]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[test/helper] Add support for maxcount and count collection len test checkers
[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 # DRM protected
1082 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1083 'only_matching': True,
1084 }
1085 ]
1086
1087 def __init__(self, *args, **kwargs):
1088 super(YoutubeIE, self).__init__(*args, **kwargs)
1089 self._player_cache = {}
1090
1091 def report_video_info_webpage_download(self, video_id):
1092 """Report attempt to download video info webpage."""
1093 self.to_screen('%s: Downloading video info webpage' % video_id)
1094
1095 def report_information_extraction(self, video_id):
1096 """Report attempt to extract video information."""
1097 self.to_screen('%s: Extracting video information' % video_id)
1098
1099 def report_unavailable_format(self, video_id, format):
1100 """Report extracted video URL."""
1101 self.to_screen('%s: Format %s not available' % (video_id, format))
1102
1103 def report_rtmp_download(self):
1104 """Indicate the download will use the RTMP protocol."""
1105 self.to_screen('RTMP download detected')
1106
1107 def _signature_cache_id(self, example_sig):
1108 """ Return a string representation of a signature """
1109 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1110
1111 def _extract_signature_function(self, video_id, player_url, example_sig):
1112 id_m = re.match(
1113 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1114 player_url)
1115 if not id_m:
1116 raise ExtractorError('Cannot identify player %r' % player_url)
1117 player_type = id_m.group('ext')
1118 player_id = id_m.group('id')
1119
1120 # Read from filesystem cache
1121 func_id = '%s_%s_%s' % (
1122 player_type, player_id, self._signature_cache_id(example_sig))
1123 assert os.path.basename(func_id) == func_id
1124
1125 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1126 if cache_spec is not None:
1127 return lambda s: ''.join(s[i] for i in cache_spec)
1128
1129 download_note = (
1130 'Downloading player %s' % player_url
1131 if self._downloader.params.get('verbose') else
1132 'Downloading %s player %s' % (player_type, player_id)
1133 )
1134 if player_type == 'js':
1135 code = self._download_webpage(
1136 player_url, video_id,
1137 note=download_note,
1138 errnote='Download of %s failed' % player_url)
1139 res = self._parse_sig_js(code)
1140 elif player_type == 'swf':
1141 urlh = self._request_webpage(
1142 player_url, video_id,
1143 note=download_note,
1144 errnote='Download of %s failed' % player_url)
1145 code = urlh.read()
1146 res = self._parse_sig_swf(code)
1147 else:
1148 assert False, 'Invalid player type %r' % player_type
1149
1150 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1151 cache_res = res(test_string)
1152 cache_spec = [ord(c) for c in cache_res]
1153
1154 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1155 return res
1156
1157 def _print_sig_code(self, func, example_sig):
1158 def gen_sig_code(idxs):
1159 def _genslice(start, end, step):
1160 starts = '' if start == 0 else str(start)
1161 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1162 steps = '' if step == 1 else (':%d' % step)
1163 return 's[%s%s%s]' % (starts, ends, steps)
1164
1165 step = None
1166 # Quelch pyflakes warnings - start will be set when step is set
1167 start = '(Never used)'
1168 for i, prev in zip(idxs[1:], idxs[:-1]):
1169 if step is not None:
1170 if i - prev == step:
1171 continue
1172 yield _genslice(start, prev, step)
1173 step = None
1174 continue
1175 if i - prev in [-1, 1]:
1176 step = i - prev
1177 start = prev
1178 continue
1179 else:
1180 yield 's[%d]' % prev
1181 if step is None:
1182 yield 's[%d]' % i
1183 else:
1184 yield _genslice(start, i, step)
1185
1186 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1187 cache_res = func(test_string)
1188 cache_spec = [ord(c) for c in cache_res]
1189 expr_code = ' + '.join(gen_sig_code(cache_spec))
1190 signature_id_tuple = '(%s)' % (
1191 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1192 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1193 ' return %s\n') % (signature_id_tuple, expr_code)
1194 self.to_screen('Extracted signature function:\n' + code)
1195
1196 def _parse_sig_js(self, jscode):
1197 funcname = self._search_regex(
1198 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1199 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1200 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1201 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1202 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1203 jscode, 'Initial JS player signature function name', group='sig')
1204
1205 jsi = JSInterpreter(jscode)
1206 initial_function = jsi.extract_function(funcname)
1207 return lambda s: initial_function([s])
1208
1209 def _parse_sig_swf(self, file_contents):
1210 swfi = SWFInterpreter(file_contents)
1211 TARGET_CLASSNAME = 'SignatureDecipher'
1212 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1213 initial_function = swfi.extract_function(searched_class, 'decipher')
1214 return lambda s: initial_function([s])
1215
1216 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1217 """Turn the encrypted s field into a working signature"""
1218
1219 if player_url is None:
1220 raise ExtractorError('Cannot decrypt signature without player_url')
1221
1222 if player_url.startswith('//'):
1223 player_url = 'https:' + player_url
1224 elif not re.match(r'https?://', player_url):
1225 player_url = compat_urlparse.urljoin(
1226 'https://www.youtube.com', player_url)
1227 try:
1228 player_id = (player_url, self._signature_cache_id(s))
1229 if player_id not in self._player_cache:
1230 func = self._extract_signature_function(
1231 video_id, player_url, s
1232 )
1233 self._player_cache[player_id] = func
1234 func = self._player_cache[player_id]
1235 if self._downloader.params.get('youtube_print_sig_code'):
1236 self._print_sig_code(func, s)
1237 return func(s)
1238 except Exception as e:
1239 tb = traceback.format_exc()
1240 raise ExtractorError(
1241 'Signature extraction failed: ' + tb, cause=e)
1242
1243 def _get_subtitles(self, video_id, webpage):
1244 try:
1245 subs_doc = self._download_xml(
1246 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1247 video_id, note=False)
1248 except ExtractorError as err:
1249 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1250 return {}
1251
1252 sub_lang_list = {}
1253 for track in subs_doc.findall('track'):
1254 lang = track.attrib['lang_code']
1255 if lang in sub_lang_list:
1256 continue
1257 sub_formats = []
1258 for ext in self._SUBTITLE_FORMATS:
1259 params = compat_urllib_parse_urlencode({
1260 'lang': lang,
1261 'v': video_id,
1262 'fmt': ext,
1263 'name': track.attrib['name'].encode('utf-8'),
1264 })
1265 sub_formats.append({
1266 'url': 'https://www.youtube.com/api/timedtext?' + params,
1267 'ext': ext,
1268 })
1269 sub_lang_list[lang] = sub_formats
1270 if not sub_lang_list:
1271 self._downloader.report_warning('video doesn\'t have subtitles')
1272 return {}
1273 return sub_lang_list
1274
1275 def _get_ytplayer_config(self, video_id, webpage):
1276 patterns = (
1277 # User data may contain arbitrary character sequences that may affect
1278 # JSON extraction with regex, e.g. when '};' is contained the second
1279 # regex won't capture the whole JSON. Yet working around by trying more
1280 # concrete regex first keeping in mind proper quoted string handling
1281 # to be implemented in future that will replace this workaround (see
1282 # https://github.com/rg3/youtube-dl/issues/7468,
1283 # https://github.com/rg3/youtube-dl/pull/7599)
1284 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1285 r';ytplayer\.config\s*=\s*({.+?});',
1286 )
1287 config = self._search_regex(
1288 patterns, webpage, 'ytplayer.config', default=None)
1289 if config:
1290 return self._parse_json(
1291 uppercase_escape(config), video_id, fatal=False)
1292
1293 def _get_automatic_captions(self, video_id, webpage):
1294 """We need the webpage for getting the captions url, pass it as an
1295 argument to speed up the process."""
1296 self.to_screen('%s: Looking for automatic captions' % video_id)
1297 player_config = self._get_ytplayer_config(video_id, webpage)
1298 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1299 if not player_config:
1300 self._downloader.report_warning(err_msg)
1301 return {}
1302 try:
1303 args = player_config['args']
1304 caption_url = args.get('ttsurl')
1305 if caption_url:
1306 timestamp = args['timestamp']
1307 # We get the available subtitles
1308 list_params = compat_urllib_parse_urlencode({
1309 'type': 'list',
1310 'tlangs': 1,
1311 'asrs': 1,
1312 })
1313 list_url = caption_url + '&' + list_params
1314 caption_list = self._download_xml(list_url, video_id)
1315 original_lang_node = caption_list.find('track')
1316 if original_lang_node is None:
1317 self._downloader.report_warning('Video doesn\'t have automatic captions')
1318 return {}
1319 original_lang = original_lang_node.attrib['lang_code']
1320 caption_kind = original_lang_node.attrib.get('kind', '')
1321
1322 sub_lang_list = {}
1323 for lang_node in caption_list.findall('target'):
1324 sub_lang = lang_node.attrib['lang_code']
1325 sub_formats = []
1326 for ext in self._SUBTITLE_FORMATS:
1327 params = compat_urllib_parse_urlencode({
1328 'lang': original_lang,
1329 'tlang': sub_lang,
1330 'fmt': ext,
1331 'ts': timestamp,
1332 'kind': caption_kind,
1333 })
1334 sub_formats.append({
1335 'url': caption_url + '&' + params,
1336 'ext': ext,
1337 })
1338 sub_lang_list[sub_lang] = sub_formats
1339 return sub_lang_list
1340
1341 def make_captions(sub_url, sub_langs):
1342 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1343 caption_qs = compat_parse_qs(parsed_sub_url.query)
1344 captions = {}
1345 for sub_lang in sub_langs:
1346 sub_formats = []
1347 for ext in self._SUBTITLE_FORMATS:
1348 caption_qs.update({
1349 'tlang': [sub_lang],
1350 'fmt': [ext],
1351 })
1352 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1353 query=compat_urllib_parse_urlencode(caption_qs, True)))
1354 sub_formats.append({
1355 'url': sub_url,
1356 'ext': ext,
1357 })
1358 captions[sub_lang] = sub_formats
1359 return captions
1360
1361 # New captions format as of 22.06.2017
1362 player_response = args.get('player_response')
1363 if player_response and isinstance(player_response, compat_str):
1364 player_response = self._parse_json(
1365 player_response, video_id, fatal=False)
1366 if player_response:
1367 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1368 base_url = renderer['captionTracks'][0]['baseUrl']
1369 sub_lang_list = []
1370 for lang in renderer['translationLanguages']:
1371 lang_code = lang.get('languageCode')
1372 if lang_code:
1373 sub_lang_list.append(lang_code)
1374 return make_captions(base_url, sub_lang_list)
1375
1376 # Some videos don't provide ttsurl but rather caption_tracks and
1377 # caption_translation_languages (e.g. 20LmZk1hakA)
1378 # Does not used anymore as of 22.06.2017
1379 caption_tracks = args['caption_tracks']
1380 caption_translation_languages = args['caption_translation_languages']
1381 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1382 sub_lang_list = []
1383 for lang in caption_translation_languages.split(','):
1384 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1385 sub_lang = lang_qs.get('lc', [None])[0]
1386 if sub_lang:
1387 sub_lang_list.append(sub_lang)
1388 return make_captions(caption_url, sub_lang_list)
1389 # An extractor error can be raise by the download process if there are
1390 # no automatic captions but there are subtitles
1391 except (KeyError, IndexError, ExtractorError):
1392 self._downloader.report_warning(err_msg)
1393 return {}
1394
1395 def _mark_watched(self, video_id, video_info, player_response):
1396 playback_url = url_or_none(try_get(
1397 player_response,
1398 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1399 video_info, lambda x: x['videostats_playback_base_url'][0]))
1400 if not playback_url:
1401 return
1402 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1403 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1404
1405 # cpn generation algorithm is reverse engineered from base.js.
1406 # In fact it works even with dummy cpn.
1407 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1408 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1409
1410 qs.update({
1411 'ver': ['2'],
1412 'cpn': [cpn],
1413 })
1414 playback_url = compat_urlparse.urlunparse(
1415 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1416
1417 self._download_webpage(
1418 playback_url, video_id, 'Marking watched',
1419 'Unable to mark watched', fatal=False)
1420
1421 @staticmethod
1422 def _extract_urls(webpage):
1423 # Embedded YouTube player
1424 entries = [
1425 unescapeHTML(mobj.group('url'))
1426 for mobj in re.finditer(r'''(?x)
1427 (?:
1428 <iframe[^>]+?src=|
1429 data-video-url=|
1430 <embed[^>]+?src=|
1431 embedSWF\(?:\s*|
1432 <object[^>]+data=|
1433 new\s+SWFObject\(
1434 )
1435 (["\'])
1436 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1437 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1438 \1''', webpage)]
1439
1440 # lazyYT YouTube embed
1441 entries.extend(list(map(
1442 unescapeHTML,
1443 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1444
1445 # Wordpress "YouTube Video Importer" plugin
1446 matches = re.findall(r'''(?x)<div[^>]+
1447 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1448 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1449 entries.extend(m[-1] for m in matches)
1450
1451 return entries
1452
1453 @staticmethod
1454 def _extract_url(webpage):
1455 urls = YoutubeIE._extract_urls(webpage)
1456 return urls[0] if urls else None
1457
1458 @classmethod
1459 def extract_id(cls, url):
1460 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1461 if mobj is None:
1462 raise ExtractorError('Invalid URL: %s' % url)
1463 video_id = mobj.group(2)
1464 return video_id
1465
1466 def _extract_annotations(self, video_id):
1467 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1468 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1469
1470 @staticmethod
1471 def _extract_chapters(description, duration):
1472 if not description:
1473 return None
1474 chapter_lines = re.findall(
1475 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1476 description)
1477 if not chapter_lines:
1478 return None
1479 chapters = []
1480 for next_num, (chapter_line, time_point) in enumerate(
1481 chapter_lines, start=1):
1482 start_time = parse_duration(time_point)
1483 if start_time is None:
1484 continue
1485 if start_time > duration:
1486 break
1487 end_time = (duration if next_num == len(chapter_lines)
1488 else parse_duration(chapter_lines[next_num][1]))
1489 if end_time is None:
1490 continue
1491 if end_time > duration:
1492 end_time = duration
1493 if start_time > end_time:
1494 break
1495 chapter_title = re.sub(
1496 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1497 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1498 chapters.append({
1499 'start_time': start_time,
1500 'end_time': end_time,
1501 'title': chapter_title,
1502 })
1503 return chapters
1504
1505 def _real_extract(self, url):
1506 url, smuggled_data = unsmuggle_url(url, {})
1507
1508 proto = (
1509 'http' if self._downloader.params.get('prefer_insecure', False)
1510 else 'https')
1511
1512 start_time = None
1513 end_time = None
1514 parsed_url = compat_urllib_parse_urlparse(url)
1515 for component in [parsed_url.fragment, parsed_url.query]:
1516 query = compat_parse_qs(component)
1517 if start_time is None and 't' in query:
1518 start_time = parse_duration(query['t'][0])
1519 if start_time is None and 'start' in query:
1520 start_time = parse_duration(query['start'][0])
1521 if end_time is None and 'end' in query:
1522 end_time = parse_duration(query['end'][0])
1523
1524 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1525 mobj = re.search(self._NEXT_URL_RE, url)
1526 if mobj:
1527 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1528 video_id = self.extract_id(url)
1529
1530 # Get video webpage
1531 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1532 video_webpage = self._download_webpage(url, video_id)
1533
1534 # Attempt to extract SWF player URL
1535 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1536 if mobj is not None:
1537 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1538 else:
1539 player_url = None
1540
1541 dash_mpds = []
1542
1543 def add_dash_mpd(video_info):
1544 dash_mpd = video_info.get('dashmpd')
1545 if dash_mpd and dash_mpd[0] not in dash_mpds:
1546 dash_mpds.append(dash_mpd[0])
1547
1548 def add_dash_mpd_pr(pl_response):
1549 dash_mpd = url_or_none(try_get(
1550 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1551 compat_str))
1552 if dash_mpd and dash_mpd not in dash_mpds:
1553 dash_mpds.append(dash_mpd)
1554
1555 is_live = None
1556 view_count = None
1557
1558 def extract_view_count(v_info):
1559 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1560
1561 player_response = {}
1562
1563 # Get video info
1564 embed_webpage = None
1565 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1566 age_gate = True
1567 # We simulate the access to the video from www.youtube.com/v/{video_id}
1568 # this can be viewed without login into Youtube
1569 url = proto + '://www.youtube.com/embed/%s' % video_id
1570 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1571 data = compat_urllib_parse_urlencode({
1572 'video_id': video_id,
1573 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1574 'sts': self._search_regex(
1575 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1576 })
1577 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1578 video_info_webpage = self._download_webpage(
1579 video_info_url, video_id,
1580 note='Refetching age-gated info webpage',
1581 errnote='unable to download video info webpage')
1582 video_info = compat_parse_qs(video_info_webpage)
1583 add_dash_mpd(video_info)
1584 else:
1585 age_gate = False
1586 video_info = None
1587 sts = None
1588 # Try looking directly into the video webpage
1589 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1590 if ytplayer_config:
1591 args = ytplayer_config['args']
1592 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1593 # Convert to the same format returned by compat_parse_qs
1594 video_info = dict((k, [v]) for k, v in args.items())
1595 add_dash_mpd(video_info)
1596 # Rental video is not rented but preview is available (e.g.
1597 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1598 # https://github.com/rg3/youtube-dl/issues/10532)
1599 if not video_info and args.get('ypc_vid'):
1600 return self.url_result(
1601 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1602 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1603 is_live = True
1604 sts = ytplayer_config.get('sts')
1605 if not player_response:
1606 pl_response = str_or_none(args.get('player_response'))
1607 if pl_response:
1608 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1609 if isinstance(pl_response, dict):
1610 player_response = pl_response
1611 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1612 add_dash_mpd_pr(player_response)
1613 # We also try looking in get_video_info since it may contain different dashmpd
1614 # URL that points to a DASH manifest with possibly different itag set (some itags
1615 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1616 # manifest pointed by get_video_info's dashmpd).
1617 # The general idea is to take a union of itags of both DASH manifests (for example
1618 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1619 self.report_video_info_webpage_download(video_id)
1620 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1621 query = {
1622 'video_id': video_id,
1623 'ps': 'default',
1624 'eurl': '',
1625 'gl': 'US',
1626 'hl': 'en',
1627 }
1628 if el:
1629 query['el'] = el
1630 if sts:
1631 query['sts'] = sts
1632 video_info_webpage = self._download_webpage(
1633 '%s://www.youtube.com/get_video_info' % proto,
1634 video_id, note=False,
1635 errnote='unable to download video info webpage',
1636 fatal=False, query=query)
1637 if not video_info_webpage:
1638 continue
1639 get_video_info = compat_parse_qs(video_info_webpage)
1640 if not player_response:
1641 pl_response = get_video_info.get('player_response', [None])[0]
1642 if isinstance(pl_response, dict):
1643 player_response = pl_response
1644 add_dash_mpd_pr(player_response)
1645 add_dash_mpd(get_video_info)
1646 if view_count is None:
1647 view_count = extract_view_count(get_video_info)
1648 if not video_info:
1649 video_info = get_video_info
1650 if 'token' in get_video_info:
1651 # Different get_video_info requests may report different results, e.g.
1652 # some may report video unavailability, but some may serve it without
1653 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1654 # the original webpage as well as el=info and el=embedded get_video_info
1655 # requests report video unavailability due to geo restriction while
1656 # el=detailpage succeeds and returns valid data). This is probably
1657 # due to YouTube measures against IP ranges of hosting providers.
1658 # Working around by preferring the first succeeded video_info containing
1659 # the token if no such video_info yet was found.
1660 if 'token' not in video_info:
1661 video_info = get_video_info
1662 break
1663
1664 def extract_unavailable_message():
1665 return self._html_search_regex(
1666 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1667 video_webpage, 'unavailable message', default=None)
1668
1669 if 'token' not in video_info:
1670 if 'reason' in video_info:
1671 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1672 regions_allowed = self._html_search_meta(
1673 'regionsAllowed', video_webpage, default=None)
1674 countries = regions_allowed.split(',') if regions_allowed else None
1675 self.raise_geo_restricted(
1676 msg=video_info['reason'][0], countries=countries)
1677 reason = video_info['reason'][0]
1678 if 'Invalid parameters' in reason:
1679 unavailable_message = extract_unavailable_message()
1680 if unavailable_message:
1681 reason = unavailable_message
1682 raise ExtractorError(
1683 'YouTube said: %s' % reason,
1684 expected=True, video_id=video_id)
1685 else:
1686 raise ExtractorError(
1687 '"token" parameter not in video info for unknown reason',
1688 video_id=video_id)
1689
1690 if video_info.get('license_info'):
1691 raise ExtractorError('This video is DRM protected.', expected=True)
1692
1693 video_details = try_get(
1694 player_response, lambda x: x['videoDetails'], dict) or {}
1695
1696 # title
1697 if 'title' in video_info:
1698 video_title = video_info['title'][0]
1699 elif 'title' in player_response:
1700 video_title = video_details['title']
1701 else:
1702 self._downloader.report_warning('Unable to extract video title')
1703 video_title = '_'
1704
1705 # description
1706 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1707 if video_description:
1708
1709 def replace_url(m):
1710 redir_url = compat_urlparse.urljoin(url, m.group(1))
1711 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1712 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1713 qs = compat_parse_qs(parsed_redir_url.query)
1714 q = qs.get('q')
1715 if q and q[0]:
1716 return q[0]
1717 return redir_url
1718
1719 description_original = video_description = re.sub(r'''(?x)
1720 <a\s+
1721 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1722 (?:title|href)="([^"]+)"\s+
1723 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1724 class="[^"]*"[^>]*>
1725 [^<]+\.{3}\s*
1726 </a>
1727 ''', replace_url, video_description)
1728 video_description = clean_html(video_description)
1729 else:
1730 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1731 if fd_mobj:
1732 video_description = unescapeHTML(fd_mobj.group(1))
1733 else:
1734 video_description = ''
1735
1736 if not smuggled_data.get('force_singlefeed', False):
1737 if not self._downloader.params.get('noplaylist'):
1738 multifeed_metadata_list = try_get(
1739 player_response,
1740 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1741 compat_str) or try_get(
1742 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1743 if multifeed_metadata_list:
1744 entries = []
1745 feed_ids = []
1746 for feed in multifeed_metadata_list.split(','):
1747 # Unquote should take place before split on comma (,) since textual
1748 # fields may contain comma as well (see
1749 # https://github.com/rg3/youtube-dl/issues/8536)
1750 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1751 entries.append({
1752 '_type': 'url_transparent',
1753 'ie_key': 'Youtube',
1754 'url': smuggle_url(
1755 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1756 {'force_singlefeed': True}),
1757 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1758 })
1759 feed_ids.append(feed_data['id'][0])
1760 self.to_screen(
1761 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1762 % (', '.join(feed_ids), video_id))
1763 return self.playlist_result(entries, video_id, video_title, video_description)
1764 else:
1765 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1766
1767 if view_count is None:
1768 view_count = extract_view_count(video_info)
1769 if view_count is None and video_details:
1770 view_count = int_or_none(video_details.get('viewCount'))
1771
1772 # Check for "rental" videos
1773 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1774 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1775
1776 def _extract_filesize(media_url):
1777 return int_or_none(self._search_regex(
1778 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1779
1780 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1781 self.report_rtmp_download()
1782 formats = [{
1783 'format_id': '_rtmp',
1784 'protocol': 'rtmp',
1785 'url': video_info['conn'][0],
1786 'player_url': player_url,
1787 }]
1788 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):
1789 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1790 if 'rtmpe%3Dyes' in encoded_url_map:
1791 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1792 formats_spec = {}
1793 fmt_list = video_info.get('fmt_list', [''])[0]
1794 if fmt_list:
1795 for fmt in fmt_list.split(','):
1796 spec = fmt.split('/')
1797 if len(spec) > 1:
1798 width_height = spec[1].split('x')
1799 if len(width_height) == 2:
1800 formats_spec[spec[0]] = {
1801 'resolution': spec[1],
1802 'width': int_or_none(width_height[0]),
1803 'height': int_or_none(width_height[1]),
1804 }
1805 q = qualities(['small', 'medium', 'hd720'])
1806 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1807 if streaming_formats:
1808 for fmt in streaming_formats:
1809 itag = str_or_none(fmt.get('itag'))
1810 if not itag:
1811 continue
1812 quality = fmt.get('quality')
1813 quality_label = fmt.get('qualityLabel') or quality
1814 formats_spec[itag] = {
1815 'asr': int_or_none(fmt.get('audioSampleRate')),
1816 'filesize': int_or_none(fmt.get('contentLength')),
1817 'format_note': quality_label,
1818 'fps': int_or_none(fmt.get('fps')),
1819 'height': int_or_none(fmt.get('height')),
1820 'quality': q(quality),
1821 # bitrate for itag 43 is always 2147483647
1822 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1823 'width': int_or_none(fmt.get('width')),
1824 }
1825 formats = []
1826 for url_data_str in encoded_url_map.split(','):
1827 url_data = compat_parse_qs(url_data_str)
1828 if 'itag' not in url_data or 'url' not in url_data:
1829 continue
1830 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1831 # Unsupported FORMAT_STREAM_TYPE_OTF
1832 if stream_type == 3:
1833 continue
1834 format_id = url_data['itag'][0]
1835 url = url_data['url'][0]
1836
1837 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1838 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1839 jsplayer_url_json = self._search_regex(
1840 ASSETS_RE,
1841 embed_webpage if age_gate else video_webpage,
1842 'JS player URL (1)', default=None)
1843 if not jsplayer_url_json and not age_gate:
1844 # We need the embed website after all
1845 if embed_webpage is None:
1846 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1847 embed_webpage = self._download_webpage(
1848 embed_url, video_id, 'Downloading embed webpage')
1849 jsplayer_url_json = self._search_regex(
1850 ASSETS_RE, embed_webpage, 'JS player URL')
1851
1852 player_url = json.loads(jsplayer_url_json)
1853 if player_url is None:
1854 player_url_json = self._search_regex(
1855 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1856 video_webpage, 'age gate player URL')
1857 player_url = json.loads(player_url_json)
1858
1859 if 'sig' in url_data:
1860 url += '&signature=' + url_data['sig'][0]
1861 elif 's' in url_data:
1862 encrypted_sig = url_data['s'][0]
1863
1864 if self._downloader.params.get('verbose'):
1865 if player_url is None:
1866 player_version = 'unknown'
1867 player_desc = 'unknown'
1868 else:
1869 if player_url.endswith('swf'):
1870 player_version = self._search_regex(
1871 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1872 'flash player', fatal=False)
1873 player_desc = 'flash player %s' % player_version
1874 else:
1875 player_version = self._search_regex(
1876 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1877 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1878 player_url,
1879 'html5 player', fatal=False)
1880 player_desc = 'html5 player %s' % player_version
1881
1882 parts_sizes = self._signature_cache_id(encrypted_sig)
1883 self.to_screen('{%s} signature length %s, %s' %
1884 (format_id, parts_sizes, player_desc))
1885
1886 signature = self._decrypt_signature(
1887 encrypted_sig, video_id, player_url, age_gate)
1888 url += '&signature=' + signature
1889 if 'ratebypass' not in url:
1890 url += '&ratebypass=yes'
1891
1892 dct = {
1893 'format_id': format_id,
1894 'url': url,
1895 'player_url': player_url,
1896 }
1897 if format_id in self._formats:
1898 dct.update(self._formats[format_id])
1899 if format_id in formats_spec:
1900 dct.update(formats_spec[format_id])
1901
1902 # Some itags are not included in DASH manifest thus corresponding formats will
1903 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1904 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1905 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1906 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1907
1908 filesize = int_or_none(url_data.get(
1909 'clen', [None])[0]) or _extract_filesize(url)
1910
1911 quality = url_data.get('quality', [None])[0]
1912
1913 more_fields = {
1914 'filesize': filesize,
1915 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1916 'width': width,
1917 'height': height,
1918 'fps': int_or_none(url_data.get('fps', [None])[0]),
1919 'format_note': url_data.get('quality_label', [None])[0] or quality,
1920 'quality': q(quality),
1921 }
1922 for key, value in more_fields.items():
1923 if value:
1924 dct[key] = value
1925 type_ = url_data.get('type', [None])[0]
1926 if type_:
1927 type_split = type_.split(';')
1928 kind_ext = type_split[0].split('/')
1929 if len(kind_ext) == 2:
1930 kind, _ = kind_ext
1931 dct['ext'] = mimetype2ext(type_split[0])
1932 if kind in ('audio', 'video'):
1933 codecs = None
1934 for mobj in re.finditer(
1935 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1936 if mobj.group('key') == 'codecs':
1937 codecs = mobj.group('val')
1938 break
1939 if codecs:
1940 dct.update(parse_codecs(codecs))
1941 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1942 dct['downloader_options'] = {
1943 # Youtube throttles chunks >~10M
1944 'http_chunk_size': 10485760,
1945 }
1946 formats.append(dct)
1947 else:
1948 manifest_url = (
1949 url_or_none(try_get(
1950 player_response,
1951 lambda x: x['streamingData']['hlsManifestUrl'],
1952 compat_str)) or
1953 url_or_none(try_get(
1954 video_info, lambda x: x['hlsvp'][0], compat_str)))
1955 if manifest_url:
1956 formats = []
1957 m3u8_formats = self._extract_m3u8_formats(
1958 manifest_url, video_id, 'mp4', fatal=False)
1959 for a_format in m3u8_formats:
1960 itag = self._search_regex(
1961 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1962 if itag:
1963 a_format['format_id'] = itag
1964 if itag in self._formats:
1965 dct = self._formats[itag].copy()
1966 dct.update(a_format)
1967 a_format = dct
1968 a_format['player_url'] = player_url
1969 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1970 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1971 formats.append(a_format)
1972 else:
1973 error_message = clean_html(video_info.get('reason', [None])[0])
1974 if not error_message:
1975 error_message = extract_unavailable_message()
1976 if error_message:
1977 raise ExtractorError(error_message, expected=True)
1978 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
1979
1980 # uploader
1981 video_uploader = try_get(
1982 video_info, lambda x: x['author'][0],
1983 compat_str) or str_or_none(video_details.get('author'))
1984 if video_uploader:
1985 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1986 else:
1987 self._downloader.report_warning('unable to extract uploader name')
1988
1989 # uploader_id
1990 video_uploader_id = None
1991 video_uploader_url = None
1992 mobj = re.search(
1993 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1994 video_webpage)
1995 if mobj is not None:
1996 video_uploader_id = mobj.group('uploader_id')
1997 video_uploader_url = mobj.group('uploader_url')
1998 else:
1999 self._downloader.report_warning('unable to extract uploader nickname')
2000
2001 channel_id = self._html_search_meta(
2002 'channelId', video_webpage, 'channel id')
2003 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2004
2005 # thumbnail image
2006 # We try first to get a high quality image:
2007 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2008 video_webpage, re.DOTALL)
2009 if m_thumb is not None:
2010 video_thumbnail = m_thumb.group(1)
2011 elif 'thumbnail_url' not in video_info:
2012 self._downloader.report_warning('unable to extract video thumbnail')
2013 video_thumbnail = None
2014 else: # don't panic if we can't find it
2015 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2016
2017 # upload date
2018 upload_date = self._html_search_meta(
2019 'datePublished', video_webpage, 'upload date', default=None)
2020 if not upload_date:
2021 upload_date = self._search_regex(
2022 [r'(?s)id="eow-date.*?>(.*?)</span>',
2023 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2024 video_webpage, 'upload date', default=None)
2025 upload_date = unified_strdate(upload_date)
2026
2027 video_license = self._html_search_regex(
2028 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2029 video_webpage, 'license', default=None)
2030
2031 m_music = re.search(
2032 r'''(?x)
2033 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2034 <ul[^>]*>\s*
2035 <li>(?P<title>.+?)
2036 by (?P<creator>.+?)
2037 (?:
2038 \(.+?\)|
2039 <a[^>]*
2040 (?:
2041 \bhref=["\']/red[^>]*>| # drop possible
2042 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2043 )
2044 .*?
2045 )?</li
2046 ''',
2047 video_webpage)
2048 if m_music:
2049 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2050 video_creator = clean_html(m_music.group('creator'))
2051 else:
2052 video_alt_title = video_creator = None
2053
2054 def extract_meta(field):
2055 return self._html_search_regex(
2056 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2057 video_webpage, field, default=None)
2058
2059 track = extract_meta('Song')
2060 artist = extract_meta('Artist')
2061
2062 m_episode = re.search(
2063 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2064 video_webpage)
2065 if m_episode:
2066 series = unescapeHTML(m_episode.group('series'))
2067 season_number = int(m_episode.group('season'))
2068 episode_number = int(m_episode.group('episode'))
2069 else:
2070 series = season_number = episode_number = None
2071
2072 m_cat_container = self._search_regex(
2073 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2074 video_webpage, 'categories', default=None)
2075 if m_cat_container:
2076 category = self._html_search_regex(
2077 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2078 default=None)
2079 video_categories = None if category is None else [category]
2080 else:
2081 video_categories = None
2082
2083 video_tags = [
2084 unescapeHTML(m.group('content'))
2085 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2086
2087 def _extract_count(count_name):
2088 return str_to_int(self._search_regex(
2089 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2090 % re.escape(count_name),
2091 video_webpage, count_name, default=None))
2092
2093 like_count = _extract_count('like')
2094 dislike_count = _extract_count('dislike')
2095
2096 if view_count is None:
2097 view_count = str_to_int(self._search_regex(
2098 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2099 'view count', default=None))
2100
2101 # subtitles
2102 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2103 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2104
2105 video_duration = try_get(
2106 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2107 if not video_duration:
2108 video_duration = int_or_none(video_details.get('lengthSeconds'))
2109 if not video_duration:
2110 video_duration = parse_duration(self._html_search_meta(
2111 'duration', video_webpage, 'video duration'))
2112
2113 # annotations
2114 video_annotations = None
2115 if self._downloader.params.get('writeannotations', False):
2116 video_annotations = self._extract_annotations(video_id)
2117
2118 chapters = self._extract_chapters(description_original, video_duration)
2119
2120 # Look for the DASH manifest
2121 if self._downloader.params.get('youtube_include_dash_manifest', True):
2122 dash_mpd_fatal = True
2123 for mpd_url in dash_mpds:
2124 dash_formats = {}
2125 try:
2126 def decrypt_sig(mobj):
2127 s = mobj.group(1)
2128 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2129 return '/signature/%s' % dec_s
2130
2131 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2132
2133 for df in self._extract_mpd_formats(
2134 mpd_url, video_id, fatal=dash_mpd_fatal,
2135 formats_dict=self._formats):
2136 if not df.get('filesize'):
2137 df['filesize'] = _extract_filesize(df['url'])
2138 # Do not overwrite DASH format found in some previous DASH manifest
2139 if df['format_id'] not in dash_formats:
2140 dash_formats[df['format_id']] = df
2141 # Additional DASH manifests may end up in HTTP Error 403 therefore
2142 # allow them to fail without bug report message if we already have
2143 # some DASH manifest succeeded. This is temporary workaround to reduce
2144 # burst of bug reports until we figure out the reason and whether it
2145 # can be fixed at all.
2146 dash_mpd_fatal = False
2147 except (ExtractorError, KeyError) as e:
2148 self.report_warning(
2149 'Skipping DASH manifest: %r' % e, video_id)
2150 if dash_formats:
2151 # Remove the formats we found through non-DASH, they
2152 # contain less info and it can be wrong, because we use
2153 # fixed values (for example the resolution). See
2154 # https://github.com/rg3/youtube-dl/issues/5774 for an
2155 # example.
2156 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2157 formats.extend(dash_formats.values())
2158
2159 # Check for malformed aspect ratio
2160 stretched_m = re.search(
2161 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2162 video_webpage)
2163 if stretched_m:
2164 w = float(stretched_m.group('w'))
2165 h = float(stretched_m.group('h'))
2166 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2167 # We will only process correct ratios.
2168 if w > 0 and h > 0:
2169 ratio = w / h
2170 for f in formats:
2171 if f.get('vcodec') != 'none':
2172 f['stretched_ratio'] = ratio
2173
2174 self._sort_formats(formats)
2175
2176 self.mark_watched(video_id, video_info, player_response)
2177
2178 return {
2179 'id': video_id,
2180 'uploader': video_uploader,
2181 'uploader_id': video_uploader_id,
2182 'uploader_url': video_uploader_url,
2183 'channel_id': channel_id,
2184 'channel_url': channel_url,
2185 'upload_date': upload_date,
2186 'license': video_license,
2187 'creator': video_creator or artist,
2188 'title': video_title,
2189 'alt_title': video_alt_title or track,
2190 'thumbnail': video_thumbnail,
2191 'description': video_description,
2192 'categories': video_categories,
2193 'tags': video_tags,
2194 'subtitles': video_subtitles,
2195 'automatic_captions': automatic_captions,
2196 'duration': video_duration,
2197 'age_limit': 18 if age_gate else 0,
2198 'annotations': video_annotations,
2199 'chapters': chapters,
2200 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2201 'view_count': view_count,
2202 'like_count': like_count,
2203 'dislike_count': dislike_count,
2204 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2205 'formats': formats,
2206 'is_live': is_live,
2207 'start_time': start_time,
2208 'end_time': end_time,
2209 'series': series,
2210 'season_number': season_number,
2211 'episode_number': episode_number,
2212 'track': track,
2213 'artist': artist,
2214 }
2215
2216
2217 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2218 IE_DESC = 'YouTube.com playlists'
2219 _VALID_URL = r"""(?x)(?:
2220 (?:https?://)?
2221 (?:\w+\.)?
2222 (?:
2223 (?:
2224 youtube\.com|
2225 invidio\.us
2226 )
2227 /
2228 (?:
2229 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2230 \? (?:.*?[&;])*? (?:p|a|list)=
2231 | p/
2232 )|
2233 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2234 )
2235 (
2236 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2237 # Top tracks, they can also include dots
2238 |(?:MC)[\w\.]*
2239 )
2240 .*
2241 |
2242 (%(playlist_id)s)
2243 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2244 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2245 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2246 IE_NAME = 'youtube:playlist'
2247 _TESTS = [{
2248 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2249 'info_dict': {
2250 'title': 'ytdl test PL',
2251 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2252 },
2253 'playlist_count': 3,
2254 }, {
2255 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2256 'info_dict': {
2257 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2258 'title': 'YDL_Empty_List',
2259 },
2260 'playlist_count': 0,
2261 'skip': 'This playlist is private',
2262 }, {
2263 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2264 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2265 'info_dict': {
2266 'title': '29C3: Not my department',
2267 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2268 },
2269 'playlist_count': 95,
2270 }, {
2271 'note': 'issue #673',
2272 'url': 'PLBB231211A4F62143',
2273 'info_dict': {
2274 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2275 'id': 'PLBB231211A4F62143',
2276 },
2277 'playlist_mincount': 26,
2278 }, {
2279 'note': 'Large playlist',
2280 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2281 'info_dict': {
2282 'title': 'Uploads from Cauchemar',
2283 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2284 },
2285 'playlist_mincount': 799,
2286 }, {
2287 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2288 'info_dict': {
2289 'title': 'YDL_safe_search',
2290 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2291 },
2292 'playlist_count': 2,
2293 'skip': 'This playlist is private',
2294 }, {
2295 'note': 'embedded',
2296 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2297 'playlist_count': 4,
2298 'info_dict': {
2299 'title': 'JODA15',
2300 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2301 }
2302 }, {
2303 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2304 'playlist_mincount': 485,
2305 'info_dict': {
2306 'title': '2017 華語最新單曲 (2/24更新)',
2307 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2308 }
2309 }, {
2310 'note': 'Embedded SWF player',
2311 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2312 'playlist_count': 4,
2313 'info_dict': {
2314 'title': 'JODA7',
2315 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2316 }
2317 }, {
2318 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2319 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2320 'info_dict': {
2321 'title': 'Uploads from Interstellar Movie',
2322 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2323 },
2324 'playlist_mincount': 21,
2325 }, {
2326 # Playlist URL that does not actually serve a playlist
2327 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2328 'info_dict': {
2329 'id': 'FqZTN594JQw',
2330 'ext': 'webm',
2331 'title': "Smiley's People 01 detective, Adventure Series, Action",
2332 'uploader': 'STREEM',
2333 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2334 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2335 'upload_date': '20150526',
2336 'license': 'Standard YouTube License',
2337 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2338 'categories': ['People & Blogs'],
2339 'tags': list,
2340 'view_count': int,
2341 'like_count': int,
2342 'dislike_count': int,
2343 },
2344 'params': {
2345 'skip_download': True,
2346 },
2347 'add_ie': [YoutubeIE.ie_key()],
2348 }, {
2349 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2350 'info_dict': {
2351 'id': 'yeWKywCrFtk',
2352 'ext': 'mp4',
2353 'title': 'Small Scale Baler and Braiding Rugs',
2354 'uploader': 'Backus-Page House Museum',
2355 'uploader_id': 'backuspagemuseum',
2356 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2357 'upload_date': '20161008',
2358 'license': 'Standard YouTube License',
2359 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2360 'categories': ['Nonprofits & Activism'],
2361 'tags': list,
2362 'like_count': int,
2363 'dislike_count': int,
2364 },
2365 'params': {
2366 'noplaylist': True,
2367 'skip_download': True,
2368 },
2369 }, {
2370 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2371 'only_matching': True,
2372 }, {
2373 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2374 'only_matching': True,
2375 }, {
2376 # music album playlist
2377 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2378 'only_matching': True,
2379 }, {
2380 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2381 'only_matching': True,
2382 }]
2383
2384 def _real_initialize(self):
2385 self._login()
2386
2387 def _extract_mix(self, playlist_id):
2388 # The mixes are generated from a single video
2389 # the id of the playlist is just 'RD' + video_id
2390 ids = []
2391 last_id = playlist_id[-11:]
2392 for n in itertools.count(1):
2393 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2394 webpage = self._download_webpage(
2395 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2396 new_ids = orderedSet(re.findall(
2397 r'''(?xs)data-video-username=".*?".*?
2398 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2399 webpage))
2400 # Fetch new pages until all the videos are repeated, it seems that
2401 # there are always 51 unique videos.
2402 new_ids = [_id for _id in new_ids if _id not in ids]
2403 if not new_ids:
2404 break
2405 ids.extend(new_ids)
2406 last_id = ids[-1]
2407
2408 url_results = self._ids_to_results(ids)
2409
2410 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2411 title_span = (
2412 search_title('playlist-title') or
2413 search_title('title long-title') or
2414 search_title('title'))
2415 title = clean_html(title_span)
2416
2417 return self.playlist_result(url_results, playlist_id, title)
2418
2419 def _extract_playlist(self, playlist_id):
2420 url = self._TEMPLATE_URL % playlist_id
2421 page = self._download_webpage(url, playlist_id)
2422
2423 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2424 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2425 match = match.strip()
2426 # Check if the playlist exists or is private
2427 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2428 if mobj:
2429 reason = mobj.group('reason')
2430 message = 'This playlist %s' % reason
2431 if 'private' in reason:
2432 message += ', use --username or --netrc to access it'
2433 message += '.'
2434 raise ExtractorError(message, expected=True)
2435 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2436 raise ExtractorError(
2437 'Invalid parameters. Maybe URL is incorrect.',
2438 expected=True)
2439 elif re.match(r'[^<]*Choose your language[^<]*', match):
2440 continue
2441 else:
2442 self.report_warning('Youtube gives an alert message: ' + match)
2443
2444 playlist_title = self._html_search_regex(
2445 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2446 page, 'title', default=None)
2447
2448 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2449 uploader = self._search_regex(
2450 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2451 page, 'uploader', default=None)
2452 mobj = re.search(
2453 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2454 page)
2455 if mobj:
2456 uploader_id = mobj.group('uploader_id')
2457 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2458 else:
2459 uploader_id = uploader_url = None
2460
2461 has_videos = True
2462
2463 if not playlist_title:
2464 try:
2465 # Some playlist URLs don't actually serve a playlist (e.g.
2466 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2467 next(self._entries(page, playlist_id))
2468 except StopIteration:
2469 has_videos = False
2470
2471 playlist = self.playlist_result(
2472 self._entries(page, playlist_id), playlist_id, playlist_title)
2473 playlist.update({
2474 'uploader': uploader,
2475 'uploader_id': uploader_id,
2476 'uploader_url': uploader_url,
2477 })
2478
2479 return has_videos, playlist
2480
2481 def _check_download_just_video(self, url, playlist_id):
2482 # Check if it's a video-specific URL
2483 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2484 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2485 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2486 'video id', default=None)
2487 if video_id:
2488 if self._downloader.params.get('noplaylist'):
2489 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2490 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2491 else:
2492 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2493 return video_id, None
2494 return None, None
2495
2496 def _real_extract(self, url):
2497 # Extract playlist id
2498 mobj = re.match(self._VALID_URL, url)
2499 if mobj is None:
2500 raise ExtractorError('Invalid URL: %s' % url)
2501 playlist_id = mobj.group(1) or mobj.group(2)
2502
2503 video_id, video = self._check_download_just_video(url, playlist_id)
2504 if video:
2505 return video
2506
2507 if playlist_id.startswith(('RD', 'UL', 'PU')):
2508 # Mixes require a custom extraction process
2509 return self._extract_mix(playlist_id)
2510
2511 has_videos, playlist = self._extract_playlist(playlist_id)
2512 if has_videos or not video_id:
2513 return playlist
2514
2515 # Some playlist URLs don't actually serve a playlist (see
2516 # https://github.com/rg3/youtube-dl/issues/10537).
2517 # Fallback to plain video extraction if there is a video id
2518 # along with playlist id.
2519 return self.url_result(video_id, 'Youtube', video_id=video_id)
2520
2521
2522 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2523 IE_DESC = 'YouTube.com channels'
2524 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2525 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2526 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2527 IE_NAME = 'youtube:channel'
2528 _TESTS = [{
2529 'note': 'paginated channel',
2530 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2531 'playlist_mincount': 91,
2532 'info_dict': {
2533 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2534 'title': 'Uploads from lex will',
2535 }
2536 }, {
2537 'note': 'Age restricted channel',
2538 # from https://www.youtube.com/user/DeusExOfficial
2539 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2540 'playlist_mincount': 64,
2541 'info_dict': {
2542 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2543 'title': 'Uploads from Deus Ex',
2544 },
2545 }, {
2546 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2547 'only_matching': True,
2548 }]
2549
2550 @classmethod
2551 def suitable(cls, url):
2552 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2553 else super(YoutubeChannelIE, cls).suitable(url))
2554
2555 def _build_template_url(self, url, channel_id):
2556 return self._TEMPLATE_URL % channel_id
2557
2558 def _real_extract(self, url):
2559 channel_id = self._match_id(url)
2560
2561 url = self._build_template_url(url, channel_id)
2562
2563 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2564 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2565 # otherwise fallback on channel by page extraction
2566 channel_page = self._download_webpage(
2567 url + '?view=57', channel_id,
2568 'Downloading channel page', fatal=False)
2569 if channel_page is False:
2570 channel_playlist_id = False
2571 else:
2572 channel_playlist_id = self._html_search_meta(
2573 'channelId', channel_page, 'channel id', default=None)
2574 if not channel_playlist_id:
2575 channel_url = self._html_search_meta(
2576 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2577 channel_page, 'channel url', default=None)
2578 if channel_url:
2579 channel_playlist_id = self._search_regex(
2580 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2581 channel_url, 'channel id', default=None)
2582 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2583 playlist_id = 'UU' + channel_playlist_id[2:]
2584 return self.url_result(
2585 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2586
2587 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2588 autogenerated = re.search(r'''(?x)
2589 class="[^"]*?(?:
2590 channel-header-autogenerated-label|
2591 yt-channel-title-autogenerated
2592 )[^"]*"''', channel_page) is not None
2593
2594 if autogenerated:
2595 # The videos are contained in a single page
2596 # the ajax pages can't be used, they are empty
2597 entries = [
2598 self.url_result(
2599 video_id, 'Youtube', video_id=video_id,
2600 video_title=video_title)
2601 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2602 return self.playlist_result(entries, channel_id)
2603
2604 try:
2605 next(self._entries(channel_page, channel_id))
2606 except StopIteration:
2607 alert_message = self._html_search_regex(
2608 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2609 channel_page, 'alert', default=None, group='alert')
2610 if alert_message:
2611 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2612
2613 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2614
2615
2616 class YoutubeUserIE(YoutubeChannelIE):
2617 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2618 _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_-]+)'
2619 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2620 IE_NAME = 'youtube:user'
2621
2622 _TESTS = [{
2623 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2624 'playlist_mincount': 320,
2625 'info_dict': {
2626 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2627 'title': 'Uploads from The Linux Foundation',
2628 }
2629 }, {
2630 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2631 # but not https://www.youtube.com/user/12minuteathlete/videos
2632 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2633 'playlist_mincount': 249,
2634 'info_dict': {
2635 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2636 'title': 'Uploads from 12 Minute Athlete',
2637 }
2638 }, {
2639 'url': 'ytuser:phihag',
2640 'only_matching': True,
2641 }, {
2642 'url': 'https://www.youtube.com/c/gametrailers',
2643 'only_matching': True,
2644 }, {
2645 'url': 'https://www.youtube.com/gametrailers',
2646 'only_matching': True,
2647 }, {
2648 # This channel is not available, geo restricted to JP
2649 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2650 'only_matching': True,
2651 }]
2652
2653 @classmethod
2654 def suitable(cls, url):
2655 # Don't return True if the url can be extracted with other youtube
2656 # extractor, the regex would is too permissive and it would match.
2657 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2658 if any(ie.suitable(url) for ie in other_yt_ies):
2659 return False
2660 else:
2661 return super(YoutubeUserIE, cls).suitable(url)
2662
2663 def _build_template_url(self, url, channel_id):
2664 mobj = re.match(self._VALID_URL, url)
2665 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2666
2667
2668 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2669 IE_DESC = 'YouTube.com live streams'
2670 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2671 IE_NAME = 'youtube:live'
2672
2673 _TESTS = [{
2674 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2675 'info_dict': {
2676 'id': 'a48o2S1cPoo',
2677 'ext': 'mp4',
2678 'title': 'The Young Turks - Live Main Show',
2679 'uploader': 'The Young Turks',
2680 'uploader_id': 'TheYoungTurks',
2681 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2682 'upload_date': '20150715',
2683 'license': 'Standard YouTube License',
2684 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2685 'categories': ['News & Politics'],
2686 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2687 'like_count': int,
2688 'dislike_count': int,
2689 },
2690 'params': {
2691 'skip_download': True,
2692 },
2693 }, {
2694 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2695 'only_matching': True,
2696 }, {
2697 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2698 'only_matching': True,
2699 }, {
2700 'url': 'https://www.youtube.com/TheYoungTurks/live',
2701 'only_matching': True,
2702 }]
2703
2704 def _real_extract(self, url):
2705 mobj = re.match(self._VALID_URL, url)
2706 channel_id = mobj.group('id')
2707 base_url = mobj.group('base_url')
2708 webpage = self._download_webpage(url, channel_id, fatal=False)
2709 if webpage:
2710 page_type = self._og_search_property(
2711 'type', webpage, 'page type', default='')
2712 video_id = self._html_search_meta(
2713 'videoId', webpage, 'video id', default=None)
2714 if page_type.startswith('video') and video_id and re.match(
2715 r'^[0-9A-Za-z_-]{11}$', video_id):
2716 return self.url_result(video_id, YoutubeIE.ie_key())
2717 return self.url_result(base_url)
2718
2719
2720 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2721 IE_DESC = 'YouTube.com user/channel playlists'
2722 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2723 IE_NAME = 'youtube:playlists'
2724
2725 _TESTS = [{
2726 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2727 'playlist_mincount': 4,
2728 'info_dict': {
2729 'id': 'ThirstForScience',
2730 'title': 'Thirst for Science',
2731 },
2732 }, {
2733 # with "Load more" button
2734 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2735 'playlist_mincount': 70,
2736 'info_dict': {
2737 'id': 'igorkle1',
2738 'title': 'Игорь Клейнер',
2739 },
2740 }, {
2741 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2742 'playlist_mincount': 17,
2743 'info_dict': {
2744 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2745 'title': 'Chem Player',
2746 },
2747 }]
2748
2749
2750 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2751 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2752
2753
2754 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2755 IE_DESC = 'YouTube.com searches'
2756 # there doesn't appear to be a real limit, for example if you search for
2757 # 'python' you get more than 8.000.000 results
2758 _MAX_RESULTS = float('inf')
2759 IE_NAME = 'youtube:search'
2760 _SEARCH_KEY = 'ytsearch'
2761 _EXTRA_QUERY_ARGS = {}
2762 _TESTS = []
2763
2764 def _get_n_results(self, query, n):
2765 """Get a specified number of results for a query"""
2766
2767 videos = []
2768 limit = n
2769
2770 url_query = {
2771 'search_query': query.encode('utf-8'),
2772 }
2773 url_query.update(self._EXTRA_QUERY_ARGS)
2774 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2775
2776 for pagenum in itertools.count(1):
2777 data = self._download_json(
2778 result_url, video_id='query "%s"' % query,
2779 note='Downloading page %s' % pagenum,
2780 errnote='Unable to download API page',
2781 query={'spf': 'navigate'})
2782 html_content = data[1]['body']['content']
2783
2784 if 'class="search-message' in html_content:
2785 raise ExtractorError(
2786 '[youtube] No video results', expected=True)
2787
2788 new_videos = list(self._process_page(html_content))
2789 videos += new_videos
2790 if not new_videos or len(videos) > limit:
2791 break
2792 next_link = self._html_search_regex(
2793 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2794 html_content, 'next link', default=None)
2795 if next_link is None:
2796 break
2797 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2798
2799 if len(videos) > n:
2800 videos = videos[:n]
2801 return self.playlist_result(videos, query)
2802
2803
2804 class YoutubeSearchDateIE(YoutubeSearchIE):
2805 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2806 _SEARCH_KEY = 'ytsearchdate'
2807 IE_DESC = 'YouTube.com searches, newest videos first'
2808 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2809
2810
2811 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2812 IE_DESC = 'YouTube.com search URLs'
2813 IE_NAME = 'youtube:search_url'
2814 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2815 _TESTS = [{
2816 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2817 'playlist_mincount': 5,
2818 'info_dict': {
2819 'title': 'youtube-dl test video',
2820 }
2821 }, {
2822 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2823 'only_matching': True,
2824 }]
2825
2826 def _real_extract(self, url):
2827 mobj = re.match(self._VALID_URL, url)
2828 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2829 webpage = self._download_webpage(url, query)
2830 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2831
2832
2833 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2834 IE_DESC = 'YouTube.com (multi-season) shows'
2835 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2836 IE_NAME = 'youtube:show'
2837 _TESTS = [{
2838 'url': 'https://www.youtube.com/show/airdisasters',
2839 'playlist_mincount': 5,
2840 'info_dict': {
2841 'id': 'airdisasters',
2842 'title': 'Air Disasters',
2843 }
2844 }]
2845
2846 def _real_extract(self, url):
2847 playlist_id = self._match_id(url)
2848 return super(YoutubeShowIE, self)._real_extract(
2849 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2850
2851
2852 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2853 """
2854 Base class for feed extractors
2855 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2856 """
2857 _LOGIN_REQUIRED = True
2858
2859 @property
2860 def IE_NAME(self):
2861 return 'youtube:%s' % self._FEED_NAME
2862
2863 def _real_initialize(self):
2864 self._login()
2865
2866 def _entries(self, page):
2867 # The extraction process is the same as for playlists, but the regex
2868 # for the video ids doesn't contain an index
2869 ids = []
2870 more_widget_html = content_html = page
2871 for page_num in itertools.count(1):
2872 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2873
2874 # 'recommended' feed has infinite 'load more' and each new portion spins
2875 # the same videos in (sometimes) slightly different order, so we'll check
2876 # for unicity and break when portion has no new videos
2877 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2878 if not new_ids:
2879 break
2880
2881 ids.extend(new_ids)
2882
2883 for entry in self._ids_to_results(new_ids):
2884 yield entry
2885
2886 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2887 if not mobj:
2888 break
2889
2890 more = self._download_json(
2891 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2892 'Downloading page #%s' % page_num,
2893 transform_source=uppercase_escape)
2894 content_html = more['content_html']
2895 more_widget_html = more['load_more_widget_html']
2896
2897 def _real_extract(self, url):
2898 page = self._download_webpage(
2899 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2900 self._PLAYLIST_TITLE)
2901 return self.playlist_result(
2902 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2903
2904
2905 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2906 IE_NAME = 'youtube:watchlater'
2907 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2908 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2909
2910 _TESTS = [{
2911 'url': 'https://www.youtube.com/playlist?list=WL',
2912 'only_matching': True,
2913 }, {
2914 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2915 'only_matching': True,
2916 }]
2917
2918 def _real_extract(self, url):
2919 _, video = self._check_download_just_video(url, 'WL')
2920 if video:
2921 return video
2922 _, playlist = self._extract_playlist('WL')
2923 return playlist
2924
2925
2926 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2927 IE_NAME = 'youtube:favorites'
2928 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2929 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2930 _LOGIN_REQUIRED = True
2931
2932 def _real_extract(self, url):
2933 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2934 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2935 return self.url_result(playlist_id, 'YoutubePlaylist')
2936
2937
2938 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2939 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2940 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2941 _FEED_NAME = 'recommended'
2942 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2943
2944
2945 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2946 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2947 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2948 _FEED_NAME = 'subscriptions'
2949 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2950
2951
2952 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2953 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2954 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2955 _FEED_NAME = 'history'
2956 _PLAYLIST_TITLE = 'Youtube History'
2957
2958
2959 class YoutubeTruncatedURLIE(InfoExtractor):
2960 IE_NAME = 'youtube:truncated_url'
2961 IE_DESC = False # Do not list
2962 _VALID_URL = r'''(?x)
2963 (?:https?://)?
2964 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2965 (?:watch\?(?:
2966 feature=[a-z_]+|
2967 annotation_id=annotation_[^&]+|
2968 x-yt-cl=[0-9]+|
2969 hl=[^&]*|
2970 t=[0-9]+
2971 )?
2972 |
2973 attribution_link\?a=[^&]+
2974 )
2975 $
2976 '''
2977
2978 _TESTS = [{
2979 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2980 'only_matching': True,
2981 }, {
2982 'url': 'https://www.youtube.com/watch?',
2983 'only_matching': True,
2984 }, {
2985 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2986 'only_matching': True,
2987 }, {
2988 'url': 'https://www.youtube.com/watch?feature=foo',
2989 'only_matching': True,
2990 }, {
2991 'url': 'https://www.youtube.com/watch?hl=en-GB',
2992 'only_matching': True,
2993 }, {
2994 'url': 'https://www.youtube.com/watch?t=2372',
2995 'only_matching': True,
2996 }]
2997
2998 def _real_extract(self, url):
2999 raise ExtractorError(
3000 'Did you forget to quote the URL? Remember that & is a meta '
3001 'character in most shells, so you want to put the URL in quotes, '
3002 'like youtube-dl '
3003 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3004 ' or simply youtube-dl BaW_jenozKc .',
3005 expected=True)
3006
3007
3008 class YoutubeTruncatedIDIE(InfoExtractor):
3009 IE_NAME = 'youtube:truncated_id'
3010 IE_DESC = False # Do not list
3011 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3012
3013 _TESTS = [{
3014 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3015 'only_matching': True,
3016 }]
3017
3018 def _real_extract(self, url):
3019 video_id = self._match_id(url)
3020 raise ExtractorError(
3021 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3022 expected=True)