',
2781 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2782 'categories': ['Nonprofits & Activism'],
2783 'tags': list,
2784 'like_count': int,
2785 'dislike_count': int,
2786 },
2787 'params': {
2788 'noplaylist': True,
2789 'skip_download': True,
2790 },
2791 }, {
2792 # https://github.com/ytdl-org/youtube-dl/issues/21844
2793 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2794 'info_dict': {
2795 'title': 'Data Analysis with Dr Mike Pound',
2796 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2797 'uploader_id': 'Computerphile',
2798 'uploader': 'Computerphile',
2799 },
2800 'playlist_mincount': 11,
2801 }, {
2802 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2803 'only_matching': True,
2804 }, {
2805 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2806 'only_matching': True,
2807 }, {
2808 # music album playlist
2809 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2810 'only_matching': True,
2811 }, {
2812 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2813 'only_matching': True,
2814 }, {
2815 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2816 'only_matching': True,
2817 }]
2818
2819 def _real_initialize(self):
2820 self._login()
2821
2822 def extract_videos_from_page(self, page):
2823 ids_in_page = []
2824 titles_in_page = []
2825
2826 for item in re.findall(
2827 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2828 attrs = extract_attributes(item)
2829 video_id = attrs['data-video-id']
2830 video_title = unescapeHTML(attrs.get('data-title'))
2831 if video_title:
2832 video_title = video_title.strip()
2833 ids_in_page.append(video_id)
2834 titles_in_page.append(video_title)
2835
2836 # Fallback with old _VIDEO_RE
2837 self.extract_videos_from_page_impl(
2838 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2839
2840 # Relaxed fallbacks
2841 self.extract_videos_from_page_impl(
2842 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2843 ids_in_page, titles_in_page)
2844 self.extract_videos_from_page_impl(
2845 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2846 ids_in_page, titles_in_page)
2847
2848 return zip(ids_in_page, titles_in_page)
2849
2850 def _extract_mix(self, playlist_id):
2851 # The mixes are generated from a single video
2852 # the id of the playlist is just 'RD' + video_id
2853 ids = []
2854 last_id = playlist_id[-11:]
2855 for n in itertools.count(1):
2856 url = 'https://www.youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2857 webpage = self._download_webpage(
2858 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2859 new_ids = orderedSet(re.findall(
2860 r'''(?xs)data-video-username=".*?".*?
2861 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2862 webpage))
2863 # Fetch new pages until all the videos are repeated, it seems that
2864 # there are always 51 unique videos.
2865 new_ids = [_id for _id in new_ids if _id not in ids]
2866 if not new_ids:
2867 break
2868 ids.extend(new_ids)
2869 last_id = ids[-1]
2870
2871 url_results = self._ids_to_results(ids)
2872
2873 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2874 title_span = (
2875 search_title('playlist-title')
2876 or search_title('title long-title')
2877 or search_title('title'))
2878 title = clean_html(title_span)
2879
2880 return self.playlist_result(url_results, playlist_id, title)
2881
2882 def _extract_playlist(self, playlist_id):
2883 url = self._TEMPLATE_URL % playlist_id
2884 page = self._download_webpage(url, playlist_id)
2885
2886 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2887 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2888 match = match.strip()
2889 # Check if the playlist exists or is private
2890 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2891 if mobj:
2892 reason = mobj.group('reason')
2893 message = 'This playlist %s' % reason
2894 if 'private' in reason:
2895 message += ', use --username or --netrc to access it'
2896 message += '.'
2897 raise ExtractorError(message, expected=True)
2898 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2899 raise ExtractorError(
2900 'Invalid parameters. Maybe URL is incorrect.',
2901 expected=True)
2902 elif re.match(r'[^<]*Choose your language[^<]*', match):
2903 continue
2904 else:
2905 self.report_warning('Youtube gives an alert message: ' + match)
2906
2907 playlist_title = self._html_search_regex(
2908 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2909 page, 'title', default=None)
2910
2911 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2912 uploader = self._html_search_regex(
2913 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2914 page, 'uploader', default=None)
2915 mobj = re.search(
2916 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2917 page)
2918 if mobj:
2919 uploader_id = mobj.group('uploader_id')
2920 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2921 else:
2922 uploader_id = uploader_url = None
2923
2924 has_videos = True
2925
2926 if not playlist_title:
2927 try:
2928 # Some playlist URLs don't actually serve a playlist (e.g.
2929 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2930 next(self._entries(page, playlist_id))
2931 except StopIteration:
2932 has_videos = False
2933
2934 playlist = self.playlist_result(
2935 self._entries(page, playlist_id), playlist_id, playlist_title)
2936 playlist.update({
2937 'uploader': uploader,
2938 'uploader_id': uploader_id,
2939 'uploader_url': uploader_url,
2940 })
2941
2942 return has_videos, playlist
2943
2944 def _check_download_just_video(self, url, playlist_id):
2945 # Check if it's a video-specific URL
2946 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2947 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2948 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2949 'video id', default=None)
2950 if video_id:
2951 if self._downloader.params.get('noplaylist'):
2952 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2953 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2954 else:
2955 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2956 return video_id, None
2957 return None, None
2958
2959 def _real_extract(self, url):
2960 # Extract playlist id
2961 mobj = re.match(self._VALID_URL, url)
2962 if mobj is None:
2963 raise ExtractorError('Invalid URL: %s' % url)
2964 playlist_id = mobj.group(1) or mobj.group(2)
2965
2966 video_id, video = self._check_download_just_video(url, playlist_id)
2967 if video:
2968 return video
2969
2970 if playlist_id.startswith(('RD', 'UL', 'PU')):
2971 # Mixes require a custom extraction process
2972 return self._extract_mix(playlist_id)
2973
2974 has_videos, playlist = self._extract_playlist(playlist_id)
2975 if has_videos or not video_id:
2976 return playlist
2977
2978 # Some playlist URLs don't actually serve a playlist (see
2979 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2980 # Fallback to plain video extraction if there is a video id
2981 # along with playlist id.
2982 return self.url_result(video_id, 'Youtube', video_id=video_id)
2983
2984
2985 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2986 IE_DESC = 'YouTube.com channels'
2987 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie|kids)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2988 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2989 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2990 IE_NAME = 'youtube:channel'
2991 _TESTS = [{
2992 'note': 'paginated channel',
2993 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2994 'playlist_mincount': 91,
2995 'info_dict': {
2996 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2997 'title': 'Uploads from lex will',
2998 'uploader': 'lex will',
2999 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
3000 }
3001 }, {
3002 'note': 'Age restricted channel',
3003 # from https://www.youtube.com/user/DeusExOfficial
3004 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
3005 'playlist_mincount': 64,
3006 'info_dict': {
3007 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
3008 'title': 'Uploads from Deus Ex',
3009 'uploader': 'Deus Ex',
3010 'uploader_id': 'DeusExOfficial',
3011 },
3012 }, {
3013 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
3014 'only_matching': True,
3015 }, {
3016 'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA',
3017 'only_matching': True,
3018 }]
3019
3020 @classmethod
3021 def suitable(cls, url):
3022 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
3023 else super(YoutubeChannelIE, cls).suitable(url))
3024
3025 def _build_template_url(self, url, channel_id):
3026 return self._TEMPLATE_URL % channel_id
3027
3028 def _real_extract(self, url):
3029 channel_id = self._match_id(url)
3030
3031 url = self._build_template_url(url, channel_id)
3032
3033 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
3034 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
3035 # otherwise fallback on channel by page extraction
3036 channel_page = self._download_webpage(
3037 url + '?view=57', channel_id,
3038 'Downloading channel page', fatal=False)
3039 if channel_page is False:
3040 channel_playlist_id = False
3041 else:
3042 channel_playlist_id = self._html_search_meta(
3043 'channelId', channel_page, 'channel id', default=None)
3044 if not channel_playlist_id:
3045 channel_url = self._html_search_meta(
3046 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
3047 channel_page, 'channel url', default=None)
3048 if channel_url:
3049 channel_playlist_id = self._search_regex(
3050 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
3051 channel_url, 'channel id', default=None)
3052 if channel_playlist_id and channel_playlist_id.startswith('UC'):
3053 playlist_id = 'UU' + channel_playlist_id[2:]
3054 return self.url_result(
3055 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
3056
3057 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
3058 autogenerated = re.search(r'''(?x)
3059 class="[^"]*?(?:
3060 channel-header-autogenerated-label|
3061 yt-channel-title-autogenerated
3062 )[^"]*"''', channel_page) is not None
3063
3064 if autogenerated:
3065 # The videos are contained in a single page
3066 # the ajax pages can't be used, they are empty
3067 entries = [
3068 self.url_result(
3069 video_id, 'Youtube', video_id=video_id,
3070 video_title=video_title)
3071 for video_id, video_title in self.extract_videos_from_page(channel_page)]
3072 return self.playlist_result(entries, channel_id)
3073
3074 try:
3075 next(self._entries(channel_page, channel_id))
3076 except StopIteration:
3077 alert_message = self._html_search_regex(
3078 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
3079 channel_page, 'alert', default=None, group='alert')
3080 if alert_message:
3081 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
3082
3083 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
3084
3085
3086 class YoutubeUserIE(YoutubeChannelIE):
3087 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
3088 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9%-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_%-]+)'
3089 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
3090 IE_NAME = 'youtube:user'
3091
3092 _TESTS = [{
3093 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
3094 'playlist_mincount': 320,
3095 'info_dict': {
3096 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
3097 'title': 'Uploads from The Linux Foundation',
3098 'uploader': 'The Linux Foundation',
3099 'uploader_id': 'TheLinuxFoundation',
3100 }
3101 }, {
3102 # Only available via https://www.youtube.com/c/12minuteathlete/videos
3103 # but not https://www.youtube.com/user/12minuteathlete/videos
3104 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
3105 'playlist_mincount': 249,
3106 'info_dict': {
3107 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
3108 'title': 'Uploads from 12 Minute Athlete',
3109 'uploader': '12 Minute Athlete',
3110 'uploader_id': 'the12minuteathlete',
3111 }
3112 }, {
3113 'url': 'ytuser:phihag',
3114 'only_matching': True,
3115 }, {
3116 'url': 'https://www.youtube.com/c/gametrailers',
3117 'only_matching': True,
3118 }, {
3119 'url': 'https://www.youtube.com/c/Pawe%C5%82Zadro%C5%BCniak',
3120 'only_matching': True,
3121 }, {
3122 'url': 'https://www.youtube.com/gametrailers',
3123 'only_matching': True,
3124 }, {
3125 # This channel is not available, geo restricted to JP
3126 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
3127 'only_matching': True,
3128 }]
3129
3130 @classmethod
3131 def suitable(cls, url):
3132 # Don't return True if the url can be extracted with other youtube
3133 # extractor, the regex would is too permissive and it would match.
3134 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
3135 if any(ie.suitable(url) for ie in other_yt_ies):
3136 return False
3137 else:
3138 return super(YoutubeUserIE, cls).suitable(url)
3139
3140 def _build_template_url(self, url, channel_id):
3141 mobj = re.match(self._VALID_URL, url)
3142 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
3143
3144
3145 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
3146 IE_DESC = 'YouTube.com live streams'
3147 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
3148 IE_NAME = 'youtube:live'
3149
3150 _TESTS = [{
3151 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
3152 'info_dict': {
3153 'id': 'a48o2S1cPoo',
3154 'ext': 'mp4',
3155 'title': 'The Young Turks - Live Main Show',
3156 'uploader': 'The Young Turks',
3157 'uploader_id': 'TheYoungTurks',
3158 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
3159 'upload_date': '20150715',
3160 'license': 'Standard YouTube License',
3161 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
3162 'categories': ['News & Politics'],
3163 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
3164 'like_count': int,
3165 'dislike_count': int,
3166 },
3167 'params': {
3168 'skip_download': True,
3169 },
3170 }, {
3171 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
3172 'only_matching': True,
3173 }, {
3174 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3175 'only_matching': True,
3176 }, {
3177 'url': 'https://www.youtube.com/TheYoungTurks/live',
3178 'only_matching': True,
3179 }]
3180
3181 def _real_extract(self, url):
3182 mobj = re.match(self._VALID_URL, url)
3183 channel_id = mobj.group('id')
3184 base_url = mobj.group('base_url')
3185 webpage = self._download_webpage(url, channel_id, fatal=False)
3186 if webpage:
3187 page_type = self._og_search_property(
3188 'type', webpage, 'page type', default='')
3189 video_id = self._html_search_meta(
3190 'videoId', webpage, 'video id', default=None)
3191 if page_type.startswith('video') and video_id and re.match(
3192 r'^[0-9A-Za-z_-]{11}$', video_id):
3193 return self.url_result(video_id, YoutubeIE.ie_key())
3194 return self.url_result(base_url)
3195
3196
3197 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
3198 IE_DESC = 'YouTube.com user/channel playlists'
3199 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel|c)/(?P<id>[^/]+)/playlists'
3200 IE_NAME = 'youtube:playlists'
3201
3202 _TESTS = [{
3203 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3204 'playlist_mincount': 4,
3205 'info_dict': {
3206 'id': 'ThirstForScience',
3207 'title': 'ThirstForScience',
3208 },
3209 }, {
3210 # with "Load more" button
3211 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3212 'playlist_mincount': 70,
3213 'info_dict': {
3214 'id': 'igorkle1',
3215 'title': 'Игорь Клейнер',
3216 },
3217 }, {
3218 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3219 'playlist_mincount': 17,
3220 'info_dict': {
3221 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3222 'title': 'Chem Player',
3223 },
3224 'skip': 'Blocked',
3225 }, {
3226 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
3227 'only_matching': True,
3228 }]
3229
3230
3231 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3232 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3233
3234
3235 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3236 IE_DESC = 'YouTube.com searches'
3237 # there doesn't appear to be a real limit, for example if you search for
3238 # 'python' you get more than 8.000.000 results
3239 _MAX_RESULTS = float('inf')
3240 IE_NAME = 'youtube:search'
3241 _SEARCH_KEY = 'ytsearch'
3242 _EXTRA_QUERY_ARGS = {}
3243 _TESTS = []
3244
3245 def _get_n_results(self, query, n):
3246 """Get a specified number of results for a query"""
3247
3248 videos = []
3249 limit = n
3250
3251 url_query = {
3252 'search_query': query.encode('utf-8'),
3253 }
3254 url_query.update(self._EXTRA_QUERY_ARGS)
3255 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3256
3257 for pagenum in itertools.count(1):
3258 data = self._download_json(
3259 result_url, video_id='query "%s"' % query,
3260 note='Downloading page %s' % pagenum,
3261 errnote='Unable to download API page',
3262 query={'spf': 'navigate'})
3263 html_content = data[1]['body']['content']
3264
3265 if 'class="search-message' in html_content:
3266 raise ExtractorError(
3267 '[youtube] No video results', expected=True)
3268
3269 new_videos = list(self._process_page(html_content))
3270 videos += new_videos
3271 if not new_videos or len(videos) > limit:
3272 break
3273 next_link = self._html_search_regex(
3274 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3275 html_content, 'next link', default=None)
3276 if next_link is None:
3277 break
3278 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3279
3280 if len(videos) > n:
3281 videos = videos[:n]
3282 return self.playlist_result(videos, query)
3283
3284
3285 class YoutubeSearchDateIE(YoutubeSearchIE):
3286 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3287 _SEARCH_KEY = 'ytsearchdate'
3288 IE_DESC = 'YouTube.com searches, newest videos first'
3289 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3290
3291
3292 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3293 IE_DESC = 'YouTube.com search URLs'
3294 IE_NAME = 'youtube:search_url'
3295 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3296 _TESTS = [{
3297 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3298 'playlist_mincount': 5,
3299 'info_dict': {
3300 'title': 'youtube-dl test video',
3301 }
3302 }, {
3303 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3304 'only_matching': True,
3305 }]
3306
3307 def _real_extract(self, url):
3308 mobj = re.match(self._VALID_URL, url)
3309 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3310 webpage = self._download_webpage(url, query)
3311 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3312
3313
3314 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3315 IE_DESC = 'YouTube.com (multi-season) shows'
3316 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3317 IE_NAME = 'youtube:show'
3318 _TESTS = [{
3319 'url': 'https://www.youtube.com/show/airdisasters',
3320 'playlist_mincount': 5,
3321 'info_dict': {
3322 'id': 'airdisasters',
3323 'title': 'Air Disasters',
3324 }
3325 }]
3326
3327 def _real_extract(self, url):
3328 playlist_id = self._match_id(url)
3329 return super(YoutubeShowIE, self)._real_extract(
3330 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3331
3332
3333 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3334 """
3335 Base class for feed extractors
3336 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3337 """
3338 _LOGIN_REQUIRED = True
3339
3340 @property
3341 def IE_NAME(self):
3342 return 'youtube:%s' % self._FEED_NAME
3343
3344 def _real_initialize(self):
3345 self._login()
3346
3347 def _entries(self, page):
3348 # The extraction process is the same as for playlists, but the regex
3349 # for the video ids doesn't contain an index
3350 ids = []
3351 more_widget_html = content_html = page
3352 for page_num in itertools.count(1):
3353 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3354
3355 # 'recommended' feed has infinite 'load more' and each new portion spins
3356 # the same videos in (sometimes) slightly different order, so we'll check
3357 # for unicity and break when portion has no new videos
3358 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3359 if not new_ids:
3360 break
3361
3362 ids.extend(new_ids)
3363
3364 for entry in self._ids_to_results(new_ids):
3365 yield entry
3366
3367 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3368 if not mobj:
3369 break
3370
3371 more = self._download_json(
3372 'https://www.youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3373 'Downloading page #%s' % page_num,
3374 transform_source=uppercase_escape,
3375 headers=self._YOUTUBE_CLIENT_HEADERS)
3376 content_html = more['content_html']
3377 more_widget_html = more['load_more_widget_html']
3378
3379 def _real_extract(self, url):
3380 page = self._download_webpage(
3381 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3382 self._PLAYLIST_TITLE)
3383 return self.playlist_result(
3384 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3385
3386
3387 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3388 IE_NAME = 'youtube:watchlater'
3389 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3390 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3391
3392 _TESTS = [{
3393 'url': 'https://www.youtube.com/playlist?list=WL',
3394 'only_matching': True,
3395 }, {
3396 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3397 'only_matching': True,
3398 }]
3399
3400 def _real_extract(self, url):
3401 _, video = self._check_download_just_video(url, 'WL')
3402 if video:
3403 return video
3404 _, playlist = self._extract_playlist('WL')
3405 return playlist
3406
3407
3408 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3409 IE_NAME = 'youtube:favorites'
3410 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3411 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3412 _LOGIN_REQUIRED = True
3413
3414 def _real_extract(self, url):
3415 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3416 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3417 return self.url_result(playlist_id, 'YoutubePlaylist')
3418
3419
3420 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3421 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3422 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3423 _FEED_NAME = 'recommended'
3424 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3425
3426
3427 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3428 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3429 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3430 _FEED_NAME = 'subscriptions'
3431 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3432
3433
3434 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3435 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3436 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3437 _FEED_NAME = 'history'
3438 _PLAYLIST_TITLE = 'Youtube History'
3439
3440
3441 class YoutubeTruncatedURLIE(InfoExtractor):
3442 IE_NAME = 'youtube:truncated_url'
3443 IE_DESC = False # Do not list
3444 _VALID_URL = r'''(?x)
3445 (?:https?://)?
3446 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3447 (?:watch\?(?:
3448 feature=[a-z_]+|
3449 annotation_id=annotation_[^&]+|
3450 x-yt-cl=[0-9]+|
3451 hl=[^&]*|
3452 t=[0-9]+
3453 )?
3454 |
3455 attribution_link\?a=[^&]+
3456 )
3457 $
3458 '''
3459
3460 _TESTS = [{
3461 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3462 'only_matching': True,
3463 }, {
3464 'url': 'https://www.youtube.com/watch?',
3465 'only_matching': True,
3466 }, {
3467 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3468 'only_matching': True,
3469 }, {
3470 'url': 'https://www.youtube.com/watch?feature=foo',
3471 'only_matching': True,
3472 }, {
3473 'url': 'https://www.youtube.com/watch?hl=en-GB',
3474 'only_matching': True,
3475 }, {
3476 'url': 'https://www.youtube.com/watch?t=2372',
3477 'only_matching': True,
3478 }]
3479
3480 def _real_extract(self, url):
3481 raise ExtractorError(
3482 'Did you forget to quote the URL? Remember that & is a meta '
3483 'character in most shells, so you want to put the URL in quotes, '
3484 'like youtube-dl '
3485 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3486 ' or simply youtube-dl BaW_jenozKc .',
3487 expected=True)
3488
3489
3490 class YoutubeTruncatedIDIE(InfoExtractor):
3491 IE_NAME = 'youtube:truncated_id'
3492 IE_DESC = False # Do not list
3493 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3494
3495 _TESTS = [{
3496 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3497 'only_matching': True,
3498 }]
3499
3500 def _real_extract(self, url):
3501 video_id = self._match_id(url)
3502 raise ExtractorError(
3503 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3504 expected=True)