]> jfr.im git - yt-dlp.git/blob - youtube_dlc/extractor/youtube.py
Merge branch 'ytdl-org-master'
[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 = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt', 'json3')
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
1270 def __init__(self, *args, **kwargs):
1271 super(YoutubeIE, self).__init__(*args, **kwargs)
1272 self._player_cache = {}
1273
1274 def report_video_info_webpage_download(self, video_id):
1275 """Report attempt to download video info webpage."""
1276 self.to_screen('%s: Downloading video info webpage' % video_id)
1277
1278 def report_information_extraction(self, video_id):
1279 """Report attempt to extract video information."""
1280 self.to_screen('%s: Extracting video information' % video_id)
1281
1282 def report_unavailable_format(self, video_id, format):
1283 """Report extracted video URL."""
1284 self.to_screen('%s: Format %s not available' % (video_id, format))
1285
1286 def report_rtmp_download(self):
1287 """Indicate the download will use the RTMP protocol."""
1288 self.to_screen('RTMP download detected')
1289
1290 def _signature_cache_id(self, example_sig):
1291 """ Return a string representation of a signature """
1292 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1293
1294 @classmethod
1295 def _extract_player_info(cls, player_url):
1296 for player_re in cls._PLAYER_INFO_RE:
1297 id_m = re.search(player_re, player_url)
1298 if id_m:
1299 break
1300 else:
1301 raise ExtractorError('Cannot identify player %r' % player_url)
1302 return id_m.group('ext'), id_m.group('id')
1303
1304 def _extract_signature_function(self, video_id, player_url, example_sig):
1305 player_type, player_id = self._extract_player_info(player_url)
1306
1307 # Read from filesystem cache
1308 func_id = '%s_%s_%s' % (
1309 player_type, player_id, self._signature_cache_id(example_sig))
1310 assert os.path.basename(func_id) == func_id
1311
1312 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1313 if cache_spec is not None:
1314 return lambda s: ''.join(s[i] for i in cache_spec)
1315
1316 download_note = (
1317 'Downloading player %s' % player_url
1318 if self._downloader.params.get('verbose') else
1319 'Downloading %s player %s' % (player_type, player_id)
1320 )
1321 if player_type == 'js':
1322 code = self._download_webpage(
1323 player_url, video_id,
1324 note=download_note,
1325 errnote='Download of %s failed' % player_url)
1326 res = self._parse_sig_js(code)
1327 elif player_type == 'swf':
1328 urlh = self._request_webpage(
1329 player_url, video_id,
1330 note=download_note,
1331 errnote='Download of %s failed' % player_url)
1332 code = urlh.read()
1333 res = self._parse_sig_swf(code)
1334 else:
1335 assert False, 'Invalid player type %r' % player_type
1336
1337 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1338 cache_res = res(test_string)
1339 cache_spec = [ord(c) for c in cache_res]
1340
1341 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1342 return res
1343
1344 def _print_sig_code(self, func, example_sig):
1345 def gen_sig_code(idxs):
1346 def _genslice(start, end, step):
1347 starts = '' if start == 0 else str(start)
1348 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1349 steps = '' if step == 1 else (':%d' % step)
1350 return 's[%s%s%s]' % (starts, ends, steps)
1351
1352 step = None
1353 # Quelch pyflakes warnings - start will be set when step is set
1354 start = '(Never used)'
1355 for i, prev in zip(idxs[1:], idxs[:-1]):
1356 if step is not None:
1357 if i - prev == step:
1358 continue
1359 yield _genslice(start, prev, step)
1360 step = None
1361 continue
1362 if i - prev in [-1, 1]:
1363 step = i - prev
1364 start = prev
1365 continue
1366 else:
1367 yield 's[%d]' % prev
1368 if step is None:
1369 yield 's[%d]' % i
1370 else:
1371 yield _genslice(start, i, step)
1372
1373 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1374 cache_res = func(test_string)
1375 cache_spec = [ord(c) for c in cache_res]
1376 expr_code = ' + '.join(gen_sig_code(cache_spec))
1377 signature_id_tuple = '(%s)' % (
1378 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1379 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1380 ' return %s\n') % (signature_id_tuple, expr_code)
1381 self.to_screen('Extracted signature function:\n' + code)
1382
1383 def _parse_sig_js(self, jscode):
1384 funcname = self._search_regex(
1385 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1386 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1387 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\(\s*""\s*\)',
1388 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1389 # Obsolete patterns
1390 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1391 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1392 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1393 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1394 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1395 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1396 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1397 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1398 jscode, 'Initial JS player signature function name', group='sig')
1399
1400 jsi = JSInterpreter(jscode)
1401 initial_function = jsi.extract_function(funcname)
1402 return lambda s: initial_function([s])
1403
1404 def _parse_sig_swf(self, file_contents):
1405 swfi = SWFInterpreter(file_contents)
1406 TARGET_CLASSNAME = 'SignatureDecipher'
1407 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1408 initial_function = swfi.extract_function(searched_class, 'decipher')
1409 return lambda s: initial_function([s])
1410
1411 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1412 """Turn the encrypted s field into a working signature"""
1413
1414 if player_url is None:
1415 raise ExtractorError('Cannot decrypt signature without player_url')
1416
1417 if player_url.startswith('//'):
1418 player_url = 'https:' + player_url
1419 elif not re.match(r'https?://', player_url):
1420 player_url = compat_urlparse.urljoin(
1421 'https://www.youtube.com', player_url)
1422 try:
1423 player_id = (player_url, self._signature_cache_id(s))
1424 if player_id not in self._player_cache:
1425 func = self._extract_signature_function(
1426 video_id, player_url, s
1427 )
1428 self._player_cache[player_id] = func
1429 func = self._player_cache[player_id]
1430 if self._downloader.params.get('youtube_print_sig_code'):
1431 self._print_sig_code(func, s)
1432 return func(s)
1433 except Exception as e:
1434 tb = traceback.format_exc()
1435 raise ExtractorError(
1436 'Signature extraction failed: ' + tb, cause=e)
1437
1438 def _get_subtitles(self, video_id, webpage, has_live_chat_replay):
1439 try:
1440 subs_doc = self._download_xml(
1441 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1442 video_id, note=False)
1443 except ExtractorError as err:
1444 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1445 return {}
1446
1447 sub_lang_list = {}
1448 for track in subs_doc.findall('track'):
1449 lang = track.attrib['lang_code']
1450 if lang in sub_lang_list:
1451 continue
1452 sub_formats = []
1453 for ext in self._SUBTITLE_FORMATS:
1454 params = compat_urllib_parse_urlencode({
1455 'lang': lang,
1456 'v': video_id,
1457 'fmt': ext,
1458 'name': track.attrib['name'].encode('utf-8'),
1459 })
1460 sub_formats.append({
1461 'url': 'https://www.youtube.com/api/timedtext?' + params,
1462 'ext': ext,
1463 })
1464 sub_lang_list[lang] = sub_formats
1465 if has_live_chat_replay:
1466 sub_lang_list['live_chat'] = [
1467 {
1468 'video_id': video_id,
1469 'ext': 'json',
1470 'protocol': 'youtube_live_chat_replay',
1471 },
1472 ]
1473 if not sub_lang_list:
1474 self._downloader.report_warning('video doesn\'t have subtitles')
1475 return {}
1476 return sub_lang_list
1477
1478 def _get_ytplayer_config(self, video_id, webpage):
1479 patterns = (
1480 # User data may contain arbitrary character sequences that may affect
1481 # JSON extraction with regex, e.g. when '};' is contained the second
1482 # regex won't capture the whole JSON. Yet working around by trying more
1483 # concrete regex first keeping in mind proper quoted string handling
1484 # to be implemented in future that will replace this workaround (see
1485 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1486 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1487 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1488 r';ytplayer\.config\s*=\s*({.+?});',
1489 )
1490 config = self._search_regex(
1491 patterns, webpage, 'ytplayer.config', default=None)
1492 if config:
1493 return self._parse_json(
1494 uppercase_escape(config), video_id, fatal=False)
1495
1496 def _get_yt_initial_data(self, video_id, webpage):
1497 config = self._search_regex(
1498 (r'window\["ytInitialData"\]\s*=\s*(.*?)(?<=});',
1499 r'var\s+ytInitialData\s*=\s*(.*?)(?<=});'),
1500 webpage, 'ytInitialData', default=None)
1501 if config:
1502 return self._parse_json(
1503 uppercase_escape(config), video_id, fatal=False)
1504
1505 def _get_automatic_captions(self, video_id, webpage):
1506 """We need the webpage for getting the captions url, pass it as an
1507 argument to speed up the process."""
1508 self.to_screen('%s: Looking for automatic captions' % video_id)
1509 player_config = self._get_ytplayer_config(video_id, webpage)
1510 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1511 if not player_config:
1512 self._downloader.report_warning(err_msg)
1513 return {}
1514 try:
1515 args = player_config['args']
1516 caption_url = args.get('ttsurl')
1517 if caption_url:
1518 timestamp = args['timestamp']
1519 # We get the available subtitles
1520 list_params = compat_urllib_parse_urlencode({
1521 'type': 'list',
1522 'tlangs': 1,
1523 'asrs': 1,
1524 })
1525 list_url = caption_url + '&' + list_params
1526 caption_list = self._download_xml(list_url, video_id)
1527 original_lang_node = caption_list.find('track')
1528 if original_lang_node is None:
1529 self._downloader.report_warning('Video doesn\'t have automatic captions')
1530 return {}
1531 original_lang = original_lang_node.attrib['lang_code']
1532 caption_kind = original_lang_node.attrib.get('kind', '')
1533
1534 sub_lang_list = {}
1535 for lang_node in caption_list.findall('target'):
1536 sub_lang = lang_node.attrib['lang_code']
1537 sub_formats = []
1538 for ext in self._SUBTITLE_FORMATS:
1539 params = compat_urllib_parse_urlencode({
1540 'lang': original_lang,
1541 'tlang': sub_lang,
1542 'fmt': ext,
1543 'ts': timestamp,
1544 'kind': caption_kind,
1545 })
1546 sub_formats.append({
1547 'url': caption_url + '&' + params,
1548 'ext': ext,
1549 })
1550 sub_lang_list[sub_lang] = sub_formats
1551 return sub_lang_list
1552
1553 def make_captions(sub_url, sub_langs):
1554 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1555 caption_qs = compat_parse_qs(parsed_sub_url.query)
1556 captions = {}
1557 for sub_lang in sub_langs:
1558 sub_formats = []
1559 for ext in self._SUBTITLE_FORMATS:
1560 caption_qs.update({
1561 'tlang': [sub_lang],
1562 'fmt': [ext],
1563 })
1564 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1565 query=compat_urllib_parse_urlencode(caption_qs, True)))
1566 sub_formats.append({
1567 'url': sub_url,
1568 'ext': ext,
1569 })
1570 captions[sub_lang] = sub_formats
1571 return captions
1572
1573 # New captions format as of 22.06.2017
1574 player_response = args.get('player_response')
1575 if player_response and isinstance(player_response, compat_str):
1576 player_response = self._parse_json(
1577 player_response, video_id, fatal=False)
1578 if player_response:
1579 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1580 caption_tracks = renderer['captionTracks']
1581 for caption_track in caption_tracks:
1582 if 'kind' not in caption_track:
1583 # not an automatic transcription
1584 continue
1585 base_url = caption_track['baseUrl']
1586 sub_lang_list = []
1587 for lang in renderer['translationLanguages']:
1588 lang_code = lang.get('languageCode')
1589 if lang_code:
1590 sub_lang_list.append(lang_code)
1591 return make_captions(base_url, sub_lang_list)
1592
1593 self._downloader.report_warning("Couldn't find automatic captions for %s" % video_id)
1594 return {}
1595 # Some videos don't provide ttsurl but rather caption_tracks and
1596 # caption_translation_languages (e.g. 20LmZk1hakA)
1597 # Does not used anymore as of 22.06.2017
1598 caption_tracks = args['caption_tracks']
1599 caption_translation_languages = args['caption_translation_languages']
1600 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1601 sub_lang_list = []
1602 for lang in caption_translation_languages.split(','):
1603 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1604 sub_lang = lang_qs.get('lc', [None])[0]
1605 if sub_lang:
1606 sub_lang_list.append(sub_lang)
1607 return make_captions(caption_url, sub_lang_list)
1608 # An extractor error can be raise by the download process if there are
1609 # no automatic captions but there are subtitles
1610 except (KeyError, IndexError, ExtractorError):
1611 self._downloader.report_warning(err_msg)
1612 return {}
1613
1614 def _mark_watched(self, video_id, video_info, player_response):
1615 playback_url = url_or_none(try_get(
1616 player_response,
1617 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1618 video_info, lambda x: x['videostats_playback_base_url'][0]))
1619 if not playback_url:
1620 return
1621 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1622 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1623
1624 # cpn generation algorithm is reverse engineered from base.js.
1625 # In fact it works even with dummy cpn.
1626 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1627 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1628
1629 qs.update({
1630 'ver': ['2'],
1631 'cpn': [cpn],
1632 })
1633 playback_url = compat_urlparse.urlunparse(
1634 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1635
1636 self._download_webpage(
1637 playback_url, video_id, 'Marking watched',
1638 'Unable to mark watched', fatal=False)
1639
1640 @staticmethod
1641 def _extract_urls(webpage):
1642 # Embedded YouTube player
1643 entries = [
1644 unescapeHTML(mobj.group('url'))
1645 for mobj in re.finditer(r'''(?x)
1646 (?:
1647 <iframe[^>]+?src=|
1648 data-video-url=|
1649 <embed[^>]+?src=|
1650 embedSWF\(?:\s*|
1651 <object[^>]+data=|
1652 new\s+SWFObject\(
1653 )
1654 (["\'])
1655 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1656 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1657 \1''', webpage)]
1658
1659 # lazyYT YouTube embed
1660 entries.extend(list(map(
1661 unescapeHTML,
1662 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1663
1664 # Wordpress "YouTube Video Importer" plugin
1665 matches = re.findall(r'''(?x)<div[^>]+
1666 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1667 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1668 entries.extend(m[-1] for m in matches)
1669
1670 return entries
1671
1672 @staticmethod
1673 def _extract_url(webpage):
1674 urls = YoutubeIE._extract_urls(webpage)
1675 return urls[0] if urls else None
1676
1677 @classmethod
1678 def extract_id(cls, url):
1679 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1680 if mobj is None:
1681 raise ExtractorError('Invalid URL: %s' % url)
1682 video_id = mobj.group(2)
1683 return video_id
1684
1685 def _extract_chapters_from_json(self, webpage, video_id, duration):
1686 if not webpage:
1687 return
1688 initial_data = self._parse_json(
1689 self._search_regex(
1690 r'window\["ytInitialData"\] = (.+);\n', webpage,
1691 'player args', default='{}'),
1692 video_id, fatal=False)
1693 if not initial_data or not isinstance(initial_data, dict):
1694 return
1695 chapters_list = try_get(
1696 initial_data,
1697 lambda x: x['playerOverlays']
1698 ['playerOverlayRenderer']
1699 ['decoratedPlayerBarRenderer']
1700 ['decoratedPlayerBarRenderer']
1701 ['playerBar']
1702 ['chapteredPlayerBarRenderer']
1703 ['chapters'],
1704 list)
1705 if not chapters_list:
1706 return
1707
1708 def chapter_time(chapter):
1709 return float_or_none(
1710 try_get(
1711 chapter,
1712 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1713 int),
1714 scale=1000)
1715 chapters = []
1716 for next_num, chapter in enumerate(chapters_list, start=1):
1717 start_time = chapter_time(chapter)
1718 if start_time is None:
1719 continue
1720 end_time = (chapter_time(chapters_list[next_num])
1721 if next_num < len(chapters_list) else duration)
1722 if end_time is None:
1723 continue
1724 title = try_get(
1725 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1726 compat_str)
1727 chapters.append({
1728 'start_time': start_time,
1729 'end_time': end_time,
1730 'title': title,
1731 })
1732 return chapters
1733
1734 @staticmethod
1735 def _extract_chapters_from_description(description, duration):
1736 if not description:
1737 return None
1738 chapter_lines = re.findall(
1739 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1740 description)
1741 if not chapter_lines:
1742 return None
1743 chapters = []
1744 for next_num, (chapter_line, time_point) in enumerate(
1745 chapter_lines, start=1):
1746 start_time = parse_duration(time_point)
1747 if start_time is None:
1748 continue
1749 if start_time > duration:
1750 break
1751 end_time = (duration if next_num == len(chapter_lines)
1752 else parse_duration(chapter_lines[next_num][1]))
1753 if end_time is None:
1754 continue
1755 if end_time > duration:
1756 end_time = duration
1757 if start_time > end_time:
1758 break
1759 chapter_title = re.sub(
1760 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1761 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1762 chapters.append({
1763 'start_time': start_time,
1764 'end_time': end_time,
1765 'title': chapter_title,
1766 })
1767 return chapters
1768
1769 def _extract_chapters(self, webpage, description, video_id, duration):
1770 return (self._extract_chapters_from_json(webpage, video_id, duration)
1771 or self._extract_chapters_from_description(description, duration))
1772
1773 def _real_extract(self, url):
1774 url, smuggled_data = unsmuggle_url(url, {})
1775
1776 proto = (
1777 'http' if self._downloader.params.get('prefer_insecure', False)
1778 else 'https')
1779
1780 start_time = None
1781 end_time = None
1782 parsed_url = compat_urllib_parse_urlparse(url)
1783 for component in [parsed_url.fragment, parsed_url.query]:
1784 query = compat_parse_qs(component)
1785 if start_time is None and 't' in query:
1786 start_time = parse_duration(query['t'][0])
1787 if start_time is None and 'start' in query:
1788 start_time = parse_duration(query['start'][0])
1789 if end_time is None and 'end' in query:
1790 end_time = parse_duration(query['end'][0])
1791
1792 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1793 mobj = re.search(self._NEXT_URL_RE, url)
1794 if mobj:
1795 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1796 video_id = self.extract_id(url)
1797
1798 # Get video webpage
1799 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1800 video_webpage, urlh = self._download_webpage_handle(url, video_id)
1801
1802 qs = compat_parse_qs(compat_urllib_parse_urlparse(urlh.geturl()).query)
1803 video_id = qs.get('v', [None])[0] or video_id
1804
1805 # Attempt to extract SWF player URL
1806 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1807 if mobj is not None:
1808 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1809 else:
1810 player_url = None
1811
1812 dash_mpds = []
1813
1814 def add_dash_mpd(video_info):
1815 dash_mpd = video_info.get('dashmpd')
1816 if dash_mpd and dash_mpd[0] not in dash_mpds:
1817 dash_mpds.append(dash_mpd[0])
1818
1819 def add_dash_mpd_pr(pl_response):
1820 dash_mpd = url_or_none(try_get(
1821 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1822 compat_str))
1823 if dash_mpd and dash_mpd not in dash_mpds:
1824 dash_mpds.append(dash_mpd)
1825
1826 is_live = None
1827 view_count = None
1828
1829 def extract_view_count(v_info):
1830 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1831
1832 def extract_player_response(player_response, video_id):
1833 pl_response = str_or_none(player_response)
1834 if not pl_response:
1835 return
1836 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1837 if isinstance(pl_response, dict):
1838 add_dash_mpd_pr(pl_response)
1839 return pl_response
1840
1841 player_response = {}
1842
1843 # Get video info
1844 video_info = {}
1845 embed_webpage = None
1846 if (self._og_search_property('restrictions:age', video_webpage, default=None) == '18+'
1847 or re.search(r'player-age-gate-content">', video_webpage) is not None):
1848 age_gate = True
1849 # We simulate the access to the video from www.youtube.com/v/{video_id}
1850 # this can be viewed without login into Youtube
1851 url = proto + '://www.youtube.com/embed/%s' % video_id
1852 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1853 data = compat_urllib_parse_urlencode({
1854 'video_id': video_id,
1855 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1856 'sts': self._search_regex(
1857 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1858 })
1859 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1860 try:
1861 video_info_webpage = self._download_webpage(
1862 video_info_url, video_id,
1863 note='Refetching age-gated info webpage',
1864 errnote='unable to download video info webpage')
1865 except ExtractorError:
1866 video_info_webpage = None
1867 if video_info_webpage:
1868 video_info = compat_parse_qs(video_info_webpage)
1869 pl_response = video_info.get('player_response', [None])[0]
1870 player_response = extract_player_response(pl_response, video_id)
1871 add_dash_mpd(video_info)
1872 view_count = extract_view_count(video_info)
1873 else:
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
1896 def extract_unavailable_message():
1897 messages = []
1898 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1899 msg = self._html_search_regex(
1900 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1901 video_webpage, 'unavailable %s' % kind, default=None)
1902 if msg:
1903 messages.append(msg)
1904 if messages:
1905 return '\n'.join(messages)
1906
1907 if not video_info and not player_response:
1908 unavailable_message = extract_unavailable_message()
1909 if not unavailable_message:
1910 unavailable_message = 'Unable to extract video data'
1911 raise ExtractorError(
1912 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1913
1914 if not isinstance(video_info, dict):
1915 video_info = {}
1916
1917 video_details = try_get(
1918 player_response, lambda x: x['videoDetails'], dict) or {}
1919
1920 microformat = try_get(
1921 player_response, lambda x: x['microformat']['playerMicroformatRenderer'], dict) or {}
1922
1923 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1924 if not video_title:
1925 self._downloader.report_warning('Unable to extract video title')
1926 video_title = '_'
1927
1928 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1929 if video_description:
1930
1931 def replace_url(m):
1932 redir_url = compat_urlparse.urljoin(url, m.group(1))
1933 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1934 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1935 qs = compat_parse_qs(parsed_redir_url.query)
1936 q = qs.get('q')
1937 if q and q[0]:
1938 return q[0]
1939 return redir_url
1940
1941 description_original = video_description = re.sub(r'''(?x)
1942 <a\s+
1943 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1944 (?:title|href)="([^"]+)"\s+
1945 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1946 class="[^"]*"[^>]*>
1947 [^<]+\.{3}\s*
1948 </a>
1949 ''', replace_url, video_description)
1950 video_description = clean_html(video_description)
1951 else:
1952 video_description = video_details.get('shortDescription') or self._html_search_meta('description', video_webpage)
1953
1954 if not smuggled_data.get('force_singlefeed', False):
1955 if not self._downloader.params.get('noplaylist'):
1956 multifeed_metadata_list = try_get(
1957 player_response,
1958 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1959 compat_str) or try_get(
1960 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1961 if multifeed_metadata_list:
1962 entries = []
1963 feed_ids = []
1964 for feed in multifeed_metadata_list.split(','):
1965 # Unquote should take place before split on comma (,) since textual
1966 # fields may contain comma as well (see
1967 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1968 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1969
1970 def feed_entry(name):
1971 return try_get(feed_data, lambda x: x[name][0], compat_str)
1972
1973 feed_id = feed_entry('id')
1974 if not feed_id:
1975 continue
1976 feed_title = feed_entry('title')
1977 title = video_title
1978 if feed_title:
1979 title += ' (%s)' % feed_title
1980 entries.append({
1981 '_type': 'url_transparent',
1982 'ie_key': 'Youtube',
1983 'url': smuggle_url(
1984 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1985 {'force_singlefeed': True}),
1986 'title': title,
1987 })
1988 feed_ids.append(feed_id)
1989 self.to_screen(
1990 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1991 % (', '.join(feed_ids), video_id))
1992 return self.playlist_result(entries, video_id, video_title, video_description)
1993 else:
1994 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1995
1996 if view_count is None:
1997 view_count = extract_view_count(video_info)
1998 if view_count is None and video_details:
1999 view_count = int_or_none(video_details.get('viewCount'))
2000 if view_count is None and microformat:
2001 view_count = int_or_none(microformat.get('viewCount'))
2002
2003 if is_live is None:
2004 is_live = bool_or_none(video_details.get('isLive'))
2005
2006 has_live_chat_replay = False
2007 if not is_live:
2008 yt_initial_data = self._get_yt_initial_data(video_id, video_webpage)
2009 try:
2010 yt_initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation']
2011 has_live_chat_replay = True
2012 except (KeyError, IndexError, TypeError):
2013 pass
2014
2015 # Check for "rental" videos
2016 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
2017 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
2018
2019 def _extract_filesize(media_url):
2020 return int_or_none(self._search_regex(
2021 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
2022
2023 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
2024 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
2025
2026 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
2027 self.report_rtmp_download()
2028 formats = [{
2029 'format_id': '_rtmp',
2030 'protocol': 'rtmp',
2031 'url': video_info['conn'][0],
2032 'player_url': player_url,
2033 }]
2034 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):
2035 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
2036 if 'rtmpe%3Dyes' in encoded_url_map:
2037 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
2038 formats = []
2039 formats_spec = {}
2040 fmt_list = video_info.get('fmt_list', [''])[0]
2041 if fmt_list:
2042 for fmt in fmt_list.split(','):
2043 spec = fmt.split('/')
2044 if len(spec) > 1:
2045 width_height = spec[1].split('x')
2046 if len(width_height) == 2:
2047 formats_spec[spec[0]] = {
2048 'resolution': spec[1],
2049 'width': int_or_none(width_height[0]),
2050 'height': int_or_none(width_height[1]),
2051 }
2052 for fmt in streaming_formats:
2053 itag = str_or_none(fmt.get('itag'))
2054 if not itag:
2055 continue
2056 quality = fmt.get('quality')
2057 quality_label = fmt.get('qualityLabel') or quality
2058 formats_spec[itag] = {
2059 'asr': int_or_none(fmt.get('audioSampleRate')),
2060 'filesize': int_or_none(fmt.get('contentLength')),
2061 'format_note': quality_label,
2062 'fps': int_or_none(fmt.get('fps')),
2063 'height': int_or_none(fmt.get('height')),
2064 # bitrate for itag 43 is always 2147483647
2065 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
2066 'width': int_or_none(fmt.get('width')),
2067 }
2068
2069 for fmt in streaming_formats:
2070 if fmt.get('drmFamilies') or fmt.get('drm_families'):
2071 continue
2072 url = url_or_none(fmt.get('url'))
2073
2074 if not url:
2075 cipher = fmt.get('cipher') or fmt.get('signatureCipher')
2076 if not cipher:
2077 continue
2078 url_data = compat_parse_qs(cipher)
2079 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
2080 if not url:
2081 continue
2082 else:
2083 cipher = None
2084 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
2085
2086 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
2087 # Unsupported FORMAT_STREAM_TYPE_OTF
2088 if stream_type == 3:
2089 continue
2090
2091 format_id = fmt.get('itag') or url_data['itag'][0]
2092 if not format_id:
2093 continue
2094 format_id = compat_str(format_id)
2095
2096 if cipher:
2097 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
2098 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
2099 jsplayer_url_json = self._search_regex(
2100 ASSETS_RE,
2101 embed_webpage if age_gate else video_webpage,
2102 'JS player URL (1)', default=None)
2103 if not jsplayer_url_json and not age_gate:
2104 # We need the embed website after all
2105 if embed_webpage is None:
2106 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
2107 embed_webpage = self._download_webpage(
2108 embed_url, video_id, 'Downloading embed webpage')
2109 jsplayer_url_json = self._search_regex(
2110 ASSETS_RE, embed_webpage, 'JS player URL')
2111
2112 player_url = json.loads(jsplayer_url_json)
2113 if player_url is None:
2114 player_url_json = self._search_regex(
2115 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2116 video_webpage, 'age gate player URL')
2117 player_url = json.loads(player_url_json)
2118
2119 if 'sig' in url_data:
2120 url += '&signature=' + url_data['sig'][0]
2121 elif 's' in url_data:
2122 encrypted_sig = url_data['s'][0]
2123
2124 if self._downloader.params.get('verbose'):
2125 if player_url is None:
2126 player_desc = 'unknown'
2127 else:
2128 player_type, player_version = self._extract_player_info(player_url)
2129 player_desc = '%s player %s' % ('flash' if player_type == 'swf' else 'html5', player_version)
2130 parts_sizes = self._signature_cache_id(encrypted_sig)
2131 self.to_screen('{%s} signature length %s, %s' %
2132 (format_id, parts_sizes, player_desc))
2133
2134 signature = self._decrypt_signature(
2135 encrypted_sig, video_id, player_url, age_gate)
2136 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2137 url += '&%s=%s' % (sp, signature)
2138 if 'ratebypass' not in url:
2139 url += '&ratebypass=yes'
2140
2141 dct = {
2142 'format_id': format_id,
2143 'url': url,
2144 'player_url': player_url,
2145 }
2146 if format_id in self._formats:
2147 dct.update(self._formats[format_id])
2148 if format_id in formats_spec:
2149 dct.update(formats_spec[format_id])
2150
2151 # Some itags are not included in DASH manifest thus corresponding formats will
2152 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2153 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2154 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2155 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2156
2157 if width is None:
2158 width = int_or_none(fmt.get('width'))
2159 if height is None:
2160 height = int_or_none(fmt.get('height'))
2161
2162 filesize = int_or_none(url_data.get(
2163 'clen', [None])[0]) or _extract_filesize(url)
2164
2165 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2166 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2167
2168 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2169 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2170 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2171
2172 more_fields = {
2173 'filesize': filesize,
2174 'tbr': tbr,
2175 'width': width,
2176 'height': height,
2177 'fps': fps,
2178 'format_note': quality_label or quality,
2179 }
2180 for key, value in more_fields.items():
2181 if value:
2182 dct[key] = value
2183 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2184 if type_:
2185 type_split = type_.split(';')
2186 kind_ext = type_split[0].split('/')
2187 if len(kind_ext) == 2:
2188 kind, _ = kind_ext
2189 dct['ext'] = mimetype2ext(type_split[0])
2190 if kind in ('audio', 'video'):
2191 codecs = None
2192 for mobj in re.finditer(
2193 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2194 if mobj.group('key') == 'codecs':
2195 codecs = mobj.group('val')
2196 break
2197 if codecs:
2198 dct.update(parse_codecs(codecs))
2199 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2200 dct['downloader_options'] = {
2201 # Youtube throttles chunks >~10M
2202 'http_chunk_size': 10485760,
2203 }
2204 formats.append(dct)
2205 else:
2206 manifest_url = (
2207 url_or_none(try_get(
2208 player_response,
2209 lambda x: x['streamingData']['hlsManifestUrl'],
2210 compat_str))
2211 or url_or_none(try_get(
2212 video_info, lambda x: x['hlsvp'][0], compat_str)))
2213 if manifest_url:
2214 formats = []
2215 m3u8_formats = self._extract_m3u8_formats(
2216 manifest_url, video_id, 'mp4', fatal=False)
2217 for a_format in m3u8_formats:
2218 itag = self._search_regex(
2219 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2220 if itag:
2221 a_format['format_id'] = itag
2222 if itag in self._formats:
2223 dct = self._formats[itag].copy()
2224 dct.update(a_format)
2225 a_format = dct
2226 a_format['player_url'] = player_url
2227 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2228 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2229 formats.append(a_format)
2230 else:
2231 error_message = extract_unavailable_message()
2232 if not error_message:
2233 error_message = clean_html(try_get(
2234 player_response, lambda x: x['playabilityStatus']['reason'],
2235 compat_str))
2236 if not error_message:
2237 error_message = clean_html(
2238 try_get(video_info, lambda x: x['reason'][0], compat_str))
2239 if error_message:
2240 raise ExtractorError(error_message, expected=True)
2241 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2242
2243 # uploader
2244 video_uploader = try_get(
2245 video_info, lambda x: x['author'][0],
2246 compat_str) or str_or_none(video_details.get('author'))
2247 if video_uploader:
2248 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2249 else:
2250 self._downloader.report_warning('unable to extract uploader name')
2251
2252 # uploader_id
2253 video_uploader_id = None
2254 video_uploader_url = None
2255 mobj = re.search(
2256 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2257 video_webpage)
2258 if mobj is not None:
2259 video_uploader_id = mobj.group('uploader_id')
2260 video_uploader_url = mobj.group('uploader_url')
2261 else:
2262 owner_profile_url = url_or_none(microformat.get('ownerProfileUrl'))
2263 if owner_profile_url:
2264 video_uploader_id = self._search_regex(
2265 r'(?:user|channel)/([^/]+)', owner_profile_url, 'uploader id',
2266 default=None)
2267 video_uploader_url = owner_profile_url
2268
2269 channel_id = (
2270 str_or_none(video_details.get('channelId'))
2271 or self._html_search_meta(
2272 'channelId', video_webpage, 'channel id', default=None)
2273 or self._search_regex(
2274 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2275 video_webpage, 'channel id', default=None, group='id'))
2276 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2277
2278 thumbnails = []
2279 thumbnails_list = try_get(
2280 video_details, lambda x: x['thumbnail']['thumbnails'], list) or []
2281 for t in thumbnails_list:
2282 if not isinstance(t, dict):
2283 continue
2284 thumbnail_url = url_or_none(t.get('url'))
2285 if not thumbnail_url:
2286 continue
2287 thumbnails.append({
2288 'url': thumbnail_url,
2289 'width': int_or_none(t.get('width')),
2290 'height': int_or_none(t.get('height')),
2291 })
2292
2293 if not thumbnails:
2294 video_thumbnail = None
2295 # We try first to get a high quality image:
2296 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2297 video_webpage, re.DOTALL)
2298 if m_thumb is not None:
2299 video_thumbnail = m_thumb.group(1)
2300 thumbnail_url = try_get(video_info, lambda x: x['thumbnail_url'][0], compat_str)
2301 if thumbnail_url:
2302 video_thumbnail = compat_urllib_parse_unquote_plus(thumbnail_url)
2303 if video_thumbnail:
2304 thumbnails.append({'url': video_thumbnail})
2305
2306 # upload date
2307 upload_date = self._html_search_meta(
2308 'datePublished', video_webpage, 'upload date', default=None)
2309 if not upload_date:
2310 upload_date = self._search_regex(
2311 [r'(?s)id="eow-date.*?>(.*?)</span>',
2312 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2313 video_webpage, 'upload date', default=None)
2314 if not upload_date:
2315 upload_date = microformat.get('publishDate') or microformat.get('uploadDate')
2316 upload_date = unified_strdate(upload_date)
2317
2318 video_license = self._html_search_regex(
2319 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2320 video_webpage, 'license', default=None)
2321
2322 m_music = re.search(
2323 r'''(?x)
2324 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2325 <ul[^>]*>\s*
2326 <li>(?P<title>.+?)
2327 by (?P<creator>.+?)
2328 (?:
2329 \(.+?\)|
2330 <a[^>]*
2331 (?:
2332 \bhref=["\']/red[^>]*>| # drop possible
2333 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2334 )
2335 .*?
2336 )?</li
2337 ''',
2338 video_webpage)
2339 if m_music:
2340 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2341 video_creator = clean_html(m_music.group('creator'))
2342 else:
2343 video_alt_title = video_creator = None
2344
2345 def extract_meta(field):
2346 return self._html_search_regex(
2347 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2348 video_webpage, field, default=None)
2349
2350 track = extract_meta('Song')
2351 artist = extract_meta('Artist')
2352 album = extract_meta('Album')
2353
2354 # Youtube Music Auto-generated description
2355 release_date = release_year = None
2356 if video_description:
2357 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)
2358 if mobj:
2359 if not track:
2360 track = mobj.group('track').strip()
2361 if not artist:
2362 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2363 if not album:
2364 album = mobj.group('album'.strip())
2365 release_year = mobj.group('release_year')
2366 release_date = mobj.group('release_date')
2367 if release_date:
2368 release_date = release_date.replace('-', '')
2369 if not release_year:
2370 release_year = int(release_date[:4])
2371 if release_year:
2372 release_year = int(release_year)
2373
2374 m_episode = re.search(
2375 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2376 video_webpage)
2377 if m_episode:
2378 series = unescapeHTML(m_episode.group('series'))
2379 season_number = int(m_episode.group('season'))
2380 episode_number = int(m_episode.group('episode'))
2381 else:
2382 series = season_number = episode_number = None
2383
2384 m_cat_container = self._search_regex(
2385 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2386 video_webpage, 'categories', default=None)
2387 category = None
2388 if m_cat_container:
2389 category = self._html_search_regex(
2390 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2391 default=None)
2392 if not category:
2393 category = try_get(
2394 microformat, lambda x: x['category'], compat_str)
2395 video_categories = None if category is None else [category]
2396
2397 video_tags = [
2398 unescapeHTML(m.group('content'))
2399 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2400 if not video_tags:
2401 video_tags = try_get(video_details, lambda x: x['keywords'], list)
2402
2403 def _extract_count(count_name):
2404 return str_to_int(self._search_regex(
2405 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2406 % re.escape(count_name),
2407 video_webpage, count_name, default=None))
2408
2409 like_count = _extract_count('like')
2410 dislike_count = _extract_count('dislike')
2411
2412 if view_count is None:
2413 view_count = str_to_int(self._search_regex(
2414 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2415 'view count', default=None))
2416
2417 average_rating = (
2418 float_or_none(video_details.get('averageRating'))
2419 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2420
2421 # subtitles
2422 video_subtitles = self.extract_subtitles(
2423 video_id, video_webpage, has_live_chat_replay)
2424 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2425
2426 video_duration = try_get(
2427 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2428 if not video_duration:
2429 video_duration = int_or_none(video_details.get('lengthSeconds'))
2430 if not video_duration:
2431 video_duration = parse_duration(self._html_search_meta(
2432 'duration', video_webpage, 'video duration'))
2433
2434 # annotations
2435 video_annotations = None
2436 if self._downloader.params.get('writeannotations', False):
2437 xsrf_token = self._search_regex(
2438 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2439 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2440 invideo_url = try_get(
2441 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2442 if xsrf_token and invideo_url:
2443 xsrf_field_name = self._search_regex(
2444 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2445 video_webpage, 'xsrf field name',
2446 group='xsrf_field_name', default='session_token')
2447 video_annotations = self._download_webpage(
2448 self._proto_relative_url(invideo_url),
2449 video_id, note='Downloading annotations',
2450 errnote='Unable to download video annotations', fatal=False,
2451 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2452
2453 chapters = self._extract_chapters(video_webpage, description_original, video_id, video_duration)
2454
2455 # Look for the DASH manifest
2456 if self._downloader.params.get('youtube_include_dash_manifest', True):
2457 dash_mpd_fatal = True
2458 for mpd_url in dash_mpds:
2459 dash_formats = {}
2460 try:
2461 def decrypt_sig(mobj):
2462 s = mobj.group(1)
2463 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2464 return '/signature/%s' % dec_s
2465
2466 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2467
2468 for df in self._extract_mpd_formats(
2469 mpd_url, video_id, fatal=dash_mpd_fatal,
2470 formats_dict=self._formats):
2471 if not df.get('filesize'):
2472 df['filesize'] = _extract_filesize(df['url'])
2473 # Do not overwrite DASH format found in some previous DASH manifest
2474 if df['format_id'] not in dash_formats:
2475 dash_formats[df['format_id']] = df
2476 # Additional DASH manifests may end up in HTTP Error 403 therefore
2477 # allow them to fail without bug report message if we already have
2478 # some DASH manifest succeeded. This is temporary workaround to reduce
2479 # burst of bug reports until we figure out the reason and whether it
2480 # can be fixed at all.
2481 dash_mpd_fatal = False
2482 except (ExtractorError, KeyError) as e:
2483 self.report_warning(
2484 'Skipping DASH manifest: %r' % e, video_id)
2485 if dash_formats:
2486 # Remove the formats we found through non-DASH, they
2487 # contain less info and it can be wrong, because we use
2488 # fixed values (for example the resolution). See
2489 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2490 # example.
2491 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2492 formats.extend(dash_formats.values())
2493
2494 # Check for malformed aspect ratio
2495 stretched_m = re.search(
2496 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2497 video_webpage)
2498 if stretched_m:
2499 w = float(stretched_m.group('w'))
2500 h = float(stretched_m.group('h'))
2501 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2502 # We will only process correct ratios.
2503 if w > 0 and h > 0:
2504 ratio = w / h
2505 for f in formats:
2506 if f.get('vcodec') != 'none':
2507 f['stretched_ratio'] = ratio
2508
2509 if not formats:
2510 if 'reason' in video_info:
2511 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2512 regions_allowed = self._html_search_meta(
2513 'regionsAllowed', video_webpage, default=None)
2514 countries = regions_allowed.split(',') if regions_allowed else None
2515 self.raise_geo_restricted(
2516 msg=video_info['reason'][0], countries=countries)
2517 reason = video_info['reason'][0]
2518 if 'Invalid parameters' in reason:
2519 unavailable_message = extract_unavailable_message()
2520 if unavailable_message:
2521 reason = unavailable_message
2522 raise ExtractorError(
2523 'YouTube said: %s' % reason,
2524 expected=True, video_id=video_id)
2525 if video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos']):
2526 raise ExtractorError('This video is DRM protected.', expected=True)
2527
2528 self._sort_formats(formats)
2529
2530 self.mark_watched(video_id, video_info, player_response)
2531
2532 return {
2533 'id': video_id,
2534 'uploader': video_uploader,
2535 'uploader_id': video_uploader_id,
2536 'uploader_url': video_uploader_url,
2537 'channel_id': channel_id,
2538 'channel_url': channel_url,
2539 'upload_date': upload_date,
2540 'license': video_license,
2541 'creator': video_creator or artist,
2542 'title': video_title,
2543 'alt_title': video_alt_title or track,
2544 'thumbnails': thumbnails,
2545 'description': video_description,
2546 'categories': video_categories,
2547 'tags': video_tags,
2548 'subtitles': video_subtitles,
2549 'automatic_captions': automatic_captions,
2550 'duration': video_duration,
2551 'age_limit': 18 if age_gate else 0,
2552 'annotations': video_annotations,
2553 'chapters': chapters,
2554 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2555 'view_count': view_count,
2556 'like_count': like_count,
2557 'dislike_count': dislike_count,
2558 'average_rating': average_rating,
2559 'formats': formats,
2560 'is_live': is_live,
2561 'start_time': start_time,
2562 'end_time': end_time,
2563 'series': series,
2564 'season_number': season_number,
2565 'episode_number': episode_number,
2566 'track': track,
2567 'artist': artist,
2568 'album': album,
2569 'release_date': release_date,
2570 'release_year': release_year,
2571 }
2572
2573
2574 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2575 IE_DESC = 'YouTube.com playlists'
2576 _VALID_URL = r"""(?x)(?:
2577 (?:https?://)?
2578 (?:\w+\.)?
2579 (?:
2580 (?:
2581 youtube(?:kids)?\.com|
2582 invidio\.us
2583 )
2584 /
2585 (?:
2586 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2587 \? (?:.*?[&;])*? (?:p|a|list)=
2588 | p/
2589 )|
2590 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2591 )
2592 (
2593 (?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2594 # Top tracks, they can also include dots
2595 |(?:MC)[\w\.]*
2596 )
2597 .*
2598 |
2599 (%(playlist_id)s)
2600 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2601 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2602 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2603 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2604 IE_NAME = 'youtube:playlist'
2605 _TESTS = [{
2606 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2607 'info_dict': {
2608 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2609 'uploader': 'Sergey M.',
2610 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2611 'title': 'youtube-dl public playlist',
2612 },
2613 'playlist_count': 1,
2614 }, {
2615 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2616 'info_dict': {
2617 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2618 'uploader': 'Sergey M.',
2619 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2620 'title': 'youtube-dl empty playlist',
2621 },
2622 'playlist_count': 0,
2623 }, {
2624 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2625 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2626 'info_dict': {
2627 'title': '29C3: Not my department',
2628 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2629 'uploader': 'Christiaan008',
2630 'uploader_id': 'ChRiStIaAn008',
2631 },
2632 'playlist_count': 96,
2633 }, {
2634 'note': 'issue #673',
2635 'url': 'PLBB231211A4F62143',
2636 'info_dict': {
2637 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2638 'id': 'PLBB231211A4F62143',
2639 'uploader': 'Wickydoo',
2640 'uploader_id': 'Wickydoo',
2641 },
2642 'playlist_mincount': 26,
2643 }, {
2644 'note': 'Large playlist',
2645 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2646 'info_dict': {
2647 'title': 'Uploads from Cauchemar',
2648 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2649 'uploader': 'Cauchemar',
2650 'uploader_id': 'Cauchemar89',
2651 },
2652 'playlist_mincount': 799,
2653 }, {
2654 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2655 'info_dict': {
2656 'title': 'YDL_safe_search',
2657 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2658 },
2659 'playlist_count': 2,
2660 'skip': 'This playlist is private',
2661 }, {
2662 'note': 'embedded',
2663 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2664 'playlist_count': 4,
2665 'info_dict': {
2666 'title': 'JODA15',
2667 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2668 'uploader': 'milan',
2669 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2670 }
2671 }, {
2672 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2673 'playlist_mincount': 485,
2674 'info_dict': {
2675 'title': '2018 Chinese New Singles (11/6 updated)',
2676 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2677 'uploader': 'LBK',
2678 'uploader_id': 'sdragonfang',
2679 }
2680 }, {
2681 'note': 'Embedded SWF player',
2682 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2683 'playlist_count': 4,
2684 'info_dict': {
2685 'title': 'JODA7',
2686 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2687 },
2688 'skip': 'This playlist does not exist',
2689 }, {
2690 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2691 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2692 'info_dict': {
2693 'title': 'Uploads from Interstellar Movie',
2694 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2695 'uploader': 'Interstellar Movie',
2696 'uploader_id': 'InterstellarMovie1',
2697 },
2698 'playlist_mincount': 21,
2699 }, {
2700 # Playlist URL that does not actually serve a playlist
2701 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2702 'info_dict': {
2703 'id': 'FqZTN594JQw',
2704 'ext': 'webm',
2705 'title': "Smiley's People 01 detective, Adventure Series, Action",
2706 'uploader': 'STREEM',
2707 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2708 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2709 'upload_date': '20150526',
2710 'license': 'Standard YouTube License',
2711 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2712 'categories': ['People & Blogs'],
2713 'tags': list,
2714 'view_count': int,
2715 'like_count': int,
2716 'dislike_count': int,
2717 },
2718 'params': {
2719 'skip_download': True,
2720 },
2721 'skip': 'This video is not available.',
2722 'add_ie': [YoutubeIE.ie_key()],
2723 }, {
2724 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2725 'info_dict': {
2726 'id': 'yeWKywCrFtk',
2727 'ext': 'mp4',
2728 'title': 'Small Scale Baler and Braiding Rugs',
2729 'uploader': 'Backus-Page House Museum',
2730 'uploader_id': 'backuspagemuseum',
2731 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2732 'upload_date': '20161008',
2733 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2734 'categories': ['Nonprofits & Activism'],
2735 'tags': list,
2736 'like_count': int,
2737 'dislike_count': int,
2738 },
2739 'params': {
2740 'noplaylist': True,
2741 'skip_download': True,
2742 },
2743 }, {
2744 # https://github.com/ytdl-org/youtube-dl/issues/21844
2745 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2746 'info_dict': {
2747 'title': 'Data Analysis with Dr Mike Pound',
2748 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2749 'uploader_id': 'Computerphile',
2750 'uploader': 'Computerphile',
2751 },
2752 'playlist_mincount': 11,
2753 }, {
2754 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2755 'only_matching': True,
2756 }, {
2757 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2758 'only_matching': True,
2759 }, {
2760 # music album playlist
2761 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2762 'only_matching': True,
2763 }, {
2764 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2765 'only_matching': True,
2766 }, {
2767 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2768 'only_matching': True,
2769 }]
2770
2771 def _real_initialize(self):
2772 self._login()
2773
2774 def extract_videos_from_page(self, page):
2775 ids_in_page = []
2776 titles_in_page = []
2777
2778 for item in re.findall(
2779 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2780 attrs = extract_attributes(item)
2781 video_id = attrs['data-video-id']
2782 video_title = unescapeHTML(attrs.get('data-title'))
2783 if video_title:
2784 video_title = video_title.strip()
2785 ids_in_page.append(video_id)
2786 titles_in_page.append(video_title)
2787
2788 # Fallback with old _VIDEO_RE
2789 self.extract_videos_from_page_impl(
2790 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2791
2792 # Relaxed fallbacks
2793 self.extract_videos_from_page_impl(
2794 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2795 ids_in_page, titles_in_page)
2796 self.extract_videos_from_page_impl(
2797 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2798 ids_in_page, titles_in_page)
2799
2800 return zip(ids_in_page, titles_in_page)
2801
2802 def _extract_mix(self, playlist_id):
2803 # The mixes are generated from a single video
2804 # the id of the playlist is just 'RD' + video_id
2805 ids = []
2806 last_id = playlist_id[-11:]
2807 for n in itertools.count(1):
2808 url = 'https://www.youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2809 webpage = self._download_webpage(
2810 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2811 new_ids = orderedSet(re.findall(
2812 r'''(?xs)data-video-username=".*?".*?
2813 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2814 webpage))
2815 # Fetch new pages until all the videos are repeated, it seems that
2816 # there are always 51 unique videos.
2817 new_ids = [_id for _id in new_ids if _id not in ids]
2818 if not new_ids:
2819 break
2820 ids.extend(new_ids)
2821 last_id = ids[-1]
2822
2823 url_results = self._ids_to_results(ids)
2824
2825 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2826 title_span = (
2827 search_title('playlist-title')
2828 or search_title('title long-title')
2829 or search_title('title'))
2830 title = clean_html(title_span)
2831
2832 return self.playlist_result(url_results, playlist_id, title)
2833
2834 def _extract_playlist(self, playlist_id):
2835 url = self._TEMPLATE_URL % playlist_id
2836 page = self._download_webpage(url, playlist_id)
2837
2838 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2839 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2840 match = match.strip()
2841 # Check if the playlist exists or is private
2842 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2843 if mobj:
2844 reason = mobj.group('reason')
2845 message = 'This playlist %s' % reason
2846 if 'private' in reason:
2847 message += ', use --username or --netrc to access it'
2848 message += '.'
2849 raise ExtractorError(message, expected=True)
2850 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2851 raise ExtractorError(
2852 'Invalid parameters. Maybe URL is incorrect.',
2853 expected=True)
2854 elif re.match(r'[^<]*Choose your language[^<]*', match):
2855 continue
2856 else:
2857 self.report_warning('Youtube gives an alert message: ' + match)
2858
2859 playlist_title = self._html_search_regex(
2860 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2861 page, 'title', default=None)
2862
2863 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2864 uploader = self._html_search_regex(
2865 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2866 page, 'uploader', default=None)
2867 mobj = re.search(
2868 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2869 page)
2870 if mobj:
2871 uploader_id = mobj.group('uploader_id')
2872 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2873 else:
2874 uploader_id = uploader_url = None
2875
2876 has_videos = True
2877
2878 if not playlist_title:
2879 try:
2880 # Some playlist URLs don't actually serve a playlist (e.g.
2881 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2882 next(self._entries(page, playlist_id))
2883 except StopIteration:
2884 has_videos = False
2885
2886 playlist = self.playlist_result(
2887 self._entries(page, playlist_id), playlist_id, playlist_title)
2888 playlist.update({
2889 'uploader': uploader,
2890 'uploader_id': uploader_id,
2891 'uploader_url': uploader_url,
2892 })
2893
2894 return has_videos, playlist
2895
2896 def _check_download_just_video(self, url, playlist_id):
2897 # Check if it's a video-specific URL
2898 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2899 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2900 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2901 'video id', default=None)
2902 if video_id:
2903 if self._downloader.params.get('noplaylist'):
2904 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2905 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2906 else:
2907 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2908 return video_id, None
2909 return None, None
2910
2911 def _real_extract(self, url):
2912 # Extract playlist id
2913 mobj = re.match(self._VALID_URL, url)
2914 if mobj is None:
2915 raise ExtractorError('Invalid URL: %s' % url)
2916 playlist_id = mobj.group(1) or mobj.group(2)
2917
2918 video_id, video = self._check_download_just_video(url, playlist_id)
2919 if video:
2920 return video
2921
2922 if playlist_id.startswith(('RD', 'UL', 'PU')):
2923 # Mixes require a custom extraction process
2924 return self._extract_mix(playlist_id)
2925
2926 has_videos, playlist = self._extract_playlist(playlist_id)
2927 if has_videos or not video_id:
2928 return playlist
2929
2930 # Some playlist URLs don't actually serve a playlist (see
2931 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2932 # Fallback to plain video extraction if there is a video id
2933 # along with playlist id.
2934 return self.url_result(video_id, 'Youtube', video_id=video_id)
2935
2936
2937 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2938 IE_DESC = 'YouTube.com channels'
2939 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie|kids)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2940 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2941 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2942 IE_NAME = 'youtube:channel'
2943 _TESTS = [{
2944 'note': 'paginated channel',
2945 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2946 'playlist_mincount': 91,
2947 'info_dict': {
2948 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2949 'title': 'Uploads from lex will',
2950 'uploader': 'lex will',
2951 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2952 }
2953 }, {
2954 'note': 'Age restricted channel',
2955 # from https://www.youtube.com/user/DeusExOfficial
2956 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2957 'playlist_mincount': 64,
2958 'info_dict': {
2959 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2960 'title': 'Uploads from Deus Ex',
2961 'uploader': 'Deus Ex',
2962 'uploader_id': 'DeusExOfficial',
2963 },
2964 }, {
2965 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2966 'only_matching': True,
2967 }, {
2968 'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA',
2969 'only_matching': True,
2970 }]
2971
2972 @classmethod
2973 def suitable(cls, url):
2974 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2975 else super(YoutubeChannelIE, cls).suitable(url))
2976
2977 def _build_template_url(self, url, channel_id):
2978 return self._TEMPLATE_URL % channel_id
2979
2980 def _real_extract(self, url):
2981 channel_id = self._match_id(url)
2982
2983 url = self._build_template_url(url, channel_id)
2984
2985 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2986 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2987 # otherwise fallback on channel by page extraction
2988 channel_page = self._download_webpage(
2989 url + '?view=57', channel_id,
2990 'Downloading channel page', fatal=False)
2991 if channel_page is False:
2992 channel_playlist_id = False
2993 else:
2994 channel_playlist_id = self._html_search_meta(
2995 'channelId', channel_page, 'channel id', default=None)
2996 if not channel_playlist_id:
2997 channel_url = self._html_search_meta(
2998 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2999 channel_page, 'channel url', default=None)
3000 if channel_url:
3001 channel_playlist_id = self._search_regex(
3002 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
3003 channel_url, 'channel id', default=None)
3004 if channel_playlist_id and channel_playlist_id.startswith('UC'):
3005 playlist_id = 'UU' + channel_playlist_id[2:]
3006 return self.url_result(
3007 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
3008
3009 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
3010 autogenerated = re.search(r'''(?x)
3011 class="[^"]*?(?:
3012 channel-header-autogenerated-label|
3013 yt-channel-title-autogenerated
3014 )[^"]*"''', channel_page) is not None
3015
3016 if autogenerated:
3017 # The videos are contained in a single page
3018 # the ajax pages can't be used, they are empty
3019 entries = [
3020 self.url_result(
3021 video_id, 'Youtube', video_id=video_id,
3022 video_title=video_title)
3023 for video_id, video_title in self.extract_videos_from_page(channel_page)]
3024 return self.playlist_result(entries, channel_id)
3025
3026 try:
3027 next(self._entries(channel_page, channel_id))
3028 except StopIteration:
3029 alert_message = self._html_search_regex(
3030 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
3031 channel_page, 'alert', default=None, group='alert')
3032 if alert_message:
3033 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
3034
3035 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
3036
3037
3038 class YoutubeUserIE(YoutubeChannelIE):
3039 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
3040 _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_%-]+)'
3041 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
3042 IE_NAME = 'youtube:user'
3043
3044 _TESTS = [{
3045 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
3046 'playlist_mincount': 320,
3047 'info_dict': {
3048 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
3049 'title': 'Uploads from The Linux Foundation',
3050 'uploader': 'The Linux Foundation',
3051 'uploader_id': 'TheLinuxFoundation',
3052 }
3053 }, {
3054 # Only available via https://www.youtube.com/c/12minuteathlete/videos
3055 # but not https://www.youtube.com/user/12minuteathlete/videos
3056 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
3057 'playlist_mincount': 249,
3058 'info_dict': {
3059 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
3060 'title': 'Uploads from 12 Minute Athlete',
3061 'uploader': '12 Minute Athlete',
3062 'uploader_id': 'the12minuteathlete',
3063 }
3064 }, {
3065 'url': 'ytuser:phihag',
3066 'only_matching': True,
3067 }, {
3068 'url': 'https://www.youtube.com/c/gametrailers',
3069 'only_matching': True,
3070 }, {
3071 'url': 'https://www.youtube.com/c/Pawe%C5%82Zadro%C5%BCniak',
3072 'only_matching': True,
3073 }, {
3074 'url': 'https://www.youtube.com/gametrailers',
3075 'only_matching': True,
3076 }, {
3077 # This channel is not available, geo restricted to JP
3078 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
3079 'only_matching': True,
3080 }]
3081
3082 @classmethod
3083 def suitable(cls, url):
3084 # Don't return True if the url can be extracted with other youtube
3085 # extractor, the regex would is too permissive and it would match.
3086 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
3087 if any(ie.suitable(url) for ie in other_yt_ies):
3088 return False
3089 else:
3090 return super(YoutubeUserIE, cls).suitable(url)
3091
3092 def _build_template_url(self, url, channel_id):
3093 mobj = re.match(self._VALID_URL, url)
3094 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
3095
3096
3097 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
3098 IE_DESC = 'YouTube.com live streams'
3099 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
3100 IE_NAME = 'youtube:live'
3101
3102 _TESTS = [{
3103 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
3104 'info_dict': {
3105 'id': 'a48o2S1cPoo',
3106 'ext': 'mp4',
3107 'title': 'The Young Turks - Live Main Show',
3108 'uploader': 'The Young Turks',
3109 'uploader_id': 'TheYoungTurks',
3110 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
3111 'upload_date': '20150715',
3112 'license': 'Standard YouTube License',
3113 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
3114 'categories': ['News & Politics'],
3115 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
3116 'like_count': int,
3117 'dislike_count': int,
3118 },
3119 'params': {
3120 'skip_download': True,
3121 },
3122 }, {
3123 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
3124 'only_matching': True,
3125 }, {
3126 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3127 'only_matching': True,
3128 }, {
3129 'url': 'https://www.youtube.com/TheYoungTurks/live',
3130 'only_matching': True,
3131 }]
3132
3133 def _real_extract(self, url):
3134 mobj = re.match(self._VALID_URL, url)
3135 channel_id = mobj.group('id')
3136 base_url = mobj.group('base_url')
3137 webpage = self._download_webpage(url, channel_id, fatal=False)
3138 if webpage:
3139 page_type = self._og_search_property(
3140 'type', webpage, 'page type', default='')
3141 video_id = self._html_search_meta(
3142 'videoId', webpage, 'video id', default=None)
3143 if page_type.startswith('video') and video_id and re.match(
3144 r'^[0-9A-Za-z_-]{11}$', video_id):
3145 return self.url_result(video_id, YoutubeIE.ie_key())
3146 return self.url_result(base_url)
3147
3148
3149 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
3150 IE_DESC = 'YouTube.com user/channel playlists'
3151 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel|c)/(?P<id>[^/]+)/playlists'
3152 IE_NAME = 'youtube:playlists'
3153
3154 _TESTS = [{
3155 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3156 'playlist_mincount': 4,
3157 'info_dict': {
3158 'id': 'ThirstForScience',
3159 'title': 'ThirstForScience',
3160 },
3161 }, {
3162 # with "Load more" button
3163 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3164 'playlist_mincount': 70,
3165 'info_dict': {
3166 'id': 'igorkle1',
3167 'title': 'Игорь Клейнер',
3168 },
3169 }, {
3170 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3171 'playlist_mincount': 17,
3172 'info_dict': {
3173 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3174 'title': 'Chem Player',
3175 },
3176 'skip': 'Blocked',
3177 }, {
3178 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists',
3179 'only_matching': True,
3180 }]
3181
3182
3183 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3184 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3185
3186
3187 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3188 IE_DESC = 'YouTube.com searches'
3189 # there doesn't appear to be a real limit, for example if you search for
3190 # 'python' you get more than 8.000.000 results
3191 _MAX_RESULTS = float('inf')
3192 IE_NAME = 'youtube:search'
3193 _SEARCH_KEY = 'ytsearch'
3194 _EXTRA_QUERY_ARGS = {}
3195 _TESTS = []
3196
3197 def _get_n_results(self, query, n):
3198 """Get a specified number of results for a query"""
3199
3200 videos = []
3201 limit = n
3202
3203 url_query = {
3204 'search_query': query.encode('utf-8'),
3205 }
3206 url_query.update(self._EXTRA_QUERY_ARGS)
3207 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3208
3209 for pagenum in itertools.count(1):
3210 data = self._download_json(
3211 result_url, video_id='query "%s"' % query,
3212 note='Downloading page %s' % pagenum,
3213 errnote='Unable to download API page',
3214 query={'spf': 'navigate'})
3215 html_content = data[1]['body']['content']
3216
3217 if 'class="search-message' in html_content:
3218 raise ExtractorError(
3219 '[youtube] No video results', expected=True)
3220
3221 new_videos = list(self._process_page(html_content))
3222 videos += new_videos
3223 if not new_videos or len(videos) > limit:
3224 break
3225 next_link = self._html_search_regex(
3226 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3227 html_content, 'next link', default=None)
3228 if next_link is None:
3229 break
3230 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3231
3232 if len(videos) > n:
3233 videos = videos[:n]
3234 return self.playlist_result(videos, query)
3235
3236
3237 class YoutubeSearchDateIE(YoutubeSearchIE):
3238 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3239 _SEARCH_KEY = 'ytsearchdate'
3240 IE_DESC = 'YouTube.com searches, newest videos first'
3241 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3242
3243
3244 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3245 IE_DESC = 'YouTube.com search URLs'
3246 IE_NAME = 'youtube:search_url'
3247 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3248 _TESTS = [{
3249 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3250 'playlist_mincount': 5,
3251 'info_dict': {
3252 'title': 'youtube-dl test video',
3253 }
3254 }, {
3255 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3256 'only_matching': True,
3257 }]
3258
3259 def _real_extract(self, url):
3260 mobj = re.match(self._VALID_URL, url)
3261 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3262 webpage = self._download_webpage(url, query)
3263 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3264
3265
3266 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3267 IE_DESC = 'YouTube.com (multi-season) shows'
3268 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3269 IE_NAME = 'youtube:show'
3270 _TESTS = [{
3271 'url': 'https://www.youtube.com/show/airdisasters',
3272 'playlist_mincount': 5,
3273 'info_dict': {
3274 'id': 'airdisasters',
3275 'title': 'Air Disasters',
3276 }
3277 }]
3278
3279 def _real_extract(self, url):
3280 playlist_id = self._match_id(url)
3281 return super(YoutubeShowIE, self)._real_extract(
3282 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3283
3284
3285 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3286 """
3287 Base class for feed extractors
3288 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3289 """
3290 _LOGIN_REQUIRED = True
3291
3292 @property
3293 def IE_NAME(self):
3294 return 'youtube:%s' % self._FEED_NAME
3295
3296 def _real_initialize(self):
3297 self._login()
3298
3299 def _entries(self, page):
3300 # The extraction process is the same as for playlists, but the regex
3301 # for the video ids doesn't contain an index
3302 ids = []
3303 more_widget_html = content_html = page
3304 for page_num in itertools.count(1):
3305 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3306
3307 # 'recommended' feed has infinite 'load more' and each new portion spins
3308 # the same videos in (sometimes) slightly different order, so we'll check
3309 # for unicity and break when portion has no new videos
3310 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3311 if not new_ids:
3312 break
3313
3314 ids.extend(new_ids)
3315
3316 for entry in self._ids_to_results(new_ids):
3317 yield entry
3318
3319 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3320 if not mobj:
3321 break
3322
3323 more = self._download_json(
3324 'https://www.youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3325 'Downloading page #%s' % page_num,
3326 transform_source=uppercase_escape,
3327 headers=self._YOUTUBE_CLIENT_HEADERS)
3328 content_html = more['content_html']
3329 more_widget_html = more['load_more_widget_html']
3330
3331 def _real_extract(self, url):
3332 page = self._download_webpage(
3333 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3334 self._PLAYLIST_TITLE)
3335 return self.playlist_result(
3336 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3337
3338
3339 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3340 IE_NAME = 'youtube:watchlater'
3341 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3342 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3343
3344 _TESTS = [{
3345 'url': 'https://www.youtube.com/playlist?list=WL',
3346 'only_matching': True,
3347 }, {
3348 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3349 'only_matching': True,
3350 }]
3351
3352 def _real_extract(self, url):
3353 _, video = self._check_download_just_video(url, 'WL')
3354 if video:
3355 return video
3356 _, playlist = self._extract_playlist('WL')
3357 return playlist
3358
3359
3360 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3361 IE_NAME = 'youtube:favorites'
3362 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3363 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3364 _LOGIN_REQUIRED = True
3365
3366 def _real_extract(self, url):
3367 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3368 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3369 return self.url_result(playlist_id, 'YoutubePlaylist')
3370
3371
3372 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3373 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3374 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3375 _FEED_NAME = 'recommended'
3376 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3377
3378
3379 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3380 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3381 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3382 _FEED_NAME = 'subscriptions'
3383 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3384
3385
3386 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3387 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3388 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3389 _FEED_NAME = 'history'
3390 _PLAYLIST_TITLE = 'Youtube History'
3391
3392
3393 class YoutubeTruncatedURLIE(InfoExtractor):
3394 IE_NAME = 'youtube:truncated_url'
3395 IE_DESC = False # Do not list
3396 _VALID_URL = r'''(?x)
3397 (?:https?://)?
3398 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3399 (?:watch\?(?:
3400 feature=[a-z_]+|
3401 annotation_id=annotation_[^&]+|
3402 x-yt-cl=[0-9]+|
3403 hl=[^&]*|
3404 t=[0-9]+
3405 )?
3406 |
3407 attribution_link\?a=[^&]+
3408 )
3409 $
3410 '''
3411
3412 _TESTS = [{
3413 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3414 'only_matching': True,
3415 }, {
3416 'url': 'https://www.youtube.com/watch?',
3417 'only_matching': True,
3418 }, {
3419 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3420 'only_matching': True,
3421 }, {
3422 'url': 'https://www.youtube.com/watch?feature=foo',
3423 'only_matching': True,
3424 }, {
3425 'url': 'https://www.youtube.com/watch?hl=en-GB',
3426 'only_matching': True,
3427 }, {
3428 'url': 'https://www.youtube.com/watch?t=2372',
3429 'only_matching': True,
3430 }]
3431
3432 def _real_extract(self, url):
3433 raise ExtractorError(
3434 'Did you forget to quote the URL? Remember that & is a meta '
3435 'character in most shells, so you want to put the URL in quotes, '
3436 'like youtube-dl '
3437 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3438 ' or simply youtube-dl BaW_jenozKc .',
3439 expected=True)
3440
3441
3442 class YoutubeTruncatedIDIE(InfoExtractor):
3443 IE_NAME = 'youtube:truncated_id'
3444 IE_DESC = False # Do not list
3445 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3446
3447 _TESTS = [{
3448 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3449 'only_matching': True,
3450 }]
3451
3452 def _real_extract(self, url):
3453 video_id = self._match_id(url)
3454 raise ExtractorError(
3455 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3456 expected=True)