]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[openload] Add support for oladblock domains (#20471)
[yt-dlp.git] / youtube_dl / 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_kwargs,
20 compat_parse_qs,
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlencode,
24 compat_urllib_parse_urlparse,
25 compat_urlparse,
26 compat_str,
27 )
28 from ..utils import (
29 clean_html,
30 error_to_compat_str,
31 ExtractorError,
32 float_or_none,
33 get_element_by_attribute,
34 get_element_by_id,
35 int_or_none,
36 mimetype2ext,
37 orderedSet,
38 parse_codecs,
39 parse_duration,
40 qualities,
41 remove_quotes,
42 remove_start,
43 smuggle_url,
44 str_or_none,
45 str_to_int,
46 try_get,
47 unescapeHTML,
48 unified_strdate,
49 unsmuggle_url,
50 uppercase_escape,
51 url_or_none,
52 urlencode_postdata,
53 )
54
55
56 class YoutubeBaseInfoExtractor(InfoExtractor):
57 """Provide base functions for Youtube extractors"""
58 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
59 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
60
61 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
62 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
63 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
64
65 _NETRC_MACHINE = 'youtube'
66 # If True it will raise an error if no login info is provided
67 _LOGIN_REQUIRED = False
68
69 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
70
71 def _set_language(self):
72 self._set_cookie(
73 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
74 # YouTube sets the expire time to about two months
75 expire_time=time.time() + 2 * 30 * 24 * 3600)
76
77 def _ids_to_results(self, ids):
78 return [
79 self.url_result(vid_id, 'Youtube', video_id=vid_id)
80 for vid_id in ids]
81
82 def _login(self):
83 """
84 Attempt to log in to YouTube.
85 True is returned if successful or skipped.
86 False is returned if login failed.
87
88 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
89 """
90 username, password = self._get_login_info()
91 # No authentication to be performed
92 if username is None:
93 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
94 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
95 return True
96
97 login_page = self._download_webpage(
98 self._LOGIN_URL, None,
99 note='Downloading login page',
100 errnote='unable to fetch login page', fatal=False)
101 if login_page is False:
102 return
103
104 login_form = self._hidden_inputs(login_page)
105
106 def req(url, f_req, note, errnote):
107 data = login_form.copy()
108 data.update({
109 'pstMsg': 1,
110 'checkConnection': 'youtube',
111 'checkedDomains': 'youtube',
112 'hl': 'en',
113 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
114 'f.req': json.dumps(f_req),
115 'flowName': 'GlifWebSignIn',
116 'flowEntry': 'ServiceLogin',
117 })
118 return self._download_json(
119 url, None, note=note, errnote=errnote,
120 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
121 fatal=False,
122 data=urlencode_postdata(data), headers={
123 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
124 'Google-Accounts-XSRF': 1,
125 })
126
127 def warn(message):
128 self._downloader.report_warning(message)
129
130 lookup_req = [
131 username,
132 None, [], None, 'US', None, None, 2, False, True,
133 [
134 None, None,
135 [2, 1, None, 1,
136 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
137 None, [], 4],
138 1, [None, None, []], None, None, None, True
139 ],
140 username,
141 ]
142
143 lookup_results = req(
144 self._LOOKUP_URL, lookup_req,
145 'Looking up account info', 'Unable to look up account info')
146
147 if lookup_results is False:
148 return False
149
150 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
151 if not user_hash:
152 warn('Unable to extract user hash')
153 return False
154
155 challenge_req = [
156 user_hash,
157 None, 1, None, [1, None, None, None, [password, None, True]],
158 [
159 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
160 1, [None, None, []], None, None, None, True
161 ]]
162
163 challenge_results = req(
164 self._CHALLENGE_URL, challenge_req,
165 'Logging in', 'Unable to log in')
166
167 if challenge_results is False:
168 return
169
170 login_res = try_get(challenge_results, lambda x: x[0][5], list)
171 if login_res:
172 login_msg = try_get(login_res, lambda x: x[5], compat_str)
173 warn(
174 'Unable to login: %s' % 'Invalid password'
175 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
176 return False
177
178 res = try_get(challenge_results, lambda x: x[0][-1], list)
179 if not res:
180 warn('Unable to extract result entry')
181 return False
182
183 login_challenge = try_get(res, lambda x: x[0][0], list)
184 if login_challenge:
185 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
186 if challenge_str == 'TWO_STEP_VERIFICATION':
187 # SEND_SUCCESS - TFA code has been successfully sent to phone
188 # QUOTA_EXCEEDED - reached the limit of TFA codes
189 status = try_get(login_challenge, lambda x: x[5], compat_str)
190 if status == 'QUOTA_EXCEEDED':
191 warn('Exceeded the limit of TFA codes, try later')
192 return False
193
194 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
195 if not tl:
196 warn('Unable to extract TL')
197 return False
198
199 tfa_code = self._get_tfa_info('2-step verification code')
200
201 if not tfa_code:
202 warn(
203 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
204 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
205 return False
206
207 tfa_code = remove_start(tfa_code, 'G-')
208
209 tfa_req = [
210 user_hash, None, 2, None,
211 [
212 9, None, None, None, None, None, None, None,
213 [None, tfa_code, True, 2]
214 ]]
215
216 tfa_results = req(
217 self._TFA_URL.format(tl), tfa_req,
218 'Submitting TFA code', 'Unable to submit TFA code')
219
220 if tfa_results is False:
221 return False
222
223 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
224 if tfa_res:
225 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
226 warn(
227 'Unable to finish TFA: %s' % 'Invalid TFA code'
228 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
229 return False
230
231 check_cookie_url = try_get(
232 tfa_results, lambda x: x[0][-1][2], compat_str)
233 else:
234 CHALLENGES = {
235 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
236 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
237 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
238 }
239 challenge = CHALLENGES.get(
240 challenge_str,
241 '%s returned error %s.' % (self.IE_NAME, challenge_str))
242 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
243 return False
244 else:
245 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
246
247 if not check_cookie_url:
248 warn('Unable to extract CheckCookie URL')
249 return False
250
251 check_cookie_results = self._download_webpage(
252 check_cookie_url, None, 'Checking cookie', fatal=False)
253
254 if check_cookie_results is False:
255 return False
256
257 if 'https://myaccount.google.com/' not in check_cookie_results:
258 warn('Unable to log in')
259 return False
260
261 return True
262
263 def _download_webpage_handle(self, *args, **kwargs):
264 query = kwargs.get('query', {}).copy()
265 query['disable_polymer'] = 'true'
266 kwargs['query'] = query
267 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
268 *args, **compat_kwargs(kwargs))
269
270 def _real_initialize(self):
271 if self._downloader is None:
272 return
273 self._set_language()
274 if not self._login():
275 return
276
277
278 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
279 # Extract entries from page with "Load more" button
280 def _entries(self, page, playlist_id):
281 more_widget_html = content_html = page
282 for page_num in itertools.count(1):
283 for entry in self._process_page(content_html):
284 yield entry
285
286 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
287 if not mobj:
288 break
289
290 more = self._download_json(
291 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
292 'Downloading page #%s' % page_num,
293 transform_source=uppercase_escape)
294 content_html = more['content_html']
295 if not content_html.strip():
296 # Some webpages show a "Load more" button but they don't
297 # have more videos
298 break
299 more_widget_html = more['load_more_widget_html']
300
301
302 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
303 def _process_page(self, content):
304 for video_id, video_title in self.extract_videos_from_page(content):
305 yield self.url_result(video_id, 'Youtube', video_id, video_title)
306
307 def extract_videos_from_page(self, page):
308 ids_in_page = []
309 titles_in_page = []
310 for mobj in re.finditer(self._VIDEO_RE, page):
311 # The link with index 0 is not the first video of the playlist (not sure if still actual)
312 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
313 continue
314 video_id = mobj.group('id')
315 video_title = unescapeHTML(mobj.group('title'))
316 if video_title:
317 video_title = video_title.strip()
318 try:
319 idx = ids_in_page.index(video_id)
320 if video_title and not titles_in_page[idx]:
321 titles_in_page[idx] = video_title
322 except ValueError:
323 ids_in_page.append(video_id)
324 titles_in_page.append(video_title)
325 return zip(ids_in_page, titles_in_page)
326
327
328 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
329 def _process_page(self, content):
330 for playlist_id in orderedSet(re.findall(
331 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
332 content)):
333 yield self.url_result(
334 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
335
336 def _real_extract(self, url):
337 playlist_id = self._match_id(url)
338 webpage = self._download_webpage(url, playlist_id)
339 title = self._og_search_title(webpage, fatal=False)
340 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
341
342
343 class YoutubeIE(YoutubeBaseInfoExtractor):
344 IE_DESC = 'YouTube.com'
345 _VALID_URL = r"""(?x)^
346 (
347 (?:https?://|//) # http(s):// or protocol-independent URL
348 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
349 (?:www\.)?deturl\.com/www\.youtube\.com/|
350 (?:www\.)?pwnyoutube\.com/|
351 (?:www\.)?hooktube\.com/|
352 (?:www\.)?yourepeat\.com/|
353 tube\.majestyc\.net/|
354 (?:(?:www|dev)\.)?invidio\.us/|
355 (?:www\.)?invidiou\.sh/|
356 (?:www\.)?invidious\.snopyta\.org/|
357 (?:www\.)?invidious\.kabi\.tk/|
358 (?:www\.)?vid\.wxzm\.sx/|
359 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
360 (?:.*?\#/)? # handle anchor (#/) redirect urls
361 (?: # the various things that can precede the ID:
362 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
363 |(?: # or the v= param in all its forms
364 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
365 (?:\?|\#!?) # the params delimiter ? or # or #!
366 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
367 v=
368 )
369 ))
370 |(?:
371 youtu\.be| # just youtu.be/xxxx
372 vid\.plus| # or vid.plus/xxxx
373 zwearz\.com/watch| # or zwearz.com/watch/xxxx
374 )/
375 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
376 )
377 )? # all until now is optional -> you can pass the naked ID
378 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
379 (?!.*?\blist=
380 (?:
381 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
382 WL # WL are handled by the watch later IE
383 )
384 )
385 (?(1).+)? # if we found the ID, everything can follow
386 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
387 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
388 _formats = {
389 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
390 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
391 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
392 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
393 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
394 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
395 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
396 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
397 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
398 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
399 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
400 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
401 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
402 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
403 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
404 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
405 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
406 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
407
408
409 # 3D videos
410 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
411 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
412 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
413 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
414 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
415 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
416 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
417
418 # Apple HTTP Live Streaming
419 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
420 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
421 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
422 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
423 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
424 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
425 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
426 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
427
428 # DASH mp4 video
429 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
430 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
431 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
432 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
433 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
434 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
435 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
436 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
437 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
438 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
439 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
440 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
441
442 # Dash mp4 audio
443 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
444 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
445 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
446 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
447 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
448 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
449 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
450
451 # Dash webm
452 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
453 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
454 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
455 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
456 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
457 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
458 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
459 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
460 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
461 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
462 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
463 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
464 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
465 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
466 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
467 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
468 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
469 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
470 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
471 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
472 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
473 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
474
475 # Dash webm audio
476 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
477 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
478
479 # Dash webm audio with opus inside
480 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
481 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
482 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
483
484 # RTMP (unnamed)
485 '_rtmp': {'protocol': 'rtmp'},
486 }
487 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
488
489 _GEO_BYPASS = False
490
491 IE_NAME = 'youtube'
492 _TESTS = [
493 {
494 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
495 'info_dict': {
496 'id': 'BaW_jenozKc',
497 'ext': 'mp4',
498 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
499 'uploader': 'Philipp Hagemeister',
500 'uploader_id': 'phihag',
501 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
502 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
503 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
504 'upload_date': '20121002',
505 '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 .',
506 'categories': ['Science & Technology'],
507 'tags': ['youtube-dl'],
508 'duration': 10,
509 'view_count': int,
510 'like_count': int,
511 'dislike_count': int,
512 'start_time': 1,
513 'end_time': 9,
514 }
515 },
516 {
517 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
518 'note': 'Test generic use_cipher_signature video (#897)',
519 'info_dict': {
520 'id': 'UxxajLWwzqY',
521 'ext': 'mp4',
522 'upload_date': '20120506',
523 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
524 'alt_title': 'I Love It (feat. Charli XCX)',
525 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
526 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
527 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
528 'iconic ep', 'iconic', 'love', 'it'],
529 'duration': 180,
530 'uploader': 'Icona Pop',
531 'uploader_id': 'IconaPop',
532 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
533 'creator': 'Icona Pop',
534 'track': 'I Love It (feat. Charli XCX)',
535 'artist': 'Icona Pop',
536 }
537 },
538 {
539 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
540 'note': 'Test VEVO video with age protection (#956)',
541 'info_dict': {
542 'id': '07FYdnEawAQ',
543 'ext': 'mp4',
544 'upload_date': '20130703',
545 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
546 'alt_title': 'Tunnel Vision',
547 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
548 'duration': 419,
549 'uploader': 'justintimberlakeVEVO',
550 'uploader_id': 'justintimberlakeVEVO',
551 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
552 'creator': 'Justin Timberlake',
553 'track': 'Tunnel Vision',
554 'artist': 'Justin Timberlake',
555 'age_limit': 18,
556 }
557 },
558 {
559 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
560 'note': 'Embed-only video (#1746)',
561 'info_dict': {
562 'id': 'yZIXLfi8CZQ',
563 'ext': 'mp4',
564 'upload_date': '20120608',
565 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
566 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
567 'uploader': 'SET India',
568 'uploader_id': 'setindia',
569 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
570 'age_limit': 18,
571 }
572 },
573 {
574 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
575 'note': 'Use the first video ID in the URL',
576 'info_dict': {
577 'id': 'BaW_jenozKc',
578 'ext': 'mp4',
579 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
580 'uploader': 'Philipp Hagemeister',
581 'uploader_id': 'phihag',
582 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
583 'upload_date': '20121002',
584 '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 .',
585 'categories': ['Science & Technology'],
586 'tags': ['youtube-dl'],
587 'duration': 10,
588 'view_count': int,
589 'like_count': int,
590 'dislike_count': int,
591 },
592 'params': {
593 'skip_download': True,
594 },
595 },
596 {
597 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
598 'note': '256k DASH audio (format 141) via DASH manifest',
599 'info_dict': {
600 'id': 'a9LDPn-MO4I',
601 'ext': 'm4a',
602 'upload_date': '20121002',
603 'uploader_id': '8KVIDEO',
604 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
605 'description': '',
606 'uploader': '8KVIDEO',
607 'title': 'UHDTV TEST 8K VIDEO.mp4'
608 },
609 'params': {
610 'youtube_include_dash_manifest': True,
611 'format': '141',
612 },
613 'skip': 'format 141 not served anymore',
614 },
615 # DASH manifest with encrypted signature
616 {
617 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
618 'info_dict': {
619 'id': 'IB3lcPjvWLA',
620 'ext': 'm4a',
621 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
622 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
623 'duration': 244,
624 'uploader': 'AfrojackVEVO',
625 'uploader_id': 'AfrojackVEVO',
626 'upload_date': '20131011',
627 },
628 'params': {
629 'youtube_include_dash_manifest': True,
630 'format': '141/bestaudio[ext=m4a]',
631 },
632 },
633 # JS player signature function name containing $
634 {
635 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
636 'info_dict': {
637 'id': 'nfWlot6h_JM',
638 'ext': 'm4a',
639 'title': 'Taylor Swift - Shake It Off',
640 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
641 'duration': 242,
642 'uploader': 'TaylorSwiftVEVO',
643 'uploader_id': 'TaylorSwiftVEVO',
644 'upload_date': '20140818',
645 'creator': 'Taylor Swift',
646 },
647 'params': {
648 'youtube_include_dash_manifest': True,
649 'format': '141/bestaudio[ext=m4a]',
650 },
651 },
652 # Controversy video
653 {
654 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
655 'info_dict': {
656 'id': 'T4XJQO3qol8',
657 'ext': 'mp4',
658 'duration': 219,
659 'upload_date': '20100909',
660 'uploader': 'Amazing Atheist',
661 'uploader_id': 'TheAmazingAtheist',
662 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
663 'title': 'Burning Everyone\'s Koran',
664 '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',
665 }
666 },
667 # Normal age-gate video (No vevo, embed allowed)
668 {
669 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
670 'info_dict': {
671 'id': 'HtVdAasjOgU',
672 'ext': 'mp4',
673 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
674 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
675 'duration': 142,
676 'uploader': 'The Witcher',
677 'uploader_id': 'WitcherGame',
678 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
679 'upload_date': '20140605',
680 'age_limit': 18,
681 },
682 },
683 # Age-gate video with encrypted signature
684 {
685 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
686 'info_dict': {
687 'id': '6kLq3WMV1nU',
688 'ext': 'mp4',
689 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
690 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
691 'duration': 246,
692 'uploader': 'LloydVEVO',
693 'uploader_id': 'LloydVEVO',
694 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
695 'upload_date': '20110629',
696 'age_limit': 18,
697 },
698 },
699 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
700 # YouTube Red ad is not captured for creator
701 {
702 'url': '__2ABJjxzNo',
703 'info_dict': {
704 'id': '__2ABJjxzNo',
705 'ext': 'mp4',
706 'duration': 266,
707 'upload_date': '20100430',
708 'uploader_id': 'deadmau5',
709 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
710 'creator': 'deadmau5',
711 'description': 'md5:12c56784b8032162bb936a5f76d55360',
712 'uploader': 'deadmau5',
713 'title': 'Deadmau5 - Some Chords (HD)',
714 'alt_title': 'Some Chords',
715 },
716 'expected_warnings': [
717 'DASH manifest missing',
718 ]
719 },
720 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
721 {
722 'url': 'lqQg6PlCWgI',
723 'info_dict': {
724 'id': 'lqQg6PlCWgI',
725 'ext': 'mp4',
726 'duration': 6085,
727 'upload_date': '20150827',
728 'uploader_id': 'olympic',
729 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
730 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
731 'uploader': 'Olympic',
732 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
733 },
734 'params': {
735 'skip_download': 'requires avconv',
736 }
737 },
738 # Non-square pixels
739 {
740 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
741 'info_dict': {
742 'id': '_b-2C3KPAM0',
743 'ext': 'mp4',
744 'stretched_ratio': 16 / 9.,
745 'duration': 85,
746 'upload_date': '20110310',
747 'uploader_id': 'AllenMeow',
748 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
749 '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',
750 'uploader': '孫ᄋᄅ',
751 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
752 },
753 },
754 # url_encoded_fmt_stream_map is empty string
755 {
756 'url': 'qEJwOuvDf7I',
757 'info_dict': {
758 'id': 'qEJwOuvDf7I',
759 'ext': 'webm',
760 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
761 'description': '',
762 'upload_date': '20150404',
763 'uploader_id': 'spbelect',
764 'uploader': 'Наблюдатели Петербурга',
765 },
766 'params': {
767 'skip_download': 'requires avconv',
768 },
769 'skip': 'This live event has ended.',
770 },
771 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
772 {
773 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
774 'info_dict': {
775 'id': 'FIl7x6_3R5Y',
776 'ext': 'webm',
777 'title': 'md5:7b81415841e02ecd4313668cde88737a',
778 'description': 'md5:116377fd2963b81ec4ce64b542173306',
779 'duration': 220,
780 'upload_date': '20150625',
781 'uploader_id': 'dorappi2000',
782 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
783 'uploader': 'dorappi2000',
784 'formats': 'mincount:31',
785 },
786 'skip': 'not actual anymore',
787 },
788 # DASH manifest with segment_list
789 {
790 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
791 'md5': '8ce563a1d667b599d21064e982ab9e31',
792 'info_dict': {
793 'id': 'CsmdDsKjzN8',
794 'ext': 'mp4',
795 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
796 'uploader': 'Airtek',
797 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
798 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
799 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
800 },
801 'params': {
802 'youtube_include_dash_manifest': True,
803 'format': '135', # bestvideo
804 },
805 'skip': 'This live event has ended.',
806 },
807 {
808 # Multifeed videos (multiple cameras), URL is for Main Camera
809 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
810 'info_dict': {
811 'id': 'jqWvoWXjCVs',
812 'title': 'teamPGP: Rocket League Noob Stream',
813 'description': 'md5:dc7872fb300e143831327f1bae3af010',
814 },
815 'playlist': [{
816 'info_dict': {
817 'id': 'jqWvoWXjCVs',
818 'ext': 'mp4',
819 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
820 'description': 'md5:dc7872fb300e143831327f1bae3af010',
821 'duration': 7335,
822 'upload_date': '20150721',
823 'uploader': 'Beer Games Beer',
824 'uploader_id': 'beergamesbeer',
825 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
826 'license': 'Standard YouTube License',
827 },
828 }, {
829 'info_dict': {
830 'id': '6h8e8xoXJzg',
831 'ext': 'mp4',
832 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
833 'description': 'md5:dc7872fb300e143831327f1bae3af010',
834 'duration': 7337,
835 'upload_date': '20150721',
836 'uploader': 'Beer Games Beer',
837 'uploader_id': 'beergamesbeer',
838 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
839 'license': 'Standard YouTube License',
840 },
841 }, {
842 'info_dict': {
843 'id': 'PUOgX5z9xZw',
844 'ext': 'mp4',
845 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
846 'description': 'md5:dc7872fb300e143831327f1bae3af010',
847 'duration': 7337,
848 'upload_date': '20150721',
849 'uploader': 'Beer Games Beer',
850 'uploader_id': 'beergamesbeer',
851 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
852 'license': 'Standard YouTube License',
853 },
854 }, {
855 'info_dict': {
856 'id': 'teuwxikvS5k',
857 'ext': 'mp4',
858 'title': 'teamPGP: Rocket League Noob Stream (zim)',
859 'description': 'md5:dc7872fb300e143831327f1bae3af010',
860 'duration': 7334,
861 'upload_date': '20150721',
862 'uploader': 'Beer Games Beer',
863 'uploader_id': 'beergamesbeer',
864 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
865 'license': 'Standard YouTube License',
866 },
867 }],
868 'params': {
869 'skip_download': True,
870 },
871 'skip': 'This video is not available.',
872 },
873 {
874 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
875 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
876 'info_dict': {
877 'id': 'gVfLd0zydlo',
878 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
879 },
880 'playlist_count': 2,
881 'skip': 'Not multifeed anymore',
882 },
883 {
884 'url': 'https://vid.plus/FlRa-iH7PGw',
885 'only_matching': True,
886 },
887 {
888 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
889 'only_matching': True,
890 },
891 {
892 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
893 # Also tests cut-off URL expansion in video description (see
894 # https://github.com/ytdl-org/youtube-dl/issues/1892,
895 # https://github.com/ytdl-org/youtube-dl/issues/8164)
896 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
897 'info_dict': {
898 'id': 'lsguqyKfVQg',
899 'ext': 'mp4',
900 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
901 'alt_title': 'Dark Walk - Position Music',
902 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
903 'duration': 133,
904 'upload_date': '20151119',
905 'uploader_id': 'IronSoulElf',
906 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
907 'uploader': 'IronSoulElf',
908 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
909 'track': 'Dark Walk - Position Music',
910 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
911 },
912 'params': {
913 'skip_download': True,
914 },
915 },
916 {
917 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
918 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
919 'only_matching': True,
920 },
921 {
922 # Video with yt:stretch=17:0
923 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
924 'info_dict': {
925 'id': 'Q39EVAstoRM',
926 'ext': 'mp4',
927 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
928 'description': 'md5:ee18a25c350637c8faff806845bddee9',
929 'upload_date': '20151107',
930 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
931 'uploader': 'CH GAMER DROID',
932 },
933 'params': {
934 'skip_download': True,
935 },
936 'skip': 'This video does not exist.',
937 },
938 {
939 # Video licensed under Creative Commons
940 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
941 'info_dict': {
942 'id': 'M4gD1WSo5mA',
943 'ext': 'mp4',
944 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
945 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
946 'duration': 721,
947 'upload_date': '20150127',
948 'uploader_id': 'BerkmanCenter',
949 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
950 'uploader': 'The Berkman Klein Center for Internet & Society',
951 'license': 'Creative Commons Attribution license (reuse allowed)',
952 },
953 'params': {
954 'skip_download': True,
955 },
956 },
957 {
958 # Channel-like uploader_url
959 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
960 'info_dict': {
961 'id': 'eQcmzGIKrzg',
962 'ext': 'mp4',
963 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
964 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
965 'duration': 4060,
966 'upload_date': '20151119',
967 'uploader': 'Bernie Sanders',
968 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
969 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
970 'license': 'Creative Commons Attribution license (reuse allowed)',
971 },
972 'params': {
973 'skip_download': True,
974 },
975 },
976 {
977 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
978 'only_matching': True,
979 },
980 {
981 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
982 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
983 'only_matching': True,
984 },
985 {
986 # Rental video preview
987 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
988 'info_dict': {
989 'id': 'uGpuVWrhIzE',
990 'ext': 'mp4',
991 'title': 'Piku - Trailer',
992 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
993 'upload_date': '20150811',
994 'uploader': 'FlixMatrix',
995 'uploader_id': 'FlixMatrixKaravan',
996 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
997 'license': 'Standard YouTube License',
998 },
999 'params': {
1000 'skip_download': True,
1001 },
1002 'skip': 'This video is not available.',
1003 },
1004 {
1005 # YouTube Red video with episode data
1006 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1007 'info_dict': {
1008 'id': 'iqKdEhx-dD4',
1009 'ext': 'mp4',
1010 'title': 'Isolation - Mind Field (Ep 1)',
1011 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1012 'duration': 2085,
1013 'upload_date': '20170118',
1014 'uploader': 'Vsauce',
1015 'uploader_id': 'Vsauce',
1016 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1017 'series': 'Mind Field',
1018 'season_number': 1,
1019 'episode_number': 1,
1020 },
1021 'params': {
1022 'skip_download': True,
1023 },
1024 'expected_warnings': [
1025 'Skipping DASH manifest',
1026 ],
1027 },
1028 {
1029 # The following content has been identified by the YouTube community
1030 # as inappropriate or offensive to some audiences.
1031 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1032 'info_dict': {
1033 'id': '6SJNVb0GnPI',
1034 'ext': 'mp4',
1035 'title': 'Race Differences in Intelligence',
1036 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1037 'duration': 965,
1038 'upload_date': '20140124',
1039 'uploader': 'New Century Foundation',
1040 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1041 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1042 },
1043 'params': {
1044 'skip_download': True,
1045 },
1046 },
1047 {
1048 # itag 212
1049 'url': '1t24XAntNCY',
1050 'only_matching': True,
1051 },
1052 {
1053 # geo restricted to JP
1054 'url': 'sJL6WA-aGkQ',
1055 'only_matching': True,
1056 },
1057 {
1058 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1059 'only_matching': True,
1060 },
1061 {
1062 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1063 'only_matching': True,
1064 },
1065 {
1066 # DRM protected
1067 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1068 'only_matching': True,
1069 },
1070 {
1071 # Video with unsupported adaptive stream type formats
1072 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1073 'info_dict': {
1074 'id': 'Z4Vy8R84T1U',
1075 'ext': 'mp4',
1076 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1077 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1078 'duration': 433,
1079 'upload_date': '20130923',
1080 'uploader': 'Amelia Putri Harwita',
1081 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1082 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1083 'formats': 'maxcount:10',
1084 },
1085 'params': {
1086 'skip_download': True,
1087 'youtube_include_dash_manifest': False,
1088 },
1089 }
1090 ]
1091
1092 def __init__(self, *args, **kwargs):
1093 super(YoutubeIE, self).__init__(*args, **kwargs)
1094 self._player_cache = {}
1095
1096 def report_video_info_webpage_download(self, video_id):
1097 """Report attempt to download video info webpage."""
1098 self.to_screen('%s: Downloading video info webpage' % video_id)
1099
1100 def report_information_extraction(self, video_id):
1101 """Report attempt to extract video information."""
1102 self.to_screen('%s: Extracting video information' % video_id)
1103
1104 def report_unavailable_format(self, video_id, format):
1105 """Report extracted video URL."""
1106 self.to_screen('%s: Format %s not available' % (video_id, format))
1107
1108 def report_rtmp_download(self):
1109 """Indicate the download will use the RTMP protocol."""
1110 self.to_screen('RTMP download detected')
1111
1112 def _signature_cache_id(self, example_sig):
1113 """ Return a string representation of a signature """
1114 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1115
1116 def _extract_signature_function(self, video_id, player_url, example_sig):
1117 id_m = re.match(
1118 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1119 player_url)
1120 if not id_m:
1121 raise ExtractorError('Cannot identify player %r' % player_url)
1122 player_type = id_m.group('ext')
1123 player_id = id_m.group('id')
1124
1125 # Read from filesystem cache
1126 func_id = '%s_%s_%s' % (
1127 player_type, player_id, self._signature_cache_id(example_sig))
1128 assert os.path.basename(func_id) == func_id
1129
1130 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1131 if cache_spec is not None:
1132 return lambda s: ''.join(s[i] for i in cache_spec)
1133
1134 download_note = (
1135 'Downloading player %s' % player_url
1136 if self._downloader.params.get('verbose') else
1137 'Downloading %s player %s' % (player_type, player_id)
1138 )
1139 if player_type == 'js':
1140 code = self._download_webpage(
1141 player_url, video_id,
1142 note=download_note,
1143 errnote='Download of %s failed' % player_url)
1144 res = self._parse_sig_js(code)
1145 elif player_type == 'swf':
1146 urlh = self._request_webpage(
1147 player_url, video_id,
1148 note=download_note,
1149 errnote='Download of %s failed' % player_url)
1150 code = urlh.read()
1151 res = self._parse_sig_swf(code)
1152 else:
1153 assert False, 'Invalid player type %r' % player_type
1154
1155 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1156 cache_res = res(test_string)
1157 cache_spec = [ord(c) for c in cache_res]
1158
1159 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1160 return res
1161
1162 def _print_sig_code(self, func, example_sig):
1163 def gen_sig_code(idxs):
1164 def _genslice(start, end, step):
1165 starts = '' if start == 0 else str(start)
1166 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1167 steps = '' if step == 1 else (':%d' % step)
1168 return 's[%s%s%s]' % (starts, ends, steps)
1169
1170 step = None
1171 # Quelch pyflakes warnings - start will be set when step is set
1172 start = '(Never used)'
1173 for i, prev in zip(idxs[1:], idxs[:-1]):
1174 if step is not None:
1175 if i - prev == step:
1176 continue
1177 yield _genslice(start, prev, step)
1178 step = None
1179 continue
1180 if i - prev in [-1, 1]:
1181 step = i - prev
1182 start = prev
1183 continue
1184 else:
1185 yield 's[%d]' % prev
1186 if step is None:
1187 yield 's[%d]' % i
1188 else:
1189 yield _genslice(start, i, step)
1190
1191 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1192 cache_res = func(test_string)
1193 cache_spec = [ord(c) for c in cache_res]
1194 expr_code = ' + '.join(gen_sig_code(cache_spec))
1195 signature_id_tuple = '(%s)' % (
1196 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1197 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1198 ' return %s\n') % (signature_id_tuple, expr_code)
1199 self.to_screen('Extracted signature function:\n' + code)
1200
1201 def _parse_sig_js(self, jscode):
1202 funcname = self._search_regex(
1203 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1204 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1205 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
1206 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1207 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1208 jscode, 'Initial JS player signature function name', group='sig')
1209
1210 jsi = JSInterpreter(jscode)
1211 initial_function = jsi.extract_function(funcname)
1212 return lambda s: initial_function([s])
1213
1214 def _parse_sig_swf(self, file_contents):
1215 swfi = SWFInterpreter(file_contents)
1216 TARGET_CLASSNAME = 'SignatureDecipher'
1217 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1218 initial_function = swfi.extract_function(searched_class, 'decipher')
1219 return lambda s: initial_function([s])
1220
1221 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1222 """Turn the encrypted s field into a working signature"""
1223
1224 if player_url is None:
1225 raise ExtractorError('Cannot decrypt signature without player_url')
1226
1227 if player_url.startswith('//'):
1228 player_url = 'https:' + player_url
1229 elif not re.match(r'https?://', player_url):
1230 player_url = compat_urlparse.urljoin(
1231 'https://www.youtube.com', player_url)
1232 try:
1233 player_id = (player_url, self._signature_cache_id(s))
1234 if player_id not in self._player_cache:
1235 func = self._extract_signature_function(
1236 video_id, player_url, s
1237 )
1238 self._player_cache[player_id] = func
1239 func = self._player_cache[player_id]
1240 if self._downloader.params.get('youtube_print_sig_code'):
1241 self._print_sig_code(func, s)
1242 return func(s)
1243 except Exception as e:
1244 tb = traceback.format_exc()
1245 raise ExtractorError(
1246 'Signature extraction failed: ' + tb, cause=e)
1247
1248 def _get_subtitles(self, video_id, webpage):
1249 try:
1250 subs_doc = self._download_xml(
1251 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1252 video_id, note=False)
1253 except ExtractorError as err:
1254 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1255 return {}
1256
1257 sub_lang_list = {}
1258 for track in subs_doc.findall('track'):
1259 lang = track.attrib['lang_code']
1260 if lang in sub_lang_list:
1261 continue
1262 sub_formats = []
1263 for ext in self._SUBTITLE_FORMATS:
1264 params = compat_urllib_parse_urlencode({
1265 'lang': lang,
1266 'v': video_id,
1267 'fmt': ext,
1268 'name': track.attrib['name'].encode('utf-8'),
1269 })
1270 sub_formats.append({
1271 'url': 'https://www.youtube.com/api/timedtext?' + params,
1272 'ext': ext,
1273 })
1274 sub_lang_list[lang] = sub_formats
1275 if not sub_lang_list:
1276 self._downloader.report_warning('video doesn\'t have subtitles')
1277 return {}
1278 return sub_lang_list
1279
1280 def _get_ytplayer_config(self, video_id, webpage):
1281 patterns = (
1282 # User data may contain arbitrary character sequences that may affect
1283 # JSON extraction with regex, e.g. when '};' is contained the second
1284 # regex won't capture the whole JSON. Yet working around by trying more
1285 # concrete regex first keeping in mind proper quoted string handling
1286 # to be implemented in future that will replace this workaround (see
1287 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1288 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1289 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1290 r';ytplayer\.config\s*=\s*({.+?});',
1291 )
1292 config = self._search_regex(
1293 patterns, webpage, 'ytplayer.config', default=None)
1294 if config:
1295 return self._parse_json(
1296 uppercase_escape(config), video_id, fatal=False)
1297
1298 def _get_automatic_captions(self, video_id, webpage):
1299 """We need the webpage for getting the captions url, pass it as an
1300 argument to speed up the process."""
1301 self.to_screen('%s: Looking for automatic captions' % video_id)
1302 player_config = self._get_ytplayer_config(video_id, webpage)
1303 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1304 if not player_config:
1305 self._downloader.report_warning(err_msg)
1306 return {}
1307 try:
1308 args = player_config['args']
1309 caption_url = args.get('ttsurl')
1310 if caption_url:
1311 timestamp = args['timestamp']
1312 # We get the available subtitles
1313 list_params = compat_urllib_parse_urlencode({
1314 'type': 'list',
1315 'tlangs': 1,
1316 'asrs': 1,
1317 })
1318 list_url = caption_url + '&' + list_params
1319 caption_list = self._download_xml(list_url, video_id)
1320 original_lang_node = caption_list.find('track')
1321 if original_lang_node is None:
1322 self._downloader.report_warning('Video doesn\'t have automatic captions')
1323 return {}
1324 original_lang = original_lang_node.attrib['lang_code']
1325 caption_kind = original_lang_node.attrib.get('kind', '')
1326
1327 sub_lang_list = {}
1328 for lang_node in caption_list.findall('target'):
1329 sub_lang = lang_node.attrib['lang_code']
1330 sub_formats = []
1331 for ext in self._SUBTITLE_FORMATS:
1332 params = compat_urllib_parse_urlencode({
1333 'lang': original_lang,
1334 'tlang': sub_lang,
1335 'fmt': ext,
1336 'ts': timestamp,
1337 'kind': caption_kind,
1338 })
1339 sub_formats.append({
1340 'url': caption_url + '&' + params,
1341 'ext': ext,
1342 })
1343 sub_lang_list[sub_lang] = sub_formats
1344 return sub_lang_list
1345
1346 def make_captions(sub_url, sub_langs):
1347 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1348 caption_qs = compat_parse_qs(parsed_sub_url.query)
1349 captions = {}
1350 for sub_lang in sub_langs:
1351 sub_formats = []
1352 for ext in self._SUBTITLE_FORMATS:
1353 caption_qs.update({
1354 'tlang': [sub_lang],
1355 'fmt': [ext],
1356 })
1357 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1358 query=compat_urllib_parse_urlencode(caption_qs, True)))
1359 sub_formats.append({
1360 'url': sub_url,
1361 'ext': ext,
1362 })
1363 captions[sub_lang] = sub_formats
1364 return captions
1365
1366 # New captions format as of 22.06.2017
1367 player_response = args.get('player_response')
1368 if player_response and isinstance(player_response, compat_str):
1369 player_response = self._parse_json(
1370 player_response, video_id, fatal=False)
1371 if player_response:
1372 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1373 base_url = renderer['captionTracks'][0]['baseUrl']
1374 sub_lang_list = []
1375 for lang in renderer['translationLanguages']:
1376 lang_code = lang.get('languageCode')
1377 if lang_code:
1378 sub_lang_list.append(lang_code)
1379 return make_captions(base_url, sub_lang_list)
1380
1381 # Some videos don't provide ttsurl but rather caption_tracks and
1382 # caption_translation_languages (e.g. 20LmZk1hakA)
1383 # Does not used anymore as of 22.06.2017
1384 caption_tracks = args['caption_tracks']
1385 caption_translation_languages = args['caption_translation_languages']
1386 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1387 sub_lang_list = []
1388 for lang in caption_translation_languages.split(','):
1389 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1390 sub_lang = lang_qs.get('lc', [None])[0]
1391 if sub_lang:
1392 sub_lang_list.append(sub_lang)
1393 return make_captions(caption_url, sub_lang_list)
1394 # An extractor error can be raise by the download process if there are
1395 # no automatic captions but there are subtitles
1396 except (KeyError, IndexError, ExtractorError):
1397 self._downloader.report_warning(err_msg)
1398 return {}
1399
1400 def _mark_watched(self, video_id, video_info, player_response):
1401 playback_url = url_or_none(try_get(
1402 player_response,
1403 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1404 video_info, lambda x: x['videostats_playback_base_url'][0]))
1405 if not playback_url:
1406 return
1407 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1408 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1409
1410 # cpn generation algorithm is reverse engineered from base.js.
1411 # In fact it works even with dummy cpn.
1412 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1413 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1414
1415 qs.update({
1416 'ver': ['2'],
1417 'cpn': [cpn],
1418 })
1419 playback_url = compat_urlparse.urlunparse(
1420 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1421
1422 self._download_webpage(
1423 playback_url, video_id, 'Marking watched',
1424 'Unable to mark watched', fatal=False)
1425
1426 @staticmethod
1427 def _extract_urls(webpage):
1428 # Embedded YouTube player
1429 entries = [
1430 unescapeHTML(mobj.group('url'))
1431 for mobj in re.finditer(r'''(?x)
1432 (?:
1433 <iframe[^>]+?src=|
1434 data-video-url=|
1435 <embed[^>]+?src=|
1436 embedSWF\(?:\s*|
1437 <object[^>]+data=|
1438 new\s+SWFObject\(
1439 )
1440 (["\'])
1441 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1442 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1443 \1''', webpage)]
1444
1445 # lazyYT YouTube embed
1446 entries.extend(list(map(
1447 unescapeHTML,
1448 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1449
1450 # Wordpress "YouTube Video Importer" plugin
1451 matches = re.findall(r'''(?x)<div[^>]+
1452 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1453 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1454 entries.extend(m[-1] for m in matches)
1455
1456 return entries
1457
1458 @staticmethod
1459 def _extract_url(webpage):
1460 urls = YoutubeIE._extract_urls(webpage)
1461 return urls[0] if urls else None
1462
1463 @classmethod
1464 def extract_id(cls, url):
1465 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1466 if mobj is None:
1467 raise ExtractorError('Invalid URL: %s' % url)
1468 video_id = mobj.group(2)
1469 return video_id
1470
1471 def _extract_annotations(self, video_id):
1472 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1473 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1474
1475 @staticmethod
1476 def _extract_chapters(description, duration):
1477 if not description:
1478 return None
1479 chapter_lines = re.findall(
1480 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1481 description)
1482 if not chapter_lines:
1483 return None
1484 chapters = []
1485 for next_num, (chapter_line, time_point) in enumerate(
1486 chapter_lines, start=1):
1487 start_time = parse_duration(time_point)
1488 if start_time is None:
1489 continue
1490 if start_time > duration:
1491 break
1492 end_time = (duration if next_num == len(chapter_lines)
1493 else parse_duration(chapter_lines[next_num][1]))
1494 if end_time is None:
1495 continue
1496 if end_time > duration:
1497 end_time = duration
1498 if start_time > end_time:
1499 break
1500 chapter_title = re.sub(
1501 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1502 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1503 chapters.append({
1504 'start_time': start_time,
1505 'end_time': end_time,
1506 'title': chapter_title,
1507 })
1508 return chapters
1509
1510 def _real_extract(self, url):
1511 url, smuggled_data = unsmuggle_url(url, {})
1512
1513 proto = (
1514 'http' if self._downloader.params.get('prefer_insecure', False)
1515 else 'https')
1516
1517 start_time = None
1518 end_time = None
1519 parsed_url = compat_urllib_parse_urlparse(url)
1520 for component in [parsed_url.fragment, parsed_url.query]:
1521 query = compat_parse_qs(component)
1522 if start_time is None and 't' in query:
1523 start_time = parse_duration(query['t'][0])
1524 if start_time is None and 'start' in query:
1525 start_time = parse_duration(query['start'][0])
1526 if end_time is None and 'end' in query:
1527 end_time = parse_duration(query['end'][0])
1528
1529 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1530 mobj = re.search(self._NEXT_URL_RE, url)
1531 if mobj:
1532 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1533 video_id = self.extract_id(url)
1534
1535 # Get video webpage
1536 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1537 video_webpage = self._download_webpage(url, video_id)
1538
1539 # Attempt to extract SWF player URL
1540 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1541 if mobj is not None:
1542 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1543 else:
1544 player_url = None
1545
1546 dash_mpds = []
1547
1548 def add_dash_mpd(video_info):
1549 dash_mpd = video_info.get('dashmpd')
1550 if dash_mpd and dash_mpd[0] not in dash_mpds:
1551 dash_mpds.append(dash_mpd[0])
1552
1553 def add_dash_mpd_pr(pl_response):
1554 dash_mpd = url_or_none(try_get(
1555 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1556 compat_str))
1557 if dash_mpd and dash_mpd not in dash_mpds:
1558 dash_mpds.append(dash_mpd)
1559
1560 is_live = None
1561 view_count = None
1562
1563 def extract_view_count(v_info):
1564 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1565
1566 player_response = {}
1567
1568 # Get video info
1569 embed_webpage = None
1570 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1571 age_gate = True
1572 # We simulate the access to the video from www.youtube.com/v/{video_id}
1573 # this can be viewed without login into Youtube
1574 url = proto + '://www.youtube.com/embed/%s' % video_id
1575 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1576 data = compat_urllib_parse_urlencode({
1577 'video_id': video_id,
1578 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1579 'sts': self._search_regex(
1580 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1581 })
1582 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1583 video_info_webpage = self._download_webpage(
1584 video_info_url, video_id,
1585 note='Refetching age-gated info webpage',
1586 errnote='unable to download video info webpage')
1587 video_info = compat_parse_qs(video_info_webpage)
1588 add_dash_mpd(video_info)
1589 else:
1590 age_gate = False
1591 video_info = None
1592 sts = None
1593 # Try looking directly into the video webpage
1594 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1595 if ytplayer_config:
1596 args = ytplayer_config['args']
1597 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1598 # Convert to the same format returned by compat_parse_qs
1599 video_info = dict((k, [v]) for k, v in args.items())
1600 add_dash_mpd(video_info)
1601 # Rental video is not rented but preview is available (e.g.
1602 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1603 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1604 if not video_info and args.get('ypc_vid'):
1605 return self.url_result(
1606 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1607 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1608 is_live = True
1609 sts = ytplayer_config.get('sts')
1610 if not player_response:
1611 pl_response = str_or_none(args.get('player_response'))
1612 if pl_response:
1613 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1614 if isinstance(pl_response, dict):
1615 player_response = pl_response
1616 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1617 add_dash_mpd_pr(player_response)
1618 # We also try looking in get_video_info since it may contain different dashmpd
1619 # URL that points to a DASH manifest with possibly different itag set (some itags
1620 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1621 # manifest pointed by get_video_info's dashmpd).
1622 # The general idea is to take a union of itags of both DASH manifests (for example
1623 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1624 self.report_video_info_webpage_download(video_id)
1625 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1626 query = {
1627 'video_id': video_id,
1628 'ps': 'default',
1629 'eurl': '',
1630 'gl': 'US',
1631 'hl': 'en',
1632 }
1633 if el:
1634 query['el'] = el
1635 if sts:
1636 query['sts'] = sts
1637 video_info_webpage = self._download_webpage(
1638 '%s://www.youtube.com/get_video_info' % proto,
1639 video_id, note=False,
1640 errnote='unable to download video info webpage',
1641 fatal=False, query=query)
1642 if not video_info_webpage:
1643 continue
1644 get_video_info = compat_parse_qs(video_info_webpage)
1645 if not player_response:
1646 pl_response = get_video_info.get('player_response', [None])[0]
1647 if isinstance(pl_response, dict):
1648 player_response = pl_response
1649 add_dash_mpd_pr(player_response)
1650 add_dash_mpd(get_video_info)
1651 if view_count is None:
1652 view_count = extract_view_count(get_video_info)
1653 if not video_info:
1654 video_info = get_video_info
1655 if 'token' in get_video_info:
1656 # Different get_video_info requests may report different results, e.g.
1657 # some may report video unavailability, but some may serve it without
1658 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1659 # the original webpage as well as el=info and el=embedded get_video_info
1660 # requests report video unavailability due to geo restriction while
1661 # el=detailpage succeeds and returns valid data). This is probably
1662 # due to YouTube measures against IP ranges of hosting providers.
1663 # Working around by preferring the first succeeded video_info containing
1664 # the token if no such video_info yet was found.
1665 if 'token' not in video_info:
1666 video_info = get_video_info
1667 break
1668
1669 def extract_unavailable_message():
1670 return self._html_search_regex(
1671 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1672 video_webpage, 'unavailable message', default=None)
1673
1674 if 'token' not in video_info:
1675 if 'reason' in video_info:
1676 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1677 regions_allowed = self._html_search_meta(
1678 'regionsAllowed', video_webpage, default=None)
1679 countries = regions_allowed.split(',') if regions_allowed else None
1680 self.raise_geo_restricted(
1681 msg=video_info['reason'][0], countries=countries)
1682 reason = video_info['reason'][0]
1683 if 'Invalid parameters' in reason:
1684 unavailable_message = extract_unavailable_message()
1685 if unavailable_message:
1686 reason = unavailable_message
1687 raise ExtractorError(
1688 'YouTube said: %s' % reason,
1689 expected=True, video_id=video_id)
1690 else:
1691 raise ExtractorError(
1692 '"token" parameter not in video info for unknown reason',
1693 video_id=video_id)
1694
1695 if video_info.get('license_info'):
1696 raise ExtractorError('This video is DRM protected.', expected=True)
1697
1698 video_details = try_get(
1699 player_response, lambda x: x['videoDetails'], dict) or {}
1700
1701 # title
1702 if 'title' in video_info:
1703 video_title = video_info['title'][0]
1704 elif 'title' in player_response:
1705 video_title = video_details['title']
1706 else:
1707 self._downloader.report_warning('Unable to extract video title')
1708 video_title = '_'
1709
1710 # description
1711 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1712 if video_description:
1713
1714 def replace_url(m):
1715 redir_url = compat_urlparse.urljoin(url, m.group(1))
1716 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1717 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1718 qs = compat_parse_qs(parsed_redir_url.query)
1719 q = qs.get('q')
1720 if q and q[0]:
1721 return q[0]
1722 return redir_url
1723
1724 description_original = video_description = re.sub(r'''(?x)
1725 <a\s+
1726 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1727 (?:title|href)="([^"]+)"\s+
1728 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1729 class="[^"]*"[^>]*>
1730 [^<]+\.{3}\s*
1731 </a>
1732 ''', replace_url, video_description)
1733 video_description = clean_html(video_description)
1734 else:
1735 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1736 if fd_mobj:
1737 video_description = unescapeHTML(fd_mobj.group(1))
1738 else:
1739 video_description = ''
1740
1741 if not smuggled_data.get('force_singlefeed', False):
1742 if not self._downloader.params.get('noplaylist'):
1743 multifeed_metadata_list = try_get(
1744 player_response,
1745 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1746 compat_str) or try_get(
1747 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1748 if multifeed_metadata_list:
1749 entries = []
1750 feed_ids = []
1751 for feed in multifeed_metadata_list.split(','):
1752 # Unquote should take place before split on comma (,) since textual
1753 # fields may contain comma as well (see
1754 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1755 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1756 entries.append({
1757 '_type': 'url_transparent',
1758 'ie_key': 'Youtube',
1759 'url': smuggle_url(
1760 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1761 {'force_singlefeed': True}),
1762 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1763 })
1764 feed_ids.append(feed_data['id'][0])
1765 self.to_screen(
1766 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1767 % (', '.join(feed_ids), video_id))
1768 return self.playlist_result(entries, video_id, video_title, video_description)
1769 else:
1770 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1771
1772 if view_count is None:
1773 view_count = extract_view_count(video_info)
1774 if view_count is None and video_details:
1775 view_count = int_or_none(video_details.get('viewCount'))
1776
1777 # Check for "rental" videos
1778 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1779 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1780
1781 def _extract_filesize(media_url):
1782 return int_or_none(self._search_regex(
1783 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1784
1785 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1786 self.report_rtmp_download()
1787 formats = [{
1788 'format_id': '_rtmp',
1789 'protocol': 'rtmp',
1790 'url': video_info['conn'][0],
1791 'player_url': player_url,
1792 }]
1793 elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1794 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1795 if 'rtmpe%3Dyes' in encoded_url_map:
1796 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1797 formats_spec = {}
1798 fmt_list = video_info.get('fmt_list', [''])[0]
1799 if fmt_list:
1800 for fmt in fmt_list.split(','):
1801 spec = fmt.split('/')
1802 if len(spec) > 1:
1803 width_height = spec[1].split('x')
1804 if len(width_height) == 2:
1805 formats_spec[spec[0]] = {
1806 'resolution': spec[1],
1807 'width': int_or_none(width_height[0]),
1808 'height': int_or_none(width_height[1]),
1809 }
1810 q = qualities(['small', 'medium', 'hd720'])
1811 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1812 if streaming_formats:
1813 for fmt in streaming_formats:
1814 itag = str_or_none(fmt.get('itag'))
1815 if not itag:
1816 continue
1817 quality = fmt.get('quality')
1818 quality_label = fmt.get('qualityLabel') or quality
1819 formats_spec[itag] = {
1820 'asr': int_or_none(fmt.get('audioSampleRate')),
1821 'filesize': int_or_none(fmt.get('contentLength')),
1822 'format_note': quality_label,
1823 'fps': int_or_none(fmt.get('fps')),
1824 'height': int_or_none(fmt.get('height')),
1825 'quality': q(quality),
1826 # bitrate for itag 43 is always 2147483647
1827 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1828 'width': int_or_none(fmt.get('width')),
1829 }
1830 formats = []
1831 for url_data_str in encoded_url_map.split(','):
1832 url_data = compat_parse_qs(url_data_str)
1833 if 'itag' not in url_data or 'url' not in url_data:
1834 continue
1835 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1836 # Unsupported FORMAT_STREAM_TYPE_OTF
1837 if stream_type == 3:
1838 continue
1839 format_id = url_data['itag'][0]
1840 url = url_data['url'][0]
1841
1842 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1843 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1844 jsplayer_url_json = self._search_regex(
1845 ASSETS_RE,
1846 embed_webpage if age_gate else video_webpage,
1847 'JS player URL (1)', default=None)
1848 if not jsplayer_url_json and not age_gate:
1849 # We need the embed website after all
1850 if embed_webpage is None:
1851 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1852 embed_webpage = self._download_webpage(
1853 embed_url, video_id, 'Downloading embed webpage')
1854 jsplayer_url_json = self._search_regex(
1855 ASSETS_RE, embed_webpage, 'JS player URL')
1856
1857 player_url = json.loads(jsplayer_url_json)
1858 if player_url is None:
1859 player_url_json = self._search_regex(
1860 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1861 video_webpage, 'age gate player URL')
1862 player_url = json.loads(player_url_json)
1863
1864 if 'sig' in url_data:
1865 url += '&signature=' + url_data['sig'][0]
1866 elif 's' in url_data:
1867 encrypted_sig = url_data['s'][0]
1868
1869 if self._downloader.params.get('verbose'):
1870 if player_url is None:
1871 player_version = 'unknown'
1872 player_desc = 'unknown'
1873 else:
1874 if player_url.endswith('swf'):
1875 player_version = self._search_regex(
1876 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1877 'flash player', fatal=False)
1878 player_desc = 'flash player %s' % player_version
1879 else:
1880 player_version = self._search_regex(
1881 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1882 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1883 player_url,
1884 'html5 player', fatal=False)
1885 player_desc = 'html5 player %s' % player_version
1886
1887 parts_sizes = self._signature_cache_id(encrypted_sig)
1888 self.to_screen('{%s} signature length %s, %s' %
1889 (format_id, parts_sizes, player_desc))
1890
1891 signature = self._decrypt_signature(
1892 encrypted_sig, video_id, player_url, age_gate)
1893 url += '&signature=' + signature
1894 if 'ratebypass' not in url:
1895 url += '&ratebypass=yes'
1896
1897 dct = {
1898 'format_id': format_id,
1899 'url': url,
1900 'player_url': player_url,
1901 }
1902 if format_id in self._formats:
1903 dct.update(self._formats[format_id])
1904 if format_id in formats_spec:
1905 dct.update(formats_spec[format_id])
1906
1907 # Some itags are not included in DASH manifest thus corresponding formats will
1908 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
1909 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1910 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1911 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1912
1913 filesize = int_or_none(url_data.get(
1914 'clen', [None])[0]) or _extract_filesize(url)
1915
1916 quality = url_data.get('quality', [None])[0]
1917
1918 more_fields = {
1919 'filesize': filesize,
1920 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1921 'width': width,
1922 'height': height,
1923 'fps': int_or_none(url_data.get('fps', [None])[0]),
1924 'format_note': url_data.get('quality_label', [None])[0] or quality,
1925 'quality': q(quality),
1926 }
1927 for key, value in more_fields.items():
1928 if value:
1929 dct[key] = value
1930 type_ = url_data.get('type', [None])[0]
1931 if type_:
1932 type_split = type_.split(';')
1933 kind_ext = type_split[0].split('/')
1934 if len(kind_ext) == 2:
1935 kind, _ = kind_ext
1936 dct['ext'] = mimetype2ext(type_split[0])
1937 if kind in ('audio', 'video'):
1938 codecs = None
1939 for mobj in re.finditer(
1940 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1941 if mobj.group('key') == 'codecs':
1942 codecs = mobj.group('val')
1943 break
1944 if codecs:
1945 dct.update(parse_codecs(codecs))
1946 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1947 dct['downloader_options'] = {
1948 # Youtube throttles chunks >~10M
1949 'http_chunk_size': 10485760,
1950 }
1951 formats.append(dct)
1952 else:
1953 manifest_url = (
1954 url_or_none(try_get(
1955 player_response,
1956 lambda x: x['streamingData']['hlsManifestUrl'],
1957 compat_str)) or
1958 url_or_none(try_get(
1959 video_info, lambda x: x['hlsvp'][0], compat_str)))
1960 if manifest_url:
1961 formats = []
1962 m3u8_formats = self._extract_m3u8_formats(
1963 manifest_url, video_id, 'mp4', fatal=False)
1964 for a_format in m3u8_formats:
1965 itag = self._search_regex(
1966 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1967 if itag:
1968 a_format['format_id'] = itag
1969 if itag in self._formats:
1970 dct = self._formats[itag].copy()
1971 dct.update(a_format)
1972 a_format = dct
1973 a_format['player_url'] = player_url
1974 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1975 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1976 formats.append(a_format)
1977 else:
1978 error_message = clean_html(video_info.get('reason', [None])[0])
1979 if not error_message:
1980 error_message = extract_unavailable_message()
1981 if error_message:
1982 raise ExtractorError(error_message, expected=True)
1983 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
1984
1985 # uploader
1986 video_uploader = try_get(
1987 video_info, lambda x: x['author'][0],
1988 compat_str) or str_or_none(video_details.get('author'))
1989 if video_uploader:
1990 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1991 else:
1992 self._downloader.report_warning('unable to extract uploader name')
1993
1994 # uploader_id
1995 video_uploader_id = None
1996 video_uploader_url = None
1997 mobj = re.search(
1998 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1999 video_webpage)
2000 if mobj is not None:
2001 video_uploader_id = mobj.group('uploader_id')
2002 video_uploader_url = mobj.group('uploader_url')
2003 else:
2004 self._downloader.report_warning('unable to extract uploader nickname')
2005
2006 channel_id = self._html_search_meta(
2007 'channelId', video_webpage, 'channel id')
2008 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2009
2010 # thumbnail image
2011 # We try first to get a high quality image:
2012 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2013 video_webpage, re.DOTALL)
2014 if m_thumb is not None:
2015 video_thumbnail = m_thumb.group(1)
2016 elif 'thumbnail_url' not in video_info:
2017 self._downloader.report_warning('unable to extract video thumbnail')
2018 video_thumbnail = None
2019 else: # don't panic if we can't find it
2020 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2021
2022 # upload date
2023 upload_date = self._html_search_meta(
2024 'datePublished', video_webpage, 'upload date', default=None)
2025 if not upload_date:
2026 upload_date = self._search_regex(
2027 [r'(?s)id="eow-date.*?>(.*?)</span>',
2028 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2029 video_webpage, 'upload date', default=None)
2030 upload_date = unified_strdate(upload_date)
2031
2032 video_license = self._html_search_regex(
2033 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2034 video_webpage, 'license', default=None)
2035
2036 m_music = re.search(
2037 r'''(?x)
2038 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2039 <ul[^>]*>\s*
2040 <li>(?P<title>.+?)
2041 by (?P<creator>.+?)
2042 (?:
2043 \(.+?\)|
2044 <a[^>]*
2045 (?:
2046 \bhref=["\']/red[^>]*>| # drop possible
2047 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2048 )
2049 .*?
2050 )?</li
2051 ''',
2052 video_webpage)
2053 if m_music:
2054 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2055 video_creator = clean_html(m_music.group('creator'))
2056 else:
2057 video_alt_title = video_creator = None
2058
2059 def extract_meta(field):
2060 return self._html_search_regex(
2061 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2062 video_webpage, field, default=None)
2063
2064 track = extract_meta('Song')
2065 artist = extract_meta('Artist')
2066
2067 m_episode = re.search(
2068 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2069 video_webpage)
2070 if m_episode:
2071 series = unescapeHTML(m_episode.group('series'))
2072 season_number = int(m_episode.group('season'))
2073 episode_number = int(m_episode.group('episode'))
2074 else:
2075 series = season_number = episode_number = None
2076
2077 m_cat_container = self._search_regex(
2078 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2079 video_webpage, 'categories', default=None)
2080 if m_cat_container:
2081 category = self._html_search_regex(
2082 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2083 default=None)
2084 video_categories = None if category is None else [category]
2085 else:
2086 video_categories = None
2087
2088 video_tags = [
2089 unescapeHTML(m.group('content'))
2090 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2091
2092 def _extract_count(count_name):
2093 return str_to_int(self._search_regex(
2094 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2095 % re.escape(count_name),
2096 video_webpage, count_name, default=None))
2097
2098 like_count = _extract_count('like')
2099 dislike_count = _extract_count('dislike')
2100
2101 if view_count is None:
2102 view_count = str_to_int(self._search_regex(
2103 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2104 'view count', default=None))
2105
2106 # subtitles
2107 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2108 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2109
2110 video_duration = try_get(
2111 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2112 if not video_duration:
2113 video_duration = int_or_none(video_details.get('lengthSeconds'))
2114 if not video_duration:
2115 video_duration = parse_duration(self._html_search_meta(
2116 'duration', video_webpage, 'video duration'))
2117
2118 # annotations
2119 video_annotations = None
2120 if self._downloader.params.get('writeannotations', False):
2121 video_annotations = self._extract_annotations(video_id)
2122
2123 chapters = self._extract_chapters(description_original, video_duration)
2124
2125 # Look for the DASH manifest
2126 if self._downloader.params.get('youtube_include_dash_manifest', True):
2127 dash_mpd_fatal = True
2128 for mpd_url in dash_mpds:
2129 dash_formats = {}
2130 try:
2131 def decrypt_sig(mobj):
2132 s = mobj.group(1)
2133 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2134 return '/signature/%s' % dec_s
2135
2136 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2137
2138 for df in self._extract_mpd_formats(
2139 mpd_url, video_id, fatal=dash_mpd_fatal,
2140 formats_dict=self._formats):
2141 if not df.get('filesize'):
2142 df['filesize'] = _extract_filesize(df['url'])
2143 # Do not overwrite DASH format found in some previous DASH manifest
2144 if df['format_id'] not in dash_formats:
2145 dash_formats[df['format_id']] = df
2146 # Additional DASH manifests may end up in HTTP Error 403 therefore
2147 # allow them to fail without bug report message if we already have
2148 # some DASH manifest succeeded. This is temporary workaround to reduce
2149 # burst of bug reports until we figure out the reason and whether it
2150 # can be fixed at all.
2151 dash_mpd_fatal = False
2152 except (ExtractorError, KeyError) as e:
2153 self.report_warning(
2154 'Skipping DASH manifest: %r' % e, video_id)
2155 if dash_formats:
2156 # Remove the formats we found through non-DASH, they
2157 # contain less info and it can be wrong, because we use
2158 # fixed values (for example the resolution). See
2159 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2160 # example.
2161 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2162 formats.extend(dash_formats.values())
2163
2164 # Check for malformed aspect ratio
2165 stretched_m = re.search(
2166 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2167 video_webpage)
2168 if stretched_m:
2169 w = float(stretched_m.group('w'))
2170 h = float(stretched_m.group('h'))
2171 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2172 # We will only process correct ratios.
2173 if w > 0 and h > 0:
2174 ratio = w / h
2175 for f in formats:
2176 if f.get('vcodec') != 'none':
2177 f['stretched_ratio'] = ratio
2178
2179 self._sort_formats(formats)
2180
2181 self.mark_watched(video_id, video_info, player_response)
2182
2183 return {
2184 'id': video_id,
2185 'uploader': video_uploader,
2186 'uploader_id': video_uploader_id,
2187 'uploader_url': video_uploader_url,
2188 'channel_id': channel_id,
2189 'channel_url': channel_url,
2190 'upload_date': upload_date,
2191 'license': video_license,
2192 'creator': video_creator or artist,
2193 'title': video_title,
2194 'alt_title': video_alt_title or track,
2195 'thumbnail': video_thumbnail,
2196 'description': video_description,
2197 'categories': video_categories,
2198 'tags': video_tags,
2199 'subtitles': video_subtitles,
2200 'automatic_captions': automatic_captions,
2201 'duration': video_duration,
2202 'age_limit': 18 if age_gate else 0,
2203 'annotations': video_annotations,
2204 'chapters': chapters,
2205 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2206 'view_count': view_count,
2207 'like_count': like_count,
2208 'dislike_count': dislike_count,
2209 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2210 'formats': formats,
2211 'is_live': is_live,
2212 'start_time': start_time,
2213 'end_time': end_time,
2214 'series': series,
2215 'season_number': season_number,
2216 'episode_number': episode_number,
2217 'track': track,
2218 'artist': artist,
2219 }
2220
2221
2222 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2223 IE_DESC = 'YouTube.com playlists'
2224 _VALID_URL = r"""(?x)(?:
2225 (?:https?://)?
2226 (?:\w+\.)?
2227 (?:
2228 (?:
2229 youtube\.com|
2230 invidio\.us
2231 )
2232 /
2233 (?:
2234 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2235 \? (?:.*?[&;])*? (?:p|a|list)=
2236 | p/
2237 )|
2238 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2239 )
2240 (
2241 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2242 # Top tracks, they can also include dots
2243 |(?:MC)[\w\.]*
2244 )
2245 .*
2246 |
2247 (%(playlist_id)s)
2248 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2249 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2250 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2251 IE_NAME = 'youtube:playlist'
2252 _TESTS = [{
2253 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2254 'info_dict': {
2255 'title': 'ytdl test PL',
2256 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2257 },
2258 'playlist_count': 3,
2259 }, {
2260 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2261 'info_dict': {
2262 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2263 'title': 'YDL_Empty_List',
2264 },
2265 'playlist_count': 0,
2266 'skip': 'This playlist is private',
2267 }, {
2268 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2269 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2270 'info_dict': {
2271 'title': '29C3: Not my department',
2272 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2273 },
2274 'playlist_count': 95,
2275 }, {
2276 'note': 'issue #673',
2277 'url': 'PLBB231211A4F62143',
2278 'info_dict': {
2279 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2280 'id': 'PLBB231211A4F62143',
2281 },
2282 'playlist_mincount': 26,
2283 }, {
2284 'note': 'Large playlist',
2285 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2286 'info_dict': {
2287 'title': 'Uploads from Cauchemar',
2288 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2289 },
2290 'playlist_mincount': 799,
2291 }, {
2292 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2293 'info_dict': {
2294 'title': 'YDL_safe_search',
2295 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2296 },
2297 'playlist_count': 2,
2298 'skip': 'This playlist is private',
2299 }, {
2300 'note': 'embedded',
2301 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2302 'playlist_count': 4,
2303 'info_dict': {
2304 'title': 'JODA15',
2305 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2306 }
2307 }, {
2308 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2309 'playlist_mincount': 485,
2310 'info_dict': {
2311 'title': '2017 華語最新單曲 (2/24更新)',
2312 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2313 }
2314 }, {
2315 'note': 'Embedded SWF player',
2316 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2317 'playlist_count': 4,
2318 'info_dict': {
2319 'title': 'JODA7',
2320 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2321 }
2322 }, {
2323 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2324 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2325 'info_dict': {
2326 'title': 'Uploads from Interstellar Movie',
2327 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2328 },
2329 'playlist_mincount': 21,
2330 }, {
2331 # Playlist URL that does not actually serve a playlist
2332 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2333 'info_dict': {
2334 'id': 'FqZTN594JQw',
2335 'ext': 'webm',
2336 'title': "Smiley's People 01 detective, Adventure Series, Action",
2337 'uploader': 'STREEM',
2338 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2339 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2340 'upload_date': '20150526',
2341 'license': 'Standard YouTube License',
2342 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2343 'categories': ['People & Blogs'],
2344 'tags': list,
2345 'view_count': int,
2346 'like_count': int,
2347 'dislike_count': int,
2348 },
2349 'params': {
2350 'skip_download': True,
2351 },
2352 'add_ie': [YoutubeIE.ie_key()],
2353 }, {
2354 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2355 'info_dict': {
2356 'id': 'yeWKywCrFtk',
2357 'ext': 'mp4',
2358 'title': 'Small Scale Baler and Braiding Rugs',
2359 'uploader': 'Backus-Page House Museum',
2360 'uploader_id': 'backuspagemuseum',
2361 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2362 'upload_date': '20161008',
2363 'license': 'Standard YouTube License',
2364 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2365 'categories': ['Nonprofits & Activism'],
2366 'tags': list,
2367 'like_count': int,
2368 'dislike_count': int,
2369 },
2370 'params': {
2371 'noplaylist': True,
2372 'skip_download': True,
2373 },
2374 }, {
2375 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2376 'only_matching': True,
2377 }, {
2378 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2379 'only_matching': True,
2380 }, {
2381 # music album playlist
2382 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2383 'only_matching': True,
2384 }, {
2385 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2386 'only_matching': True,
2387 }]
2388
2389 def _real_initialize(self):
2390 self._login()
2391
2392 def _extract_mix(self, playlist_id):
2393 # The mixes are generated from a single video
2394 # the id of the playlist is just 'RD' + video_id
2395 ids = []
2396 last_id = playlist_id[-11:]
2397 for n in itertools.count(1):
2398 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2399 webpage = self._download_webpage(
2400 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2401 new_ids = orderedSet(re.findall(
2402 r'''(?xs)data-video-username=".*?".*?
2403 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2404 webpage))
2405 # Fetch new pages until all the videos are repeated, it seems that
2406 # there are always 51 unique videos.
2407 new_ids = [_id for _id in new_ids if _id not in ids]
2408 if not new_ids:
2409 break
2410 ids.extend(new_ids)
2411 last_id = ids[-1]
2412
2413 url_results = self._ids_to_results(ids)
2414
2415 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2416 title_span = (
2417 search_title('playlist-title') or
2418 search_title('title long-title') or
2419 search_title('title'))
2420 title = clean_html(title_span)
2421
2422 return self.playlist_result(url_results, playlist_id, title)
2423
2424 def _extract_playlist(self, playlist_id):
2425 url = self._TEMPLATE_URL % playlist_id
2426 page = self._download_webpage(url, playlist_id)
2427
2428 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2429 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2430 match = match.strip()
2431 # Check if the playlist exists or is private
2432 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2433 if mobj:
2434 reason = mobj.group('reason')
2435 message = 'This playlist %s' % reason
2436 if 'private' in reason:
2437 message += ', use --username or --netrc to access it'
2438 message += '.'
2439 raise ExtractorError(message, expected=True)
2440 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2441 raise ExtractorError(
2442 'Invalid parameters. Maybe URL is incorrect.',
2443 expected=True)
2444 elif re.match(r'[^<]*Choose your language[^<]*', match):
2445 continue
2446 else:
2447 self.report_warning('Youtube gives an alert message: ' + match)
2448
2449 playlist_title = self._html_search_regex(
2450 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2451 page, 'title', default=None)
2452
2453 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2454 uploader = self._search_regex(
2455 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2456 page, 'uploader', default=None)
2457 mobj = re.search(
2458 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2459 page)
2460 if mobj:
2461 uploader_id = mobj.group('uploader_id')
2462 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2463 else:
2464 uploader_id = uploader_url = None
2465
2466 has_videos = True
2467
2468 if not playlist_title:
2469 try:
2470 # Some playlist URLs don't actually serve a playlist (e.g.
2471 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2472 next(self._entries(page, playlist_id))
2473 except StopIteration:
2474 has_videos = False
2475
2476 playlist = self.playlist_result(
2477 self._entries(page, playlist_id), playlist_id, playlist_title)
2478 playlist.update({
2479 'uploader': uploader,
2480 'uploader_id': uploader_id,
2481 'uploader_url': uploader_url,
2482 })
2483
2484 return has_videos, playlist
2485
2486 def _check_download_just_video(self, url, playlist_id):
2487 # Check if it's a video-specific URL
2488 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2489 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2490 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2491 'video id', default=None)
2492 if video_id:
2493 if self._downloader.params.get('noplaylist'):
2494 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2495 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2496 else:
2497 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2498 return video_id, None
2499 return None, None
2500
2501 def _real_extract(self, url):
2502 # Extract playlist id
2503 mobj = re.match(self._VALID_URL, url)
2504 if mobj is None:
2505 raise ExtractorError('Invalid URL: %s' % url)
2506 playlist_id = mobj.group(1) or mobj.group(2)
2507
2508 video_id, video = self._check_download_just_video(url, playlist_id)
2509 if video:
2510 return video
2511
2512 if playlist_id.startswith(('RD', 'UL', 'PU')):
2513 # Mixes require a custom extraction process
2514 return self._extract_mix(playlist_id)
2515
2516 has_videos, playlist = self._extract_playlist(playlist_id)
2517 if has_videos or not video_id:
2518 return playlist
2519
2520 # Some playlist URLs don't actually serve a playlist (see
2521 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2522 # Fallback to plain video extraction if there is a video id
2523 # along with playlist id.
2524 return self.url_result(video_id, 'Youtube', video_id=video_id)
2525
2526
2527 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2528 IE_DESC = 'YouTube.com channels'
2529 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2530 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2531 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2532 IE_NAME = 'youtube:channel'
2533 _TESTS = [{
2534 'note': 'paginated channel',
2535 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2536 'playlist_mincount': 91,
2537 'info_dict': {
2538 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2539 'title': 'Uploads from lex will',
2540 }
2541 }, {
2542 'note': 'Age restricted channel',
2543 # from https://www.youtube.com/user/DeusExOfficial
2544 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2545 'playlist_mincount': 64,
2546 'info_dict': {
2547 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2548 'title': 'Uploads from Deus Ex',
2549 },
2550 }, {
2551 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2552 'only_matching': True,
2553 }]
2554
2555 @classmethod
2556 def suitable(cls, url):
2557 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2558 else super(YoutubeChannelIE, cls).suitable(url))
2559
2560 def _build_template_url(self, url, channel_id):
2561 return self._TEMPLATE_URL % channel_id
2562
2563 def _real_extract(self, url):
2564 channel_id = self._match_id(url)
2565
2566 url = self._build_template_url(url, channel_id)
2567
2568 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2569 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2570 # otherwise fallback on channel by page extraction
2571 channel_page = self._download_webpage(
2572 url + '?view=57', channel_id,
2573 'Downloading channel page', fatal=False)
2574 if channel_page is False:
2575 channel_playlist_id = False
2576 else:
2577 channel_playlist_id = self._html_search_meta(
2578 'channelId', channel_page, 'channel id', default=None)
2579 if not channel_playlist_id:
2580 channel_url = self._html_search_meta(
2581 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2582 channel_page, 'channel url', default=None)
2583 if channel_url:
2584 channel_playlist_id = self._search_regex(
2585 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2586 channel_url, 'channel id', default=None)
2587 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2588 playlist_id = 'UU' + channel_playlist_id[2:]
2589 return self.url_result(
2590 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2591
2592 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2593 autogenerated = re.search(r'''(?x)
2594 class="[^"]*?(?:
2595 channel-header-autogenerated-label|
2596 yt-channel-title-autogenerated
2597 )[^"]*"''', channel_page) is not None
2598
2599 if autogenerated:
2600 # The videos are contained in a single page
2601 # the ajax pages can't be used, they are empty
2602 entries = [
2603 self.url_result(
2604 video_id, 'Youtube', video_id=video_id,
2605 video_title=video_title)
2606 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2607 return self.playlist_result(entries, channel_id)
2608
2609 try:
2610 next(self._entries(channel_page, channel_id))
2611 except StopIteration:
2612 alert_message = self._html_search_regex(
2613 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2614 channel_page, 'alert', default=None, group='alert')
2615 if alert_message:
2616 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2617
2618 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2619
2620
2621 class YoutubeUserIE(YoutubeChannelIE):
2622 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2623 _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_-]+)'
2624 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2625 IE_NAME = 'youtube:user'
2626
2627 _TESTS = [{
2628 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2629 'playlist_mincount': 320,
2630 'info_dict': {
2631 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2632 'title': 'Uploads from The Linux Foundation',
2633 }
2634 }, {
2635 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2636 # but not https://www.youtube.com/user/12minuteathlete/videos
2637 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2638 'playlist_mincount': 249,
2639 'info_dict': {
2640 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2641 'title': 'Uploads from 12 Minute Athlete',
2642 }
2643 }, {
2644 'url': 'ytuser:phihag',
2645 'only_matching': True,
2646 }, {
2647 'url': 'https://www.youtube.com/c/gametrailers',
2648 'only_matching': True,
2649 }, {
2650 'url': 'https://www.youtube.com/gametrailers',
2651 'only_matching': True,
2652 }, {
2653 # This channel is not available, geo restricted to JP
2654 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2655 'only_matching': True,
2656 }]
2657
2658 @classmethod
2659 def suitable(cls, url):
2660 # Don't return True if the url can be extracted with other youtube
2661 # extractor, the regex would is too permissive and it would match.
2662 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2663 if any(ie.suitable(url) for ie in other_yt_ies):
2664 return False
2665 else:
2666 return super(YoutubeUserIE, cls).suitable(url)
2667
2668 def _build_template_url(self, url, channel_id):
2669 mobj = re.match(self._VALID_URL, url)
2670 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2671
2672
2673 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2674 IE_DESC = 'YouTube.com live streams'
2675 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2676 IE_NAME = 'youtube:live'
2677
2678 _TESTS = [{
2679 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2680 'info_dict': {
2681 'id': 'a48o2S1cPoo',
2682 'ext': 'mp4',
2683 'title': 'The Young Turks - Live Main Show',
2684 'uploader': 'The Young Turks',
2685 'uploader_id': 'TheYoungTurks',
2686 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2687 'upload_date': '20150715',
2688 'license': 'Standard YouTube License',
2689 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2690 'categories': ['News & Politics'],
2691 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2692 'like_count': int,
2693 'dislike_count': int,
2694 },
2695 'params': {
2696 'skip_download': True,
2697 },
2698 }, {
2699 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2700 'only_matching': True,
2701 }, {
2702 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2703 'only_matching': True,
2704 }, {
2705 'url': 'https://www.youtube.com/TheYoungTurks/live',
2706 'only_matching': True,
2707 }]
2708
2709 def _real_extract(self, url):
2710 mobj = re.match(self._VALID_URL, url)
2711 channel_id = mobj.group('id')
2712 base_url = mobj.group('base_url')
2713 webpage = self._download_webpage(url, channel_id, fatal=False)
2714 if webpage:
2715 page_type = self._og_search_property(
2716 'type', webpage, 'page type', default='')
2717 video_id = self._html_search_meta(
2718 'videoId', webpage, 'video id', default=None)
2719 if page_type.startswith('video') and video_id and re.match(
2720 r'^[0-9A-Za-z_-]{11}$', video_id):
2721 return self.url_result(video_id, YoutubeIE.ie_key())
2722 return self.url_result(base_url)
2723
2724
2725 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2726 IE_DESC = 'YouTube.com user/channel playlists'
2727 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2728 IE_NAME = 'youtube:playlists'
2729
2730 _TESTS = [{
2731 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2732 'playlist_mincount': 4,
2733 'info_dict': {
2734 'id': 'ThirstForScience',
2735 'title': 'Thirst for Science',
2736 },
2737 }, {
2738 # with "Load more" button
2739 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2740 'playlist_mincount': 70,
2741 'info_dict': {
2742 'id': 'igorkle1',
2743 'title': 'Игорь Клейнер',
2744 },
2745 }, {
2746 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2747 'playlist_mincount': 17,
2748 'info_dict': {
2749 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2750 'title': 'Chem Player',
2751 },
2752 }]
2753
2754
2755 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2756 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2757
2758
2759 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2760 IE_DESC = 'YouTube.com searches'
2761 # there doesn't appear to be a real limit, for example if you search for
2762 # 'python' you get more than 8.000.000 results
2763 _MAX_RESULTS = float('inf')
2764 IE_NAME = 'youtube:search'
2765 _SEARCH_KEY = 'ytsearch'
2766 _EXTRA_QUERY_ARGS = {}
2767 _TESTS = []
2768
2769 def _get_n_results(self, query, n):
2770 """Get a specified number of results for a query"""
2771
2772 videos = []
2773 limit = n
2774
2775 url_query = {
2776 'search_query': query.encode('utf-8'),
2777 }
2778 url_query.update(self._EXTRA_QUERY_ARGS)
2779 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2780
2781 for pagenum in itertools.count(1):
2782 data = self._download_json(
2783 result_url, video_id='query "%s"' % query,
2784 note='Downloading page %s' % pagenum,
2785 errnote='Unable to download API page',
2786 query={'spf': 'navigate'})
2787 html_content = data[1]['body']['content']
2788
2789 if 'class="search-message' in html_content:
2790 raise ExtractorError(
2791 '[youtube] No video results', expected=True)
2792
2793 new_videos = list(self._process_page(html_content))
2794 videos += new_videos
2795 if not new_videos or len(videos) > limit:
2796 break
2797 next_link = self._html_search_regex(
2798 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2799 html_content, 'next link', default=None)
2800 if next_link is None:
2801 break
2802 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2803
2804 if len(videos) > n:
2805 videos = videos[:n]
2806 return self.playlist_result(videos, query)
2807
2808
2809 class YoutubeSearchDateIE(YoutubeSearchIE):
2810 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2811 _SEARCH_KEY = 'ytsearchdate'
2812 IE_DESC = 'YouTube.com searches, newest videos first'
2813 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2814
2815
2816 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2817 IE_DESC = 'YouTube.com search URLs'
2818 IE_NAME = 'youtube:search_url'
2819 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2820 _TESTS = [{
2821 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2822 'playlist_mincount': 5,
2823 'info_dict': {
2824 'title': 'youtube-dl test video',
2825 }
2826 }, {
2827 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2828 'only_matching': True,
2829 }]
2830
2831 def _real_extract(self, url):
2832 mobj = re.match(self._VALID_URL, url)
2833 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2834 webpage = self._download_webpage(url, query)
2835 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2836
2837
2838 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2839 IE_DESC = 'YouTube.com (multi-season) shows'
2840 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2841 IE_NAME = 'youtube:show'
2842 _TESTS = [{
2843 'url': 'https://www.youtube.com/show/airdisasters',
2844 'playlist_mincount': 5,
2845 'info_dict': {
2846 'id': 'airdisasters',
2847 'title': 'Air Disasters',
2848 }
2849 }]
2850
2851 def _real_extract(self, url):
2852 playlist_id = self._match_id(url)
2853 return super(YoutubeShowIE, self)._real_extract(
2854 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2855
2856
2857 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2858 """
2859 Base class for feed extractors
2860 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2861 """
2862 _LOGIN_REQUIRED = True
2863
2864 @property
2865 def IE_NAME(self):
2866 return 'youtube:%s' % self._FEED_NAME
2867
2868 def _real_initialize(self):
2869 self._login()
2870
2871 def _entries(self, page):
2872 # The extraction process is the same as for playlists, but the regex
2873 # for the video ids doesn't contain an index
2874 ids = []
2875 more_widget_html = content_html = page
2876 for page_num in itertools.count(1):
2877 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2878
2879 # 'recommended' feed has infinite 'load more' and each new portion spins
2880 # the same videos in (sometimes) slightly different order, so we'll check
2881 # for unicity and break when portion has no new videos
2882 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2883 if not new_ids:
2884 break
2885
2886 ids.extend(new_ids)
2887
2888 for entry in self._ids_to_results(new_ids):
2889 yield entry
2890
2891 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2892 if not mobj:
2893 break
2894
2895 more = self._download_json(
2896 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2897 'Downloading page #%s' % page_num,
2898 transform_source=uppercase_escape)
2899 content_html = more['content_html']
2900 more_widget_html = more['load_more_widget_html']
2901
2902 def _real_extract(self, url):
2903 page = self._download_webpage(
2904 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2905 self._PLAYLIST_TITLE)
2906 return self.playlist_result(
2907 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2908
2909
2910 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2911 IE_NAME = 'youtube:watchlater'
2912 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2913 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2914
2915 _TESTS = [{
2916 'url': 'https://www.youtube.com/playlist?list=WL',
2917 'only_matching': True,
2918 }, {
2919 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2920 'only_matching': True,
2921 }]
2922
2923 def _real_extract(self, url):
2924 _, video = self._check_download_just_video(url, 'WL')
2925 if video:
2926 return video
2927 _, playlist = self._extract_playlist('WL')
2928 return playlist
2929
2930
2931 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2932 IE_NAME = 'youtube:favorites'
2933 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2934 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2935 _LOGIN_REQUIRED = True
2936
2937 def _real_extract(self, url):
2938 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2939 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2940 return self.url_result(playlist_id, 'YoutubePlaylist')
2941
2942
2943 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2944 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2945 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2946 _FEED_NAME = 'recommended'
2947 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2948
2949
2950 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2951 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2952 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2953 _FEED_NAME = 'subscriptions'
2954 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2955
2956
2957 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2958 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2959 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2960 _FEED_NAME = 'history'
2961 _PLAYLIST_TITLE = 'Youtube History'
2962
2963
2964 class YoutubeTruncatedURLIE(InfoExtractor):
2965 IE_NAME = 'youtube:truncated_url'
2966 IE_DESC = False # Do not list
2967 _VALID_URL = r'''(?x)
2968 (?:https?://)?
2969 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2970 (?:watch\?(?:
2971 feature=[a-z_]+|
2972 annotation_id=annotation_[^&]+|
2973 x-yt-cl=[0-9]+|
2974 hl=[^&]*|
2975 t=[0-9]+
2976 )?
2977 |
2978 attribution_link\?a=[^&]+
2979 )
2980 $
2981 '''
2982
2983 _TESTS = [{
2984 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2985 'only_matching': True,
2986 }, {
2987 'url': 'https://www.youtube.com/watch?',
2988 'only_matching': True,
2989 }, {
2990 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2991 'only_matching': True,
2992 }, {
2993 'url': 'https://www.youtube.com/watch?feature=foo',
2994 'only_matching': True,
2995 }, {
2996 'url': 'https://www.youtube.com/watch?hl=en-GB',
2997 'only_matching': True,
2998 }, {
2999 'url': 'https://www.youtube.com/watch?t=2372',
3000 'only_matching': True,
3001 }]
3002
3003 def _real_extract(self, url):
3004 raise ExtractorError(
3005 'Did you forget to quote the URL? Remember that & is a meta '
3006 'character in most shells, so you want to put the URL in quotes, '
3007 'like youtube-dl '
3008 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3009 ' or simply youtube-dl BaW_jenozKc .',
3010 expected=True)
3011
3012
3013 class YoutubeTruncatedIDIE(InfoExtractor):
3014 IE_NAME = 'youtube:truncated_id'
3015 IE_DESC = False # Do not list
3016 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3017
3018 _TESTS = [{
3019 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3020 'only_matching': True,
3021 }]
3022
3023 def _real_extract(self, url):
3024 video_id = self._match_id(url)
3025 raise ExtractorError(
3026 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3027 expected=True)