]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[youtube] Fix mark watched (closes #18546)
[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\.)?invidio\.us/|
355 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
356 (?:.*?\#/)? # handle anchor (#/) redirect urls
357 (?: # the various things that can precede the ID:
358 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
359 |(?: # or the v= param in all its forms
360 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
361 (?:\?|\#!?) # the params delimiter ? or # or #!
362 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
363 v=
364 )
365 ))
366 |(?:
367 youtu\.be| # just youtu.be/xxxx
368 vid\.plus| # or vid.plus/xxxx
369 zwearz\.com/watch| # or zwearz.com/watch/xxxx
370 )/
371 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
372 )
373 )? # all until now is optional -> you can pass the naked ID
374 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
375 (?!.*?\blist=
376 (?:
377 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
378 WL # WL are handled by the watch later IE
379 )
380 )
381 (?(1).+)? # if we found the ID, everything can follow
382 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
383 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
384 _formats = {
385 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
386 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
387 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
388 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
389 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
390 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
391 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
392 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
393 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
394 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
395 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
396 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
397 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
398 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
399 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
400 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
401 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
402 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
403
404
405 # 3D videos
406 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
407 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
408 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
409 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
410 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
411 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
412 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
413
414 # Apple HTTP Live Streaming
415 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
416 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
417 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
418 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
419 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
420 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
421 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
422 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
423
424 # DASH mp4 video
425 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
426 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
427 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
428 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
429 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
430 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
431 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
432 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
433 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
434 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
435 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
436 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
437
438 # Dash mp4 audio
439 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
440 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
441 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
442 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
443 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
444 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
445 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
446
447 # Dash webm
448 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
449 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
450 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
451 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
452 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
453 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
454 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
455 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
456 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
457 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
458 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
459 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
460 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
461 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
462 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
463 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
464 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
465 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
466 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
467 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
468 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
469 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
470
471 # Dash webm audio
472 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
473 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
474
475 # Dash webm audio with opus inside
476 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
477 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
478 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
479
480 # RTMP (unnamed)
481 '_rtmp': {'protocol': 'rtmp'},
482 }
483 _SUBTITLE_FORMATS = ('ttml', 'vtt')
484
485 _GEO_BYPASS = False
486
487 IE_NAME = 'youtube'
488 _TESTS = [
489 {
490 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
491 'info_dict': {
492 'id': 'BaW_jenozKc',
493 'ext': 'mp4',
494 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
495 'uploader': 'Philipp Hagemeister',
496 'uploader_id': 'phihag',
497 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
498 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
499 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
500 'upload_date': '20121002',
501 'license': 'Standard YouTube License',
502 '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 .',
503 'categories': ['Science & Technology'],
504 'tags': ['youtube-dl'],
505 'duration': 10,
506 'view_count': int,
507 'like_count': int,
508 'dislike_count': int,
509 'start_time': 1,
510 'end_time': 9,
511 }
512 },
513 {
514 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
515 'note': 'Test generic use_cipher_signature video (#897)',
516 'info_dict': {
517 'id': 'UxxajLWwzqY',
518 'ext': 'mp4',
519 'upload_date': '20120506',
520 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
521 'alt_title': 'I Love It (feat. Charli XCX)',
522 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
523 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
524 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
525 'iconic ep', 'iconic', 'love', 'it'],
526 'duration': 180,
527 'uploader': 'Icona Pop',
528 'uploader_id': 'IconaPop',
529 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
530 'license': 'Standard YouTube License',
531 'creator': 'Icona Pop',
532 'track': 'I Love It (feat. Charli XCX)',
533 'artist': 'Icona Pop',
534 }
535 },
536 {
537 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
538 'note': 'Test VEVO video with age protection (#956)',
539 'info_dict': {
540 'id': '07FYdnEawAQ',
541 'ext': 'mp4',
542 'upload_date': '20130703',
543 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
544 'alt_title': 'Tunnel Vision',
545 'description': 'md5:64249768eec3bc4276236606ea996373',
546 'duration': 419,
547 'uploader': 'justintimberlakeVEVO',
548 'uploader_id': 'justintimberlakeVEVO',
549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
550 'license': 'Standard YouTube License',
551 'creator': 'Justin Timberlake',
552 'track': 'Tunnel Vision',
553 'artist': 'Justin Timberlake',
554 'age_limit': 18,
555 }
556 },
557 {
558 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
559 'note': 'Embed-only video (#1746)',
560 'info_dict': {
561 'id': 'yZIXLfi8CZQ',
562 'ext': 'mp4',
563 'upload_date': '20120608',
564 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
565 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
566 'uploader': 'SET India',
567 'uploader_id': 'setindia',
568 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
569 'license': 'Standard YouTube License',
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 'license': 'Standard YouTube License',
585 '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 .',
586 'categories': ['Science & Technology'],
587 'tags': ['youtube-dl'],
588 'duration': 10,
589 'view_count': int,
590 'like_count': int,
591 'dislike_count': int,
592 },
593 'params': {
594 'skip_download': True,
595 },
596 },
597 {
598 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
599 'note': '256k DASH audio (format 141) via DASH manifest',
600 'info_dict': {
601 'id': 'a9LDPn-MO4I',
602 'ext': 'm4a',
603 'upload_date': '20121002',
604 'uploader_id': '8KVIDEO',
605 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
606 'description': '',
607 'uploader': '8KVIDEO',
608 'license': 'Standard YouTube License',
609 'title': 'UHDTV TEST 8K VIDEO.mp4'
610 },
611 'params': {
612 'youtube_include_dash_manifest': True,
613 'format': '141',
614 },
615 'skip': 'format 141 not served anymore',
616 },
617 # DASH manifest with encrypted signature
618 {
619 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
620 'info_dict': {
621 'id': 'IB3lcPjvWLA',
622 'ext': 'm4a',
623 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
624 'description': 'md5:1900ed86ee514927b9e00fbead6969a5',
625 'duration': 244,
626 'uploader': 'AfrojackVEVO',
627 'uploader_id': 'AfrojackVEVO',
628 'upload_date': '20131011',
629 'license': 'Standard YouTube License',
630 },
631 'params': {
632 'youtube_include_dash_manifest': True,
633 'format': '141/bestaudio[ext=m4a]',
634 },
635 },
636 # JS player signature function name containing $
637 {
638 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
639 'info_dict': {
640 'id': 'nfWlot6h_JM',
641 'ext': 'm4a',
642 'title': 'Taylor Swift - Shake It Off',
643 'alt_title': 'Shake It Off',
644 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
645 'duration': 242,
646 'uploader': 'TaylorSwiftVEVO',
647 'uploader_id': 'TaylorSwiftVEVO',
648 'upload_date': '20140818',
649 'license': 'Standard YouTube License',
650 'creator': 'Taylor Swift',
651 },
652 'params': {
653 'youtube_include_dash_manifest': True,
654 'format': '141/bestaudio[ext=m4a]',
655 },
656 },
657 # Controversy video
658 {
659 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
660 'info_dict': {
661 'id': 'T4XJQO3qol8',
662 'ext': 'mp4',
663 'duration': 219,
664 'upload_date': '20100909',
665 'uploader': 'TJ Kirk',
666 'uploader_id': 'TheAmazingAtheist',
667 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
668 'license': 'Standard YouTube License',
669 'title': 'Burning Everyone\'s Koran',
670 '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',
671 }
672 },
673 # Normal age-gate video (No vevo, embed allowed)
674 {
675 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
676 'info_dict': {
677 'id': 'HtVdAasjOgU',
678 'ext': 'mp4',
679 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
680 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
681 'duration': 142,
682 'uploader': 'The Witcher',
683 'uploader_id': 'WitcherGame',
684 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
685 'upload_date': '20140605',
686 'license': 'Standard YouTube License',
687 'age_limit': 18,
688 },
689 },
690 # Age-gate video with encrypted signature
691 {
692 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
693 'info_dict': {
694 'id': '6kLq3WMV1nU',
695 'ext': 'webm',
696 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
697 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
698 'duration': 246,
699 'uploader': 'LloydVEVO',
700 'uploader_id': 'LloydVEVO',
701 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
702 'upload_date': '20110629',
703 'license': 'Standard YouTube License',
704 'age_limit': 18,
705 },
706 },
707 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
708 # YouTube Red ad is not captured for creator
709 {
710 'url': '__2ABJjxzNo',
711 'info_dict': {
712 'id': '__2ABJjxzNo',
713 'ext': 'mp4',
714 'duration': 266,
715 'upload_date': '20100430',
716 'uploader_id': 'deadmau5',
717 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
718 'creator': 'deadmau5',
719 'description': 'md5:12c56784b8032162bb936a5f76d55360',
720 'uploader': 'deadmau5',
721 'license': 'Standard YouTube License',
722 'title': 'Deadmau5 - Some Chords (HD)',
723 'alt_title': 'Some Chords',
724 },
725 'expected_warnings': [
726 'DASH manifest missing',
727 ]
728 },
729 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
730 {
731 'url': 'lqQg6PlCWgI',
732 'info_dict': {
733 'id': 'lqQg6PlCWgI',
734 'ext': 'mp4',
735 'duration': 6085,
736 'upload_date': '20150827',
737 'uploader_id': 'olympic',
738 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
739 'license': 'Standard YouTube License',
740 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
741 'uploader': 'Olympic',
742 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
743 },
744 'params': {
745 'skip_download': 'requires avconv',
746 }
747 },
748 # Non-square pixels
749 {
750 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
751 'info_dict': {
752 'id': '_b-2C3KPAM0',
753 'ext': 'mp4',
754 'stretched_ratio': 16 / 9.,
755 'duration': 85,
756 'upload_date': '20110310',
757 'uploader_id': 'AllenMeow',
758 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
759 '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',
760 'uploader': '孫ᄋᄅ',
761 'license': 'Standard YouTube License',
762 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
763 },
764 },
765 # url_encoded_fmt_stream_map is empty string
766 {
767 'url': 'qEJwOuvDf7I',
768 'info_dict': {
769 'id': 'qEJwOuvDf7I',
770 'ext': 'webm',
771 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
772 'description': '',
773 'upload_date': '20150404',
774 'uploader_id': 'spbelect',
775 'uploader': 'Наблюдатели Петербурга',
776 },
777 'params': {
778 'skip_download': 'requires avconv',
779 },
780 'skip': 'This live event has ended.',
781 },
782 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
783 {
784 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
785 'info_dict': {
786 'id': 'FIl7x6_3R5Y',
787 'ext': 'webm',
788 'title': 'md5:7b81415841e02ecd4313668cde88737a',
789 'description': 'md5:116377fd2963b81ec4ce64b542173306',
790 'duration': 220,
791 'upload_date': '20150625',
792 'uploader_id': 'dorappi2000',
793 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
794 'uploader': 'dorappi2000',
795 'license': 'Standard YouTube License',
796 'formats': 'mincount:31',
797 },
798 'skip': 'not actual anymore',
799 },
800 # DASH manifest with segment_list
801 {
802 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
803 'md5': '8ce563a1d667b599d21064e982ab9e31',
804 'info_dict': {
805 'id': 'CsmdDsKjzN8',
806 'ext': 'mp4',
807 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
808 'uploader': 'Airtek',
809 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
810 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
811 'license': 'Standard YouTube License',
812 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
813 },
814 'params': {
815 'youtube_include_dash_manifest': True,
816 'format': '135', # bestvideo
817 },
818 'skip': 'This live event has ended.',
819 },
820 {
821 # Multifeed videos (multiple cameras), URL is for Main Camera
822 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
823 'info_dict': {
824 'id': 'jqWvoWXjCVs',
825 'title': 'teamPGP: Rocket League Noob Stream',
826 'description': 'md5:dc7872fb300e143831327f1bae3af010',
827 },
828 'playlist': [{
829 'info_dict': {
830 'id': 'jqWvoWXjCVs',
831 'ext': 'mp4',
832 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
833 'description': 'md5:dc7872fb300e143831327f1bae3af010',
834 'duration': 7335,
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': '6h8e8xoXJzg',
844 'ext': 'mp4',
845 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
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': 'PUOgX5z9xZw',
857 'ext': 'mp4',
858 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
859 'description': 'md5:dc7872fb300e143831327f1bae3af010',
860 'duration': 7337,
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 'info_dict': {
869 'id': 'teuwxikvS5k',
870 'ext': 'mp4',
871 'title': 'teamPGP: Rocket League Noob Stream (zim)',
872 'description': 'md5:dc7872fb300e143831327f1bae3af010',
873 'duration': 7334,
874 'upload_date': '20150721',
875 'uploader': 'Beer Games Beer',
876 'uploader_id': 'beergamesbeer',
877 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
878 'license': 'Standard YouTube License',
879 },
880 }],
881 'params': {
882 'skip_download': True,
883 },
884 },
885 {
886 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
887 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
888 'info_dict': {
889 'id': 'gVfLd0zydlo',
890 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
891 },
892 'playlist_count': 2,
893 'skip': 'Not multifeed anymore',
894 },
895 {
896 'url': 'https://vid.plus/FlRa-iH7PGw',
897 'only_matching': True,
898 },
899 {
900 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
901 'only_matching': True,
902 },
903 {
904 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
905 # Also tests cut-off URL expansion in video description (see
906 # https://github.com/rg3/youtube-dl/issues/1892,
907 # https://github.com/rg3/youtube-dl/issues/8164)
908 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
909 'info_dict': {
910 'id': 'lsguqyKfVQg',
911 'ext': 'mp4',
912 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
913 'alt_title': 'Dark Walk - Position Music',
914 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
915 'duration': 133,
916 'upload_date': '20151119',
917 'uploader_id': 'IronSoulElf',
918 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
919 'uploader': 'IronSoulElf',
920 'license': 'Standard YouTube License',
921 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
922 'track': 'Dark Walk - Position Music',
923 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
924 },
925 'params': {
926 'skip_download': True,
927 },
928 },
929 {
930 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
931 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
932 'only_matching': True,
933 },
934 {
935 # Video with yt:stretch=17:0
936 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
937 'info_dict': {
938 'id': 'Q39EVAstoRM',
939 'ext': 'mp4',
940 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
941 'description': 'md5:ee18a25c350637c8faff806845bddee9',
942 'upload_date': '20151107',
943 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
944 'uploader': 'CH GAMER DROID',
945 },
946 'params': {
947 'skip_download': True,
948 },
949 'skip': 'This video does not exist.',
950 },
951 {
952 # Video licensed under Creative Commons
953 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
954 'info_dict': {
955 'id': 'M4gD1WSo5mA',
956 'ext': 'mp4',
957 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
958 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
959 'duration': 721,
960 'upload_date': '20150127',
961 'uploader_id': 'BerkmanCenter',
962 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
963 'uploader': 'The Berkman Klein Center for Internet & Society',
964 'license': 'Creative Commons Attribution license (reuse allowed)',
965 },
966 'params': {
967 'skip_download': True,
968 },
969 },
970 {
971 # Channel-like uploader_url
972 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
973 'info_dict': {
974 'id': 'eQcmzGIKrzg',
975 'ext': 'mp4',
976 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
977 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
978 'duration': 4060,
979 'upload_date': '20151119',
980 'uploader': 'Bernie Sanders',
981 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
982 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
983 'license': 'Creative Commons Attribution license (reuse allowed)',
984 },
985 'params': {
986 'skip_download': True,
987 },
988 },
989 {
990 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
991 'only_matching': True,
992 },
993 {
994 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
995 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
996 'only_matching': True,
997 },
998 {
999 # Rental video preview
1000 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1001 'info_dict': {
1002 'id': 'uGpuVWrhIzE',
1003 'ext': 'mp4',
1004 'title': 'Piku - Trailer',
1005 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1006 'upload_date': '20150811',
1007 'uploader': 'FlixMatrix',
1008 'uploader_id': 'FlixMatrixKaravan',
1009 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1010 'license': 'Standard YouTube License',
1011 },
1012 'params': {
1013 'skip_download': True,
1014 },
1015 'skip': 'This video is not available.',
1016 },
1017 {
1018 # YouTube Red video with episode data
1019 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1020 'info_dict': {
1021 'id': 'iqKdEhx-dD4',
1022 'ext': 'mp4',
1023 'title': 'Isolation - Mind Field (Ep 1)',
1024 'description': 'md5:25b78d2f64ae81719f5c96319889b736',
1025 'duration': 2085,
1026 'upload_date': '20170118',
1027 'uploader': 'Vsauce',
1028 'uploader_id': 'Vsauce',
1029 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1030 'license': 'Standard YouTube License',
1031 'series': 'Mind Field',
1032 'season_number': 1,
1033 'episode_number': 1,
1034 },
1035 'params': {
1036 'skip_download': True,
1037 },
1038 'expected_warnings': [
1039 'Skipping DASH manifest',
1040 ],
1041 },
1042 {
1043 # The following content has been identified by the YouTube community
1044 # as inappropriate or offensive to some audiences.
1045 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1046 'info_dict': {
1047 'id': '6SJNVb0GnPI',
1048 'ext': 'mp4',
1049 'title': 'Race Differences in Intelligence',
1050 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1051 'duration': 965,
1052 'upload_date': '20140124',
1053 'uploader': 'New Century Foundation',
1054 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1055 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1056 'license': 'Standard YouTube License',
1057 },
1058 'params': {
1059 'skip_download': True,
1060 },
1061 },
1062 {
1063 # itag 212
1064 'url': '1t24XAntNCY',
1065 'only_matching': True,
1066 },
1067 {
1068 # geo restricted to JP
1069 'url': 'sJL6WA-aGkQ',
1070 'only_matching': True,
1071 },
1072 {
1073 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1074 'only_matching': True,
1075 },
1076 {
1077 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1078 'only_matching': True,
1079 },
1080 ]
1081
1082 def __init__(self, *args, **kwargs):
1083 super(YoutubeIE, self).__init__(*args, **kwargs)
1084 self._player_cache = {}
1085
1086 def report_video_info_webpage_download(self, video_id):
1087 """Report attempt to download video info webpage."""
1088 self.to_screen('%s: Downloading video info webpage' % video_id)
1089
1090 def report_information_extraction(self, video_id):
1091 """Report attempt to extract video information."""
1092 self.to_screen('%s: Extracting video information' % video_id)
1093
1094 def report_unavailable_format(self, video_id, format):
1095 """Report extracted video URL."""
1096 self.to_screen('%s: Format %s not available' % (video_id, format))
1097
1098 def report_rtmp_download(self):
1099 """Indicate the download will use the RTMP protocol."""
1100 self.to_screen('RTMP download detected')
1101
1102 def _signature_cache_id(self, example_sig):
1103 """ Return a string representation of a signature """
1104 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1105
1106 def _extract_signature_function(self, video_id, player_url, example_sig):
1107 id_m = re.match(
1108 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1109 player_url)
1110 if not id_m:
1111 raise ExtractorError('Cannot identify player %r' % player_url)
1112 player_type = id_m.group('ext')
1113 player_id = id_m.group('id')
1114
1115 # Read from filesystem cache
1116 func_id = '%s_%s_%s' % (
1117 player_type, player_id, self._signature_cache_id(example_sig))
1118 assert os.path.basename(func_id) == func_id
1119
1120 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1121 if cache_spec is not None:
1122 return lambda s: ''.join(s[i] for i in cache_spec)
1123
1124 download_note = (
1125 'Downloading player %s' % player_url
1126 if self._downloader.params.get('verbose') else
1127 'Downloading %s player %s' % (player_type, player_id)
1128 )
1129 if player_type == 'js':
1130 code = self._download_webpage(
1131 player_url, video_id,
1132 note=download_note,
1133 errnote='Download of %s failed' % player_url)
1134 res = self._parse_sig_js(code)
1135 elif player_type == 'swf':
1136 urlh = self._request_webpage(
1137 player_url, video_id,
1138 note=download_note,
1139 errnote='Download of %s failed' % player_url)
1140 code = urlh.read()
1141 res = self._parse_sig_swf(code)
1142 else:
1143 assert False, 'Invalid player type %r' % player_type
1144
1145 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1146 cache_res = res(test_string)
1147 cache_spec = [ord(c) for c in cache_res]
1148
1149 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1150 return res
1151
1152 def _print_sig_code(self, func, example_sig):
1153 def gen_sig_code(idxs):
1154 def _genslice(start, end, step):
1155 starts = '' if start == 0 else str(start)
1156 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1157 steps = '' if step == 1 else (':%d' % step)
1158 return 's[%s%s%s]' % (starts, ends, steps)
1159
1160 step = None
1161 # Quelch pyflakes warnings - start will be set when step is set
1162 start = '(Never used)'
1163 for i, prev in zip(idxs[1:], idxs[:-1]):
1164 if step is not None:
1165 if i - prev == step:
1166 continue
1167 yield _genslice(start, prev, step)
1168 step = None
1169 continue
1170 if i - prev in [-1, 1]:
1171 step = i - prev
1172 start = prev
1173 continue
1174 else:
1175 yield 's[%d]' % prev
1176 if step is None:
1177 yield 's[%d]' % i
1178 else:
1179 yield _genslice(start, i, step)
1180
1181 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1182 cache_res = func(test_string)
1183 cache_spec = [ord(c) for c in cache_res]
1184 expr_code = ' + '.join(gen_sig_code(cache_spec))
1185 signature_id_tuple = '(%s)' % (
1186 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1187 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1188 ' return %s\n') % (signature_id_tuple, expr_code)
1189 self.to_screen('Extracted signature function:\n' + code)
1190
1191 def _parse_sig_js(self, jscode):
1192 funcname = self._search_regex(
1193 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1194 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1195 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1196 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1197 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1198 jscode, 'Initial JS player signature function name', group='sig')
1199
1200 jsi = JSInterpreter(jscode)
1201 initial_function = jsi.extract_function(funcname)
1202 return lambda s: initial_function([s])
1203
1204 def _parse_sig_swf(self, file_contents):
1205 swfi = SWFInterpreter(file_contents)
1206 TARGET_CLASSNAME = 'SignatureDecipher'
1207 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1208 initial_function = swfi.extract_function(searched_class, 'decipher')
1209 return lambda s: initial_function([s])
1210
1211 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1212 """Turn the encrypted s field into a working signature"""
1213
1214 if player_url is None:
1215 raise ExtractorError('Cannot decrypt signature without player_url')
1216
1217 if player_url.startswith('//'):
1218 player_url = 'https:' + player_url
1219 elif not re.match(r'https?://', player_url):
1220 player_url = compat_urlparse.urljoin(
1221 'https://www.youtube.com', player_url)
1222 try:
1223 player_id = (player_url, self._signature_cache_id(s))
1224 if player_id not in self._player_cache:
1225 func = self._extract_signature_function(
1226 video_id, player_url, s
1227 )
1228 self._player_cache[player_id] = func
1229 func = self._player_cache[player_id]
1230 if self._downloader.params.get('youtube_print_sig_code'):
1231 self._print_sig_code(func, s)
1232 return func(s)
1233 except Exception as e:
1234 tb = traceback.format_exc()
1235 raise ExtractorError(
1236 'Signature extraction failed: ' + tb, cause=e)
1237
1238 def _get_subtitles(self, video_id, webpage):
1239 try:
1240 subs_doc = self._download_xml(
1241 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1242 video_id, note=False)
1243 except ExtractorError as err:
1244 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1245 return {}
1246
1247 sub_lang_list = {}
1248 for track in subs_doc.findall('track'):
1249 lang = track.attrib['lang_code']
1250 if lang in sub_lang_list:
1251 continue
1252 sub_formats = []
1253 for ext in self._SUBTITLE_FORMATS:
1254 params = compat_urllib_parse_urlencode({
1255 'lang': lang,
1256 'v': video_id,
1257 'fmt': ext,
1258 'name': track.attrib['name'].encode('utf-8'),
1259 })
1260 sub_formats.append({
1261 'url': 'https://www.youtube.com/api/timedtext?' + params,
1262 'ext': ext,
1263 })
1264 sub_lang_list[lang] = sub_formats
1265 if not sub_lang_list:
1266 self._downloader.report_warning('video doesn\'t have subtitles')
1267 return {}
1268 return sub_lang_list
1269
1270 def _get_ytplayer_config(self, video_id, webpage):
1271 patterns = (
1272 # User data may contain arbitrary character sequences that may affect
1273 # JSON extraction with regex, e.g. when '};' is contained the second
1274 # regex won't capture the whole JSON. Yet working around by trying more
1275 # concrete regex first keeping in mind proper quoted string handling
1276 # to be implemented in future that will replace this workaround (see
1277 # https://github.com/rg3/youtube-dl/issues/7468,
1278 # https://github.com/rg3/youtube-dl/pull/7599)
1279 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1280 r';ytplayer\.config\s*=\s*({.+?});',
1281 )
1282 config = self._search_regex(
1283 patterns, webpage, 'ytplayer.config', default=None)
1284 if config:
1285 return self._parse_json(
1286 uppercase_escape(config), video_id, fatal=False)
1287
1288 def _get_automatic_captions(self, video_id, webpage):
1289 """We need the webpage for getting the captions url, pass it as an
1290 argument to speed up the process."""
1291 self.to_screen('%s: Looking for automatic captions' % video_id)
1292 player_config = self._get_ytplayer_config(video_id, webpage)
1293 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1294 if not player_config:
1295 self._downloader.report_warning(err_msg)
1296 return {}
1297 try:
1298 args = player_config['args']
1299 caption_url = args.get('ttsurl')
1300 if caption_url:
1301 timestamp = args['timestamp']
1302 # We get the available subtitles
1303 list_params = compat_urllib_parse_urlencode({
1304 'type': 'list',
1305 'tlangs': 1,
1306 'asrs': 1,
1307 })
1308 list_url = caption_url + '&' + list_params
1309 caption_list = self._download_xml(list_url, video_id)
1310 original_lang_node = caption_list.find('track')
1311 if original_lang_node is None:
1312 self._downloader.report_warning('Video doesn\'t have automatic captions')
1313 return {}
1314 original_lang = original_lang_node.attrib['lang_code']
1315 caption_kind = original_lang_node.attrib.get('kind', '')
1316
1317 sub_lang_list = {}
1318 for lang_node in caption_list.findall('target'):
1319 sub_lang = lang_node.attrib['lang_code']
1320 sub_formats = []
1321 for ext in self._SUBTITLE_FORMATS:
1322 params = compat_urllib_parse_urlencode({
1323 'lang': original_lang,
1324 'tlang': sub_lang,
1325 'fmt': ext,
1326 'ts': timestamp,
1327 'kind': caption_kind,
1328 })
1329 sub_formats.append({
1330 'url': caption_url + '&' + params,
1331 'ext': ext,
1332 })
1333 sub_lang_list[sub_lang] = sub_formats
1334 return sub_lang_list
1335
1336 def make_captions(sub_url, sub_langs):
1337 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1338 caption_qs = compat_parse_qs(parsed_sub_url.query)
1339 captions = {}
1340 for sub_lang in sub_langs:
1341 sub_formats = []
1342 for ext in self._SUBTITLE_FORMATS:
1343 caption_qs.update({
1344 'tlang': [sub_lang],
1345 'fmt': [ext],
1346 })
1347 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1348 query=compat_urllib_parse_urlencode(caption_qs, True)))
1349 sub_formats.append({
1350 'url': sub_url,
1351 'ext': ext,
1352 })
1353 captions[sub_lang] = sub_formats
1354 return captions
1355
1356 # New captions format as of 22.06.2017
1357 player_response = args.get('player_response')
1358 if player_response and isinstance(player_response, compat_str):
1359 player_response = self._parse_json(
1360 player_response, video_id, fatal=False)
1361 if player_response:
1362 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1363 base_url = renderer['captionTracks'][0]['baseUrl']
1364 sub_lang_list = []
1365 for lang in renderer['translationLanguages']:
1366 lang_code = lang.get('languageCode')
1367 if lang_code:
1368 sub_lang_list.append(lang_code)
1369 return make_captions(base_url, sub_lang_list)
1370
1371 # Some videos don't provide ttsurl but rather caption_tracks and
1372 # caption_translation_languages (e.g. 20LmZk1hakA)
1373 # Does not used anymore as of 22.06.2017
1374 caption_tracks = args['caption_tracks']
1375 caption_translation_languages = args['caption_translation_languages']
1376 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1377 sub_lang_list = []
1378 for lang in caption_translation_languages.split(','):
1379 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1380 sub_lang = lang_qs.get('lc', [None])[0]
1381 if sub_lang:
1382 sub_lang_list.append(sub_lang)
1383 return make_captions(caption_url, sub_lang_list)
1384 # An extractor error can be raise by the download process if there are
1385 # no automatic captions but there are subtitles
1386 except (KeyError, IndexError, ExtractorError):
1387 self._downloader.report_warning(err_msg)
1388 return {}
1389
1390 def _mark_watched(self, video_id, video_info, player_response):
1391 playback_url = url_or_none(try_get(
1392 player_response,
1393 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1394 video_info, lambda x: x['videostats_playback_base_url'][0]))
1395 if not playback_url:
1396 return
1397 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1398 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1399
1400 # cpn generation algorithm is reverse engineered from base.js.
1401 # In fact it works even with dummy cpn.
1402 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1403 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1404
1405 qs.update({
1406 'ver': ['2'],
1407 'cpn': [cpn],
1408 })
1409 playback_url = compat_urlparse.urlunparse(
1410 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1411
1412 self._download_webpage(
1413 playback_url, video_id, 'Marking watched',
1414 'Unable to mark watched', fatal=False)
1415
1416 @staticmethod
1417 def _extract_urls(webpage):
1418 # Embedded YouTube player
1419 entries = [
1420 unescapeHTML(mobj.group('url'))
1421 for mobj in re.finditer(r'''(?x)
1422 (?:
1423 <iframe[^>]+?src=|
1424 data-video-url=|
1425 <embed[^>]+?src=|
1426 embedSWF\(?:\s*|
1427 <object[^>]+data=|
1428 new\s+SWFObject\(
1429 )
1430 (["\'])
1431 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1432 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1433 \1''', webpage)]
1434
1435 # lazyYT YouTube embed
1436 entries.extend(list(map(
1437 unescapeHTML,
1438 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1439
1440 # Wordpress "YouTube Video Importer" plugin
1441 matches = re.findall(r'''(?x)<div[^>]+
1442 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1443 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1444 entries.extend(m[-1] for m in matches)
1445
1446 return entries
1447
1448 @staticmethod
1449 def _extract_url(webpage):
1450 urls = YoutubeIE._extract_urls(webpage)
1451 return urls[0] if urls else None
1452
1453 @classmethod
1454 def extract_id(cls, url):
1455 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1456 if mobj is None:
1457 raise ExtractorError('Invalid URL: %s' % url)
1458 video_id = mobj.group(2)
1459 return video_id
1460
1461 def _extract_annotations(self, video_id):
1462 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1463 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1464
1465 @staticmethod
1466 def _extract_chapters(description, duration):
1467 if not description:
1468 return None
1469 chapter_lines = re.findall(
1470 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1471 description)
1472 if not chapter_lines:
1473 return None
1474 chapters = []
1475 for next_num, (chapter_line, time_point) in enumerate(
1476 chapter_lines, start=1):
1477 start_time = parse_duration(time_point)
1478 if start_time is None:
1479 continue
1480 if start_time > duration:
1481 break
1482 end_time = (duration if next_num == len(chapter_lines)
1483 else parse_duration(chapter_lines[next_num][1]))
1484 if end_time is None:
1485 continue
1486 if end_time > duration:
1487 end_time = duration
1488 if start_time > end_time:
1489 break
1490 chapter_title = re.sub(
1491 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1492 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1493 chapters.append({
1494 'start_time': start_time,
1495 'end_time': end_time,
1496 'title': chapter_title,
1497 })
1498 return chapters
1499
1500 def _real_extract(self, url):
1501 url, smuggled_data = unsmuggle_url(url, {})
1502
1503 proto = (
1504 'http' if self._downloader.params.get('prefer_insecure', False)
1505 else 'https')
1506
1507 start_time = None
1508 end_time = None
1509 parsed_url = compat_urllib_parse_urlparse(url)
1510 for component in [parsed_url.fragment, parsed_url.query]:
1511 query = compat_parse_qs(component)
1512 if start_time is None and 't' in query:
1513 start_time = parse_duration(query['t'][0])
1514 if start_time is None and 'start' in query:
1515 start_time = parse_duration(query['start'][0])
1516 if end_time is None and 'end' in query:
1517 end_time = parse_duration(query['end'][0])
1518
1519 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1520 mobj = re.search(self._NEXT_URL_RE, url)
1521 if mobj:
1522 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1523 video_id = self.extract_id(url)
1524
1525 # Get video webpage
1526 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1527 video_webpage = self._download_webpage(url, video_id)
1528
1529 # Attempt to extract SWF player URL
1530 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1531 if mobj is not None:
1532 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1533 else:
1534 player_url = None
1535
1536 dash_mpds = []
1537
1538 def add_dash_mpd(video_info):
1539 dash_mpd = video_info.get('dashmpd')
1540 if dash_mpd and dash_mpd[0] not in dash_mpds:
1541 dash_mpds.append(dash_mpd[0])
1542
1543 is_live = None
1544 view_count = None
1545
1546 def extract_view_count(v_info):
1547 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1548
1549 player_response = {}
1550
1551 # Get video info
1552 embed_webpage = None
1553 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1554 age_gate = True
1555 # We simulate the access to the video from www.youtube.com/v/{video_id}
1556 # this can be viewed without login into Youtube
1557 url = proto + '://www.youtube.com/embed/%s' % video_id
1558 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1559 data = compat_urllib_parse_urlencode({
1560 'video_id': video_id,
1561 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1562 'sts': self._search_regex(
1563 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1564 })
1565 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1566 video_info_webpage = self._download_webpage(
1567 video_info_url, video_id,
1568 note='Refetching age-gated info webpage',
1569 errnote='unable to download video info webpage')
1570 video_info = compat_parse_qs(video_info_webpage)
1571 add_dash_mpd(video_info)
1572 else:
1573 age_gate = False
1574 video_info = None
1575 sts = None
1576 # Try looking directly into the video webpage
1577 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1578 if ytplayer_config:
1579 args = ytplayer_config['args']
1580 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1581 # Convert to the same format returned by compat_parse_qs
1582 video_info = dict((k, [v]) for k, v in args.items())
1583 add_dash_mpd(video_info)
1584 # Rental video is not rented but preview is available (e.g.
1585 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1586 # https://github.com/rg3/youtube-dl/issues/10532)
1587 if not video_info and args.get('ypc_vid'):
1588 return self.url_result(
1589 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1590 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1591 is_live = True
1592 sts = ytplayer_config.get('sts')
1593 if not player_response:
1594 pl_response = str_or_none(args.get('player_response'))
1595 if pl_response:
1596 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1597 if isinstance(pl_response, dict):
1598 player_response = pl_response
1599 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1600 # We also try looking in get_video_info since it may contain different dashmpd
1601 # URL that points to a DASH manifest with possibly different itag set (some itags
1602 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1603 # manifest pointed by get_video_info's dashmpd).
1604 # The general idea is to take a union of itags of both DASH manifests (for example
1605 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1606 self.report_video_info_webpage_download(video_id)
1607 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1608 query = {
1609 'video_id': video_id,
1610 'ps': 'default',
1611 'eurl': '',
1612 'gl': 'US',
1613 'hl': 'en',
1614 }
1615 if el:
1616 query['el'] = el
1617 if sts:
1618 query['sts'] = sts
1619 video_info_webpage = self._download_webpage(
1620 '%s://www.youtube.com/get_video_info' % proto,
1621 video_id, note=False,
1622 errnote='unable to download video info webpage',
1623 fatal=False, query=query)
1624 if not video_info_webpage:
1625 continue
1626 get_video_info = compat_parse_qs(video_info_webpage)
1627 if not player_response:
1628 pl_response = get_video_info.get('player_response', [None])[0]
1629 if isinstance(pl_response, dict):
1630 player_response = pl_response
1631 add_dash_mpd(get_video_info)
1632 if view_count is None:
1633 view_count = extract_view_count(get_video_info)
1634 if not video_info:
1635 video_info = get_video_info
1636 if 'token' in get_video_info:
1637 # Different get_video_info requests may report different results, e.g.
1638 # some may report video unavailability, but some may serve it without
1639 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1640 # the original webpage as well as el=info and el=embedded get_video_info
1641 # requests report video unavailability due to geo restriction while
1642 # el=detailpage succeeds and returns valid data). This is probably
1643 # due to YouTube measures against IP ranges of hosting providers.
1644 # Working around by preferring the first succeeded video_info containing
1645 # the token if no such video_info yet was found.
1646 if 'token' not in video_info:
1647 video_info = get_video_info
1648 break
1649
1650 def extract_unavailable_message():
1651 return self._html_search_regex(
1652 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1653 video_webpage, 'unavailable message', default=None)
1654
1655 if 'token' not in video_info:
1656 if 'reason' in video_info:
1657 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1658 regions_allowed = self._html_search_meta(
1659 'regionsAllowed', video_webpage, default=None)
1660 countries = regions_allowed.split(',') if regions_allowed else None
1661 self.raise_geo_restricted(
1662 msg=video_info['reason'][0], countries=countries)
1663 reason = video_info['reason'][0]
1664 if 'Invalid parameters' in reason:
1665 unavailable_message = extract_unavailable_message()
1666 if unavailable_message:
1667 reason = unavailable_message
1668 raise ExtractorError(
1669 'YouTube said: %s' % reason,
1670 expected=True, video_id=video_id)
1671 else:
1672 raise ExtractorError(
1673 '"token" parameter not in video info for unknown reason',
1674 video_id=video_id)
1675
1676 video_details = try_get(
1677 player_response, lambda x: x['videoDetails'], dict) or {}
1678
1679 # title
1680 if 'title' in video_info:
1681 video_title = video_info['title'][0]
1682 elif 'title' in player_response:
1683 video_title = video_details['title']
1684 else:
1685 self._downloader.report_warning('Unable to extract video title')
1686 video_title = '_'
1687
1688 # description
1689 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1690 if video_description:
1691
1692 def replace_url(m):
1693 redir_url = compat_urlparse.urljoin(url, m.group(1))
1694 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1695 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1696 qs = compat_parse_qs(parsed_redir_url.query)
1697 q = qs.get('q')
1698 if q and q[0]:
1699 return q[0]
1700 return redir_url
1701
1702 description_original = video_description = re.sub(r'''(?x)
1703 <a\s+
1704 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1705 (?:title|href)="([^"]+)"\s+
1706 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1707 class="[^"]*"[^>]*>
1708 [^<]+\.{3}\s*
1709 </a>
1710 ''', replace_url, video_description)
1711 video_description = clean_html(video_description)
1712 else:
1713 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1714 if fd_mobj:
1715 video_description = unescapeHTML(fd_mobj.group(1))
1716 else:
1717 video_description = ''
1718
1719 if not smuggled_data.get('force_singlefeed', False):
1720 if not self._downloader.params.get('noplaylist'):
1721 multifeed_metadata_list = try_get(
1722 player_response,
1723 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1724 compat_str) or try_get(
1725 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1726 if multifeed_metadata_list:
1727 entries = []
1728 feed_ids = []
1729 for feed in multifeed_metadata_list.split(','):
1730 # Unquote should take place before split on comma (,) since textual
1731 # fields may contain comma as well (see
1732 # https://github.com/rg3/youtube-dl/issues/8536)
1733 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1734 entries.append({
1735 '_type': 'url_transparent',
1736 'ie_key': 'Youtube',
1737 'url': smuggle_url(
1738 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1739 {'force_singlefeed': True}),
1740 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1741 })
1742 feed_ids.append(feed_data['id'][0])
1743 self.to_screen(
1744 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1745 % (', '.join(feed_ids), video_id))
1746 return self.playlist_result(entries, video_id, video_title, video_description)
1747 else:
1748 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1749
1750 if view_count is None:
1751 view_count = extract_view_count(video_info)
1752 if view_count is None and video_details:
1753 view_count = int_or_none(video_details.get('viewCount'))
1754
1755 # Check for "rental" videos
1756 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1757 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1758
1759 def _extract_filesize(media_url):
1760 return int_or_none(self._search_regex(
1761 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1762
1763 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1764 self.report_rtmp_download()
1765 formats = [{
1766 'format_id': '_rtmp',
1767 'protocol': 'rtmp',
1768 'url': video_info['conn'][0],
1769 'player_url': player_url,
1770 }]
1771 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):
1772 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1773 if 'rtmpe%3Dyes' in encoded_url_map:
1774 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1775 formats_spec = {}
1776 fmt_list = video_info.get('fmt_list', [''])[0]
1777 if fmt_list:
1778 for fmt in fmt_list.split(','):
1779 spec = fmt.split('/')
1780 if len(spec) > 1:
1781 width_height = spec[1].split('x')
1782 if len(width_height) == 2:
1783 formats_spec[spec[0]] = {
1784 'resolution': spec[1],
1785 'width': int_or_none(width_height[0]),
1786 'height': int_or_none(width_height[1]),
1787 }
1788 q = qualities(['small', 'medium', 'hd720'])
1789 formats = []
1790 for url_data_str in encoded_url_map.split(','):
1791 url_data = compat_parse_qs(url_data_str)
1792 if 'itag' not in url_data or 'url' not in url_data:
1793 continue
1794 format_id = url_data['itag'][0]
1795 url = url_data['url'][0]
1796
1797 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1798 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1799 jsplayer_url_json = self._search_regex(
1800 ASSETS_RE,
1801 embed_webpage if age_gate else video_webpage,
1802 'JS player URL (1)', default=None)
1803 if not jsplayer_url_json and not age_gate:
1804 # We need the embed website after all
1805 if embed_webpage is None:
1806 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1807 embed_webpage = self._download_webpage(
1808 embed_url, video_id, 'Downloading embed webpage')
1809 jsplayer_url_json = self._search_regex(
1810 ASSETS_RE, embed_webpage, 'JS player URL')
1811
1812 player_url = json.loads(jsplayer_url_json)
1813 if player_url is None:
1814 player_url_json = self._search_regex(
1815 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1816 video_webpage, 'age gate player URL')
1817 player_url = json.loads(player_url_json)
1818
1819 if 'sig' in url_data:
1820 url += '&signature=' + url_data['sig'][0]
1821 elif 's' in url_data:
1822 encrypted_sig = url_data['s'][0]
1823
1824 if self._downloader.params.get('verbose'):
1825 if player_url is None:
1826 player_version = 'unknown'
1827 player_desc = 'unknown'
1828 else:
1829 if player_url.endswith('swf'):
1830 player_version = self._search_regex(
1831 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1832 'flash player', fatal=False)
1833 player_desc = 'flash player %s' % player_version
1834 else:
1835 player_version = self._search_regex(
1836 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1837 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1838 player_url,
1839 'html5 player', fatal=False)
1840 player_desc = 'html5 player %s' % player_version
1841
1842 parts_sizes = self._signature_cache_id(encrypted_sig)
1843 self.to_screen('{%s} signature length %s, %s' %
1844 (format_id, parts_sizes, player_desc))
1845
1846 signature = self._decrypt_signature(
1847 encrypted_sig, video_id, player_url, age_gate)
1848 url += '&signature=' + signature
1849 if 'ratebypass' not in url:
1850 url += '&ratebypass=yes'
1851
1852 dct = {
1853 'format_id': format_id,
1854 'url': url,
1855 'player_url': player_url,
1856 }
1857 if format_id in self._formats:
1858 dct.update(self._formats[format_id])
1859 if format_id in formats_spec:
1860 dct.update(formats_spec[format_id])
1861
1862 # Some itags are not included in DASH manifest thus corresponding formats will
1863 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1864 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1865 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1866 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1867
1868 filesize = int_or_none(url_data.get(
1869 'clen', [None])[0]) or _extract_filesize(url)
1870
1871 quality = url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0]
1872
1873 more_fields = {
1874 'filesize': filesize,
1875 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1876 'width': width,
1877 'height': height,
1878 'fps': int_or_none(url_data.get('fps', [None])[0]),
1879 'format_note': quality,
1880 'quality': q(quality),
1881 }
1882 for key, value in more_fields.items():
1883 if value:
1884 dct[key] = value
1885 type_ = url_data.get('type', [None])[0]
1886 if type_:
1887 type_split = type_.split(';')
1888 kind_ext = type_split[0].split('/')
1889 if len(kind_ext) == 2:
1890 kind, _ = kind_ext
1891 dct['ext'] = mimetype2ext(type_split[0])
1892 if kind in ('audio', 'video'):
1893 codecs = None
1894 for mobj in re.finditer(
1895 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1896 if mobj.group('key') == 'codecs':
1897 codecs = mobj.group('val')
1898 break
1899 if codecs:
1900 dct.update(parse_codecs(codecs))
1901 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1902 dct['downloader_options'] = {
1903 # Youtube throttles chunks >~10M
1904 'http_chunk_size': 10485760,
1905 }
1906 formats.append(dct)
1907 elif video_info.get('hlsvp'):
1908 manifest_url = video_info['hlsvp'][0]
1909 formats = []
1910 m3u8_formats = self._extract_m3u8_formats(
1911 manifest_url, video_id, 'mp4', fatal=False)
1912 for a_format in m3u8_formats:
1913 itag = self._search_regex(
1914 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1915 if itag:
1916 a_format['format_id'] = itag
1917 if itag in self._formats:
1918 dct = self._formats[itag].copy()
1919 dct.update(a_format)
1920 a_format = dct
1921 a_format['player_url'] = player_url
1922 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1923 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1924 formats.append(a_format)
1925 else:
1926 error_message = clean_html(video_info.get('reason', [None])[0])
1927 if not error_message:
1928 error_message = extract_unavailable_message()
1929 if error_message:
1930 raise ExtractorError(error_message, expected=True)
1931 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1932
1933 # uploader
1934 video_uploader = try_get(
1935 video_info, lambda x: x['author'][0],
1936 compat_str) or str_or_none(video_details.get('author'))
1937 if video_uploader:
1938 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1939 else:
1940 self._downloader.report_warning('unable to extract uploader name')
1941
1942 # uploader_id
1943 video_uploader_id = None
1944 video_uploader_url = None
1945 mobj = re.search(
1946 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1947 video_webpage)
1948 if mobj is not None:
1949 video_uploader_id = mobj.group('uploader_id')
1950 video_uploader_url = mobj.group('uploader_url')
1951 else:
1952 self._downloader.report_warning('unable to extract uploader nickname')
1953
1954 channel_id = self._html_search_meta(
1955 'channelId', video_webpage, 'channel id')
1956 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
1957
1958 # thumbnail image
1959 # We try first to get a high quality image:
1960 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1961 video_webpage, re.DOTALL)
1962 if m_thumb is not None:
1963 video_thumbnail = m_thumb.group(1)
1964 elif 'thumbnail_url' not in video_info:
1965 self._downloader.report_warning('unable to extract video thumbnail')
1966 video_thumbnail = None
1967 else: # don't panic if we can't find it
1968 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1969
1970 # upload date
1971 upload_date = self._html_search_meta(
1972 'datePublished', video_webpage, 'upload date', default=None)
1973 if not upload_date:
1974 upload_date = self._search_regex(
1975 [r'(?s)id="eow-date.*?>(.*?)</span>',
1976 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1977 video_webpage, 'upload date', default=None)
1978 upload_date = unified_strdate(upload_date)
1979
1980 video_license = self._html_search_regex(
1981 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1982 video_webpage, 'license', default=None)
1983
1984 m_music = re.search(
1985 r'''(?x)
1986 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1987 <ul[^>]*>\s*
1988 <li>(?P<title>.+?)
1989 by (?P<creator>.+?)
1990 (?:
1991 \(.+?\)|
1992 <a[^>]*
1993 (?:
1994 \bhref=["\']/red[^>]*>| # drop possible
1995 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1996 )
1997 .*?
1998 )?</li
1999 ''',
2000 video_webpage)
2001 if m_music:
2002 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2003 video_creator = clean_html(m_music.group('creator'))
2004 else:
2005 video_alt_title = video_creator = None
2006
2007 def extract_meta(field):
2008 return self._html_search_regex(
2009 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2010 video_webpage, field, default=None)
2011
2012 track = extract_meta('Song')
2013 artist = extract_meta('Artist')
2014
2015 m_episode = re.search(
2016 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2017 video_webpage)
2018 if m_episode:
2019 series = m_episode.group('series')
2020 season_number = int(m_episode.group('season'))
2021 episode_number = int(m_episode.group('episode'))
2022 else:
2023 series = season_number = episode_number = None
2024
2025 m_cat_container = self._search_regex(
2026 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2027 video_webpage, 'categories', default=None)
2028 if m_cat_container:
2029 category = self._html_search_regex(
2030 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2031 default=None)
2032 video_categories = None if category is None else [category]
2033 else:
2034 video_categories = None
2035
2036 video_tags = [
2037 unescapeHTML(m.group('content'))
2038 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2039
2040 def _extract_count(count_name):
2041 return str_to_int(self._search_regex(
2042 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2043 % re.escape(count_name),
2044 video_webpage, count_name, default=None))
2045
2046 like_count = _extract_count('like')
2047 dislike_count = _extract_count('dislike')
2048
2049 if view_count is None:
2050 view_count = str_to_int(self._search_regex(
2051 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2052 'view count', default=None))
2053
2054 # subtitles
2055 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2056 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2057
2058 video_duration = try_get(
2059 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2060 if not video_duration:
2061 video_duration = int_or_none(video_details.get('lengthSeconds'))
2062 if not video_duration:
2063 video_duration = parse_duration(self._html_search_meta(
2064 'duration', video_webpage, 'video duration'))
2065
2066 # annotations
2067 video_annotations = None
2068 if self._downloader.params.get('writeannotations', False):
2069 video_annotations = self._extract_annotations(video_id)
2070
2071 chapters = self._extract_chapters(description_original, video_duration)
2072
2073 # Look for the DASH manifest
2074 if self._downloader.params.get('youtube_include_dash_manifest', True):
2075 dash_mpd_fatal = True
2076 for mpd_url in dash_mpds:
2077 dash_formats = {}
2078 try:
2079 def decrypt_sig(mobj):
2080 s = mobj.group(1)
2081 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2082 return '/signature/%s' % dec_s
2083
2084 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2085
2086 for df in self._extract_mpd_formats(
2087 mpd_url, video_id, fatal=dash_mpd_fatal,
2088 formats_dict=self._formats):
2089 if not df.get('filesize'):
2090 df['filesize'] = _extract_filesize(df['url'])
2091 # Do not overwrite DASH format found in some previous DASH manifest
2092 if df['format_id'] not in dash_formats:
2093 dash_formats[df['format_id']] = df
2094 # Additional DASH manifests may end up in HTTP Error 403 therefore
2095 # allow them to fail without bug report message if we already have
2096 # some DASH manifest succeeded. This is temporary workaround to reduce
2097 # burst of bug reports until we figure out the reason and whether it
2098 # can be fixed at all.
2099 dash_mpd_fatal = False
2100 except (ExtractorError, KeyError) as e:
2101 self.report_warning(
2102 'Skipping DASH manifest: %r' % e, video_id)
2103 if dash_formats:
2104 # Remove the formats we found through non-DASH, they
2105 # contain less info and it can be wrong, because we use
2106 # fixed values (for example the resolution). See
2107 # https://github.com/rg3/youtube-dl/issues/5774 for an
2108 # example.
2109 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2110 formats.extend(dash_formats.values())
2111
2112 # Check for malformed aspect ratio
2113 stretched_m = re.search(
2114 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2115 video_webpage)
2116 if stretched_m:
2117 w = float(stretched_m.group('w'))
2118 h = float(stretched_m.group('h'))
2119 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2120 # We will only process correct ratios.
2121 if w > 0 and h > 0:
2122 ratio = w / h
2123 for f in formats:
2124 if f.get('vcodec') != 'none':
2125 f['stretched_ratio'] = ratio
2126
2127 self._sort_formats(formats)
2128
2129 self.mark_watched(video_id, video_info, player_response)
2130
2131 return {
2132 'id': video_id,
2133 'uploader': video_uploader,
2134 'uploader_id': video_uploader_id,
2135 'uploader_url': video_uploader_url,
2136 'channel_id': channel_id,
2137 'channel_url': channel_url,
2138 'upload_date': upload_date,
2139 'license': video_license,
2140 'creator': video_creator or artist,
2141 'title': video_title,
2142 'alt_title': video_alt_title or track,
2143 'thumbnail': video_thumbnail,
2144 'description': video_description,
2145 'categories': video_categories,
2146 'tags': video_tags,
2147 'subtitles': video_subtitles,
2148 'automatic_captions': automatic_captions,
2149 'duration': video_duration,
2150 'age_limit': 18 if age_gate else 0,
2151 'annotations': video_annotations,
2152 'chapters': chapters,
2153 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2154 'view_count': view_count,
2155 'like_count': like_count,
2156 'dislike_count': dislike_count,
2157 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2158 'formats': formats,
2159 'is_live': is_live,
2160 'start_time': start_time,
2161 'end_time': end_time,
2162 'series': series,
2163 'season_number': season_number,
2164 'episode_number': episode_number,
2165 'track': track,
2166 'artist': artist,
2167 }
2168
2169
2170 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2171 IE_DESC = 'YouTube.com playlists'
2172 _VALID_URL = r"""(?x)(?:
2173 (?:https?://)?
2174 (?:\w+\.)?
2175 (?:
2176 (?:
2177 youtube\.com|
2178 invidio\.us
2179 )
2180 /
2181 (?:
2182 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2183 \? (?:.*?[&;])*? (?:p|a|list)=
2184 | p/
2185 )|
2186 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2187 )
2188 (
2189 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2190 # Top tracks, they can also include dots
2191 |(?:MC)[\w\.]*
2192 )
2193 .*
2194 |
2195 (%(playlist_id)s)
2196 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2197 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2198 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2199 IE_NAME = 'youtube:playlist'
2200 _TESTS = [{
2201 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2202 'info_dict': {
2203 'title': 'ytdl test PL',
2204 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2205 },
2206 'playlist_count': 3,
2207 }, {
2208 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2209 'info_dict': {
2210 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2211 'title': 'YDL_Empty_List',
2212 },
2213 'playlist_count': 0,
2214 'skip': 'This playlist is private',
2215 }, {
2216 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2217 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2218 'info_dict': {
2219 'title': '29C3: Not my department',
2220 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2221 },
2222 'playlist_count': 95,
2223 }, {
2224 'note': 'issue #673',
2225 'url': 'PLBB231211A4F62143',
2226 'info_dict': {
2227 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2228 'id': 'PLBB231211A4F62143',
2229 },
2230 'playlist_mincount': 26,
2231 }, {
2232 'note': 'Large playlist',
2233 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2234 'info_dict': {
2235 'title': 'Uploads from Cauchemar',
2236 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2237 },
2238 'playlist_mincount': 799,
2239 }, {
2240 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2241 'info_dict': {
2242 'title': 'YDL_safe_search',
2243 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2244 },
2245 'playlist_count': 2,
2246 'skip': 'This playlist is private',
2247 }, {
2248 'note': 'embedded',
2249 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2250 'playlist_count': 4,
2251 'info_dict': {
2252 'title': 'JODA15',
2253 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2254 }
2255 }, {
2256 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2257 'playlist_mincount': 485,
2258 'info_dict': {
2259 'title': '2017 華語最新單曲 (2/24更新)',
2260 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2261 }
2262 }, {
2263 'note': 'Embedded SWF player',
2264 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2265 'playlist_count': 4,
2266 'info_dict': {
2267 'title': 'JODA7',
2268 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2269 }
2270 }, {
2271 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2272 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2273 'info_dict': {
2274 'title': 'Uploads from Interstellar Movie',
2275 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2276 },
2277 'playlist_mincount': 21,
2278 }, {
2279 # Playlist URL that does not actually serve a playlist
2280 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2281 'info_dict': {
2282 'id': 'FqZTN594JQw',
2283 'ext': 'webm',
2284 'title': "Smiley's People 01 detective, Adventure Series, Action",
2285 'uploader': 'STREEM',
2286 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2287 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2288 'upload_date': '20150526',
2289 'license': 'Standard YouTube License',
2290 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2291 'categories': ['People & Blogs'],
2292 'tags': list,
2293 'view_count': int,
2294 'like_count': int,
2295 'dislike_count': int,
2296 },
2297 'params': {
2298 'skip_download': True,
2299 },
2300 'add_ie': [YoutubeIE.ie_key()],
2301 }, {
2302 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2303 'info_dict': {
2304 'id': 'yeWKywCrFtk',
2305 'ext': 'mp4',
2306 'title': 'Small Scale Baler and Braiding Rugs',
2307 'uploader': 'Backus-Page House Museum',
2308 'uploader_id': 'backuspagemuseum',
2309 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2310 'upload_date': '20161008',
2311 'license': 'Standard YouTube License',
2312 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2313 'categories': ['Nonprofits & Activism'],
2314 'tags': list,
2315 'like_count': int,
2316 'dislike_count': int,
2317 },
2318 'params': {
2319 'noplaylist': True,
2320 'skip_download': True,
2321 },
2322 }, {
2323 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2324 'only_matching': True,
2325 }, {
2326 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2327 'only_matching': True,
2328 }, {
2329 # music album playlist
2330 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2331 'only_matching': True,
2332 }, {
2333 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2334 'only_matching': True,
2335 }]
2336
2337 def _real_initialize(self):
2338 self._login()
2339
2340 def _extract_mix(self, playlist_id):
2341 # The mixes are generated from a single video
2342 # the id of the playlist is just 'RD' + video_id
2343 ids = []
2344 last_id = playlist_id[-11:]
2345 for n in itertools.count(1):
2346 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2347 webpage = self._download_webpage(
2348 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2349 new_ids = orderedSet(re.findall(
2350 r'''(?xs)data-video-username=".*?".*?
2351 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2352 webpage))
2353 # Fetch new pages until all the videos are repeated, it seems that
2354 # there are always 51 unique videos.
2355 new_ids = [_id for _id in new_ids if _id not in ids]
2356 if not new_ids:
2357 break
2358 ids.extend(new_ids)
2359 last_id = ids[-1]
2360
2361 url_results = self._ids_to_results(ids)
2362
2363 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2364 title_span = (
2365 search_title('playlist-title') or
2366 search_title('title long-title') or
2367 search_title('title'))
2368 title = clean_html(title_span)
2369
2370 return self.playlist_result(url_results, playlist_id, title)
2371
2372 def _extract_playlist(self, playlist_id):
2373 url = self._TEMPLATE_URL % playlist_id
2374 page = self._download_webpage(url, playlist_id)
2375
2376 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2377 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2378 match = match.strip()
2379 # Check if the playlist exists or is private
2380 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2381 if mobj:
2382 reason = mobj.group('reason')
2383 message = 'This playlist %s' % reason
2384 if 'private' in reason:
2385 message += ', use --username or --netrc to access it'
2386 message += '.'
2387 raise ExtractorError(message, expected=True)
2388 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2389 raise ExtractorError(
2390 'Invalid parameters. Maybe URL is incorrect.',
2391 expected=True)
2392 elif re.match(r'[^<]*Choose your language[^<]*', match):
2393 continue
2394 else:
2395 self.report_warning('Youtube gives an alert message: ' + match)
2396
2397 playlist_title = self._html_search_regex(
2398 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2399 page, 'title', default=None)
2400
2401 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2402 uploader = self._search_regex(
2403 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2404 page, 'uploader', default=None)
2405 mobj = re.search(
2406 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2407 page)
2408 if mobj:
2409 uploader_id = mobj.group('uploader_id')
2410 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2411 else:
2412 uploader_id = uploader_url = None
2413
2414 has_videos = True
2415
2416 if not playlist_title:
2417 try:
2418 # Some playlist URLs don't actually serve a playlist (e.g.
2419 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2420 next(self._entries(page, playlist_id))
2421 except StopIteration:
2422 has_videos = False
2423
2424 playlist = self.playlist_result(
2425 self._entries(page, playlist_id), playlist_id, playlist_title)
2426 playlist.update({
2427 'uploader': uploader,
2428 'uploader_id': uploader_id,
2429 'uploader_url': uploader_url,
2430 })
2431
2432 return has_videos, playlist
2433
2434 def _check_download_just_video(self, url, playlist_id):
2435 # Check if it's a video-specific URL
2436 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2437 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2438 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2439 'video id', default=None)
2440 if video_id:
2441 if self._downloader.params.get('noplaylist'):
2442 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2443 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2444 else:
2445 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2446 return video_id, None
2447 return None, None
2448
2449 def _real_extract(self, url):
2450 # Extract playlist id
2451 mobj = re.match(self._VALID_URL, url)
2452 if mobj is None:
2453 raise ExtractorError('Invalid URL: %s' % url)
2454 playlist_id = mobj.group(1) or mobj.group(2)
2455
2456 video_id, video = self._check_download_just_video(url, playlist_id)
2457 if video:
2458 return video
2459
2460 if playlist_id.startswith(('RD', 'UL', 'PU')):
2461 # Mixes require a custom extraction process
2462 return self._extract_mix(playlist_id)
2463
2464 has_videos, playlist = self._extract_playlist(playlist_id)
2465 if has_videos or not video_id:
2466 return playlist
2467
2468 # Some playlist URLs don't actually serve a playlist (see
2469 # https://github.com/rg3/youtube-dl/issues/10537).
2470 # Fallback to plain video extraction if there is a video id
2471 # along with playlist id.
2472 return self.url_result(video_id, 'Youtube', video_id=video_id)
2473
2474
2475 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2476 IE_DESC = 'YouTube.com channels'
2477 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2478 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2479 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2480 IE_NAME = 'youtube:channel'
2481 _TESTS = [{
2482 'note': 'paginated channel',
2483 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2484 'playlist_mincount': 91,
2485 'info_dict': {
2486 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2487 'title': 'Uploads from lex will',
2488 }
2489 }, {
2490 'note': 'Age restricted channel',
2491 # from https://www.youtube.com/user/DeusExOfficial
2492 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2493 'playlist_mincount': 64,
2494 'info_dict': {
2495 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2496 'title': 'Uploads from Deus Ex',
2497 },
2498 }, {
2499 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2500 'only_matching': True,
2501 }]
2502
2503 @classmethod
2504 def suitable(cls, url):
2505 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2506 else super(YoutubeChannelIE, cls).suitable(url))
2507
2508 def _build_template_url(self, url, channel_id):
2509 return self._TEMPLATE_URL % channel_id
2510
2511 def _real_extract(self, url):
2512 channel_id = self._match_id(url)
2513
2514 url = self._build_template_url(url, channel_id)
2515
2516 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2517 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2518 # otherwise fallback on channel by page extraction
2519 channel_page = self._download_webpage(
2520 url + '?view=57', channel_id,
2521 'Downloading channel page', fatal=False)
2522 if channel_page is False:
2523 channel_playlist_id = False
2524 else:
2525 channel_playlist_id = self._html_search_meta(
2526 'channelId', channel_page, 'channel id', default=None)
2527 if not channel_playlist_id:
2528 channel_url = self._html_search_meta(
2529 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2530 channel_page, 'channel url', default=None)
2531 if channel_url:
2532 channel_playlist_id = self._search_regex(
2533 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2534 channel_url, 'channel id', default=None)
2535 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2536 playlist_id = 'UU' + channel_playlist_id[2:]
2537 return self.url_result(
2538 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2539
2540 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2541 autogenerated = re.search(r'''(?x)
2542 class="[^"]*?(?:
2543 channel-header-autogenerated-label|
2544 yt-channel-title-autogenerated
2545 )[^"]*"''', channel_page) is not None
2546
2547 if autogenerated:
2548 # The videos are contained in a single page
2549 # the ajax pages can't be used, they are empty
2550 entries = [
2551 self.url_result(
2552 video_id, 'Youtube', video_id=video_id,
2553 video_title=video_title)
2554 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2555 return self.playlist_result(entries, channel_id)
2556
2557 try:
2558 next(self._entries(channel_page, channel_id))
2559 except StopIteration:
2560 alert_message = self._html_search_regex(
2561 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2562 channel_page, 'alert', default=None, group='alert')
2563 if alert_message:
2564 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2565
2566 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2567
2568
2569 class YoutubeUserIE(YoutubeChannelIE):
2570 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2571 _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_-]+)'
2572 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2573 IE_NAME = 'youtube:user'
2574
2575 _TESTS = [{
2576 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2577 'playlist_mincount': 320,
2578 'info_dict': {
2579 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2580 'title': 'Uploads from The Linux Foundation',
2581 }
2582 }, {
2583 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2584 # but not https://www.youtube.com/user/12minuteathlete/videos
2585 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2586 'playlist_mincount': 249,
2587 'info_dict': {
2588 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2589 'title': 'Uploads from 12 Minute Athlete',
2590 }
2591 }, {
2592 'url': 'ytuser:phihag',
2593 'only_matching': True,
2594 }, {
2595 'url': 'https://www.youtube.com/c/gametrailers',
2596 'only_matching': True,
2597 }, {
2598 'url': 'https://www.youtube.com/gametrailers',
2599 'only_matching': True,
2600 }, {
2601 # This channel is not available, geo restricted to JP
2602 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2603 'only_matching': True,
2604 }]
2605
2606 @classmethod
2607 def suitable(cls, url):
2608 # Don't return True if the url can be extracted with other youtube
2609 # extractor, the regex would is too permissive and it would match.
2610 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2611 if any(ie.suitable(url) for ie in other_yt_ies):
2612 return False
2613 else:
2614 return super(YoutubeUserIE, cls).suitable(url)
2615
2616 def _build_template_url(self, url, channel_id):
2617 mobj = re.match(self._VALID_URL, url)
2618 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2619
2620
2621 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2622 IE_DESC = 'YouTube.com live streams'
2623 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2624 IE_NAME = 'youtube:live'
2625
2626 _TESTS = [{
2627 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2628 'info_dict': {
2629 'id': 'a48o2S1cPoo',
2630 'ext': 'mp4',
2631 'title': 'The Young Turks - Live Main Show',
2632 'uploader': 'The Young Turks',
2633 'uploader_id': 'TheYoungTurks',
2634 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2635 'upload_date': '20150715',
2636 'license': 'Standard YouTube License',
2637 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2638 'categories': ['News & Politics'],
2639 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2640 'like_count': int,
2641 'dislike_count': int,
2642 },
2643 'params': {
2644 'skip_download': True,
2645 },
2646 }, {
2647 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2648 'only_matching': True,
2649 }, {
2650 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2651 'only_matching': True,
2652 }, {
2653 'url': 'https://www.youtube.com/TheYoungTurks/live',
2654 'only_matching': True,
2655 }]
2656
2657 def _real_extract(self, url):
2658 mobj = re.match(self._VALID_URL, url)
2659 channel_id = mobj.group('id')
2660 base_url = mobj.group('base_url')
2661 webpage = self._download_webpage(url, channel_id, fatal=False)
2662 if webpage:
2663 page_type = self._og_search_property(
2664 'type', webpage, 'page type', default='')
2665 video_id = self._html_search_meta(
2666 'videoId', webpage, 'video id', default=None)
2667 if page_type.startswith('video') and video_id and re.match(
2668 r'^[0-9A-Za-z_-]{11}$', video_id):
2669 return self.url_result(video_id, YoutubeIE.ie_key())
2670 return self.url_result(base_url)
2671
2672
2673 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2674 IE_DESC = 'YouTube.com user/channel playlists'
2675 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2676 IE_NAME = 'youtube:playlists'
2677
2678 _TESTS = [{
2679 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2680 'playlist_mincount': 4,
2681 'info_dict': {
2682 'id': 'ThirstForScience',
2683 'title': 'Thirst for Science',
2684 },
2685 }, {
2686 # with "Load more" button
2687 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2688 'playlist_mincount': 70,
2689 'info_dict': {
2690 'id': 'igorkle1',
2691 'title': 'Игорь Клейнер',
2692 },
2693 }, {
2694 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2695 'playlist_mincount': 17,
2696 'info_dict': {
2697 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2698 'title': 'Chem Player',
2699 },
2700 }]
2701
2702
2703 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2704 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2705
2706
2707 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2708 IE_DESC = 'YouTube.com searches'
2709 # there doesn't appear to be a real limit, for example if you search for
2710 # 'python' you get more than 8.000.000 results
2711 _MAX_RESULTS = float('inf')
2712 IE_NAME = 'youtube:search'
2713 _SEARCH_KEY = 'ytsearch'
2714 _EXTRA_QUERY_ARGS = {}
2715 _TESTS = []
2716
2717 def _get_n_results(self, query, n):
2718 """Get a specified number of results for a query"""
2719
2720 videos = []
2721 limit = n
2722
2723 url_query = {
2724 'search_query': query.encode('utf-8'),
2725 }
2726 url_query.update(self._EXTRA_QUERY_ARGS)
2727 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2728
2729 for pagenum in itertools.count(1):
2730 data = self._download_json(
2731 result_url, video_id='query "%s"' % query,
2732 note='Downloading page %s' % pagenum,
2733 errnote='Unable to download API page',
2734 query={'spf': 'navigate'})
2735 html_content = data[1]['body']['content']
2736
2737 if 'class="search-message' in html_content:
2738 raise ExtractorError(
2739 '[youtube] No video results', expected=True)
2740
2741 new_videos = list(self._process_page(html_content))
2742 videos += new_videos
2743 if not new_videos or len(videos) > limit:
2744 break
2745 next_link = self._html_search_regex(
2746 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2747 html_content, 'next link', default=None)
2748 if next_link is None:
2749 break
2750 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2751
2752 if len(videos) > n:
2753 videos = videos[:n]
2754 return self.playlist_result(videos, query)
2755
2756
2757 class YoutubeSearchDateIE(YoutubeSearchIE):
2758 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2759 _SEARCH_KEY = 'ytsearchdate'
2760 IE_DESC = 'YouTube.com searches, newest videos first'
2761 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2762
2763
2764 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2765 IE_DESC = 'YouTube.com search URLs'
2766 IE_NAME = 'youtube:search_url'
2767 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2768 _TESTS = [{
2769 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2770 'playlist_mincount': 5,
2771 'info_dict': {
2772 'title': 'youtube-dl test video',
2773 }
2774 }, {
2775 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2776 'only_matching': True,
2777 }]
2778
2779 def _real_extract(self, url):
2780 mobj = re.match(self._VALID_URL, url)
2781 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2782 webpage = self._download_webpage(url, query)
2783 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2784
2785
2786 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2787 IE_DESC = 'YouTube.com (multi-season) shows'
2788 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2789 IE_NAME = 'youtube:show'
2790 _TESTS = [{
2791 'url': 'https://www.youtube.com/show/airdisasters',
2792 'playlist_mincount': 5,
2793 'info_dict': {
2794 'id': 'airdisasters',
2795 'title': 'Air Disasters',
2796 }
2797 }]
2798
2799 def _real_extract(self, url):
2800 playlist_id = self._match_id(url)
2801 return super(YoutubeShowIE, self)._real_extract(
2802 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2803
2804
2805 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2806 """
2807 Base class for feed extractors
2808 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2809 """
2810 _LOGIN_REQUIRED = True
2811
2812 @property
2813 def IE_NAME(self):
2814 return 'youtube:%s' % self._FEED_NAME
2815
2816 def _real_initialize(self):
2817 self._login()
2818
2819 def _entries(self, page):
2820 # The extraction process is the same as for playlists, but the regex
2821 # for the video ids doesn't contain an index
2822 ids = []
2823 more_widget_html = content_html = page
2824 for page_num in itertools.count(1):
2825 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2826
2827 # 'recommended' feed has infinite 'load more' and each new portion spins
2828 # the same videos in (sometimes) slightly different order, so we'll check
2829 # for unicity and break when portion has no new videos
2830 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2831 if not new_ids:
2832 break
2833
2834 ids.extend(new_ids)
2835
2836 for entry in self._ids_to_results(new_ids):
2837 yield entry
2838
2839 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2840 if not mobj:
2841 break
2842
2843 more = self._download_json(
2844 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2845 'Downloading page #%s' % page_num,
2846 transform_source=uppercase_escape)
2847 content_html = more['content_html']
2848 more_widget_html = more['load_more_widget_html']
2849
2850 def _real_extract(self, url):
2851 page = self._download_webpage(
2852 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2853 self._PLAYLIST_TITLE)
2854 return self.playlist_result(
2855 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2856
2857
2858 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2859 IE_NAME = 'youtube:watchlater'
2860 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2861 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2862
2863 _TESTS = [{
2864 'url': 'https://www.youtube.com/playlist?list=WL',
2865 'only_matching': True,
2866 }, {
2867 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2868 'only_matching': True,
2869 }]
2870
2871 def _real_extract(self, url):
2872 _, video = self._check_download_just_video(url, 'WL')
2873 if video:
2874 return video
2875 _, playlist = self._extract_playlist('WL')
2876 return playlist
2877
2878
2879 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2880 IE_NAME = 'youtube:favorites'
2881 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2882 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2883 _LOGIN_REQUIRED = True
2884
2885 def _real_extract(self, url):
2886 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2887 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2888 return self.url_result(playlist_id, 'YoutubePlaylist')
2889
2890
2891 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2892 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2893 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2894 _FEED_NAME = 'recommended'
2895 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2896
2897
2898 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2899 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2900 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2901 _FEED_NAME = 'subscriptions'
2902 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2903
2904
2905 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2906 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2907 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2908 _FEED_NAME = 'history'
2909 _PLAYLIST_TITLE = 'Youtube History'
2910
2911
2912 class YoutubeTruncatedURLIE(InfoExtractor):
2913 IE_NAME = 'youtube:truncated_url'
2914 IE_DESC = False # Do not list
2915 _VALID_URL = r'''(?x)
2916 (?:https?://)?
2917 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2918 (?:watch\?(?:
2919 feature=[a-z_]+|
2920 annotation_id=annotation_[^&]+|
2921 x-yt-cl=[0-9]+|
2922 hl=[^&]*|
2923 t=[0-9]+
2924 )?
2925 |
2926 attribution_link\?a=[^&]+
2927 )
2928 $
2929 '''
2930
2931 _TESTS = [{
2932 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2933 'only_matching': True,
2934 }, {
2935 'url': 'https://www.youtube.com/watch?',
2936 'only_matching': True,
2937 }, {
2938 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2939 'only_matching': True,
2940 }, {
2941 'url': 'https://www.youtube.com/watch?feature=foo',
2942 'only_matching': True,
2943 }, {
2944 'url': 'https://www.youtube.com/watch?hl=en-GB',
2945 'only_matching': True,
2946 }, {
2947 'url': 'https://www.youtube.com/watch?t=2372',
2948 'only_matching': True,
2949 }]
2950
2951 def _real_extract(self, url):
2952 raise ExtractorError(
2953 'Did you forget to quote the URL? Remember that & is a meta '
2954 'character in most shells, so you want to put the URL in quotes, '
2955 'like youtube-dl '
2956 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2957 ' or simply youtube-dl BaW_jenozKc .',
2958 expected=True)
2959
2960
2961 class YoutubeTruncatedIDIE(InfoExtractor):
2962 IE_NAME = 'youtube:truncated_id'
2963 IE_DESC = False # Do not list
2964 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2965
2966 _TESTS = [{
2967 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2968 'only_matching': True,
2969 }]
2970
2971 def _real_extract(self, url):
2972 video_id = self._match_id(url)
2973 raise ExtractorError(
2974 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2975 expected=True)