]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
New upstream version 2019.01.16
[youtubedl] / 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 "\'/\\Ƥā†­š•',
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 'description': 'test chars: "\'/\\Ƥā†­š•\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 .',
502 'categories': ['Science & Technology'],
503 'tags': ['youtube-dl'],
504 'duration': 10,
505 'view_count': int,
506 'like_count': int,
507 'dislike_count': int,
508 'start_time': 1,
509 'end_time': 9,
510 }
511 },
512 {
513 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
514 'note': 'Test generic use_cipher_signature video (#897)',
515 'info_dict': {
516 'id': 'UxxajLWwzqY',
517 'ext': 'mp4',
518 'upload_date': '20120506',
519 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
520 'alt_title': 'I Love It (feat. Charli XCX)',
521 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
522 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
523 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
524 'iconic ep', 'iconic', 'love', 'it'],
525 'duration': 180,
526 'uploader': 'Icona Pop',
527 'uploader_id': 'IconaPop',
528 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
529 'creator': 'Icona Pop',
530 'track': 'I Love It (feat. Charli XCX)',
531 'artist': 'Icona Pop',
532 }
533 },
534 {
535 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
536 'note': 'Test VEVO video with age protection (#956)',
537 'info_dict': {
538 'id': '07FYdnEawAQ',
539 'ext': 'mp4',
540 'upload_date': '20130703',
541 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
542 'alt_title': 'Tunnel Vision',
543 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
544 'duration': 419,
545 'uploader': 'justintimberlakeVEVO',
546 'uploader_id': 'justintimberlakeVEVO',
547 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
548 'creator': 'Justin Timberlake',
549 'track': 'Tunnel Vision',
550 'artist': 'Justin Timberlake',
551 'age_limit': 18,
552 }
553 },
554 {
555 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
556 'note': 'Embed-only video (#1746)',
557 'info_dict': {
558 'id': 'yZIXLfi8CZQ',
559 'ext': 'mp4',
560 'upload_date': '20120608',
561 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
562 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
563 'uploader': 'SET India',
564 'uploader_id': 'setindia',
565 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
566 'age_limit': 18,
567 }
568 },
569 {
570 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
571 'note': 'Use the first video ID in the URL',
572 'info_dict': {
573 'id': 'BaW_jenozKc',
574 'ext': 'mp4',
575 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
576 'uploader': 'Philipp Hagemeister',
577 'uploader_id': 'phihag',
578 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
579 'upload_date': '20121002',
580 'description': 'test chars: "\'/\\Ƥā†­š•\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 .',
581 'categories': ['Science & Technology'],
582 'tags': ['youtube-dl'],
583 'duration': 10,
584 'view_count': int,
585 'like_count': int,
586 'dislike_count': int,
587 },
588 'params': {
589 'skip_download': True,
590 },
591 },
592 {
593 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
594 'note': '256k DASH audio (format 141) via DASH manifest',
595 'info_dict': {
596 'id': 'a9LDPn-MO4I',
597 'ext': 'm4a',
598 'upload_date': '20121002',
599 'uploader_id': '8KVIDEO',
600 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
601 'description': '',
602 'uploader': '8KVIDEO',
603 'title': 'UHDTV TEST 8K VIDEO.mp4'
604 },
605 'params': {
606 'youtube_include_dash_manifest': True,
607 'format': '141',
608 },
609 'skip': 'format 141 not served anymore',
610 },
611 # DASH manifest with encrypted signature
612 {
613 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
614 'info_dict': {
615 'id': 'IB3lcPjvWLA',
616 'ext': 'm4a',
617 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
618 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
619 'duration': 244,
620 'uploader': 'AfrojackVEVO',
621 'uploader_id': 'AfrojackVEVO',
622 'upload_date': '20131011',
623 },
624 'params': {
625 'youtube_include_dash_manifest': True,
626 'format': '141/bestaudio[ext=m4a]',
627 },
628 },
629 # JS player signature function name containing $
630 {
631 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
632 'info_dict': {
633 'id': 'nfWlot6h_JM',
634 'ext': 'm4a',
635 'title': 'Taylor Swift - Shake It Off',
636 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
637 'duration': 242,
638 'uploader': 'TaylorSwiftVEVO',
639 'uploader_id': 'TaylorSwiftVEVO',
640 'upload_date': '20140818',
641 'creator': 'Taylor Swift',
642 },
643 'params': {
644 'youtube_include_dash_manifest': True,
645 'format': '141/bestaudio[ext=m4a]',
646 },
647 },
648 # Controversy video
649 {
650 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
651 'info_dict': {
652 'id': 'T4XJQO3qol8',
653 'ext': 'mp4',
654 'duration': 219,
655 'upload_date': '20100909',
656 'uploader': 'Amazing Atheist',
657 'uploader_id': 'TheAmazingAtheist',
658 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
659 'title': 'Burning Everyone\'s Koran',
660 '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',
661 }
662 },
663 # Normal age-gate video (No vevo, embed allowed)
664 {
665 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
666 'info_dict': {
667 'id': 'HtVdAasjOgU',
668 'ext': 'mp4',
669 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
670 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
671 'duration': 142,
672 'uploader': 'The Witcher',
673 'uploader_id': 'WitcherGame',
674 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
675 'upload_date': '20140605',
676 'age_limit': 18,
677 },
678 },
679 # Age-gate video with encrypted signature
680 {
681 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
682 'info_dict': {
683 'id': '6kLq3WMV1nU',
684 'ext': 'mp4',
685 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
686 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
687 'duration': 246,
688 'uploader': 'LloydVEVO',
689 'uploader_id': 'LloydVEVO',
690 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
691 'upload_date': '20110629',
692 'age_limit': 18,
693 },
694 },
695 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
696 # YouTube Red ad is not captured for creator
697 {
698 'url': '__2ABJjxzNo',
699 'info_dict': {
700 'id': '__2ABJjxzNo',
701 'ext': 'mp4',
702 'duration': 266,
703 'upload_date': '20100430',
704 'uploader_id': 'deadmau5',
705 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
706 'creator': 'deadmau5',
707 'description': 'md5:12c56784b8032162bb936a5f76d55360',
708 'uploader': 'deadmau5',
709 'title': 'Deadmau5 - Some Chords (HD)',
710 'alt_title': 'Some Chords',
711 },
712 'expected_warnings': [
713 'DASH manifest missing',
714 ]
715 },
716 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
717 {
718 'url': 'lqQg6PlCWgI',
719 'info_dict': {
720 'id': 'lqQg6PlCWgI',
721 'ext': 'mp4',
722 'duration': 6085,
723 'upload_date': '20150827',
724 'uploader_id': 'olympic',
725 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
726 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
727 'uploader': 'Olympic',
728 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
729 },
730 'params': {
731 'skip_download': 'requires avconv',
732 }
733 },
734 # Non-square pixels
735 {
736 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
737 'info_dict': {
738 'id': '_b-2C3KPAM0',
739 'ext': 'mp4',
740 'stretched_ratio': 16 / 9.,
741 'duration': 85,
742 'upload_date': '20110310',
743 'uploader_id': 'AllenMeow',
744 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
745 'description': 'made by Wacom from Korea | 字幕&åŠ ę²¹ę·»é†‹ by TY\'s Allen | ę„Ÿč¬heylisa00cavey1001同å­øē†±ęƒ…ęä¾›ę¢—åŠēæ»č­Æ',
746 'uploader': 'å­«į„‹į„…',
747 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
748 },
749 },
750 # url_encoded_fmt_stream_map is empty string
751 {
752 'url': 'qEJwOuvDf7I',
753 'info_dict': {
754 'id': 'qEJwOuvDf7I',
755 'ext': 'webm',
756 'title': 'ŠžŠ±ŃŃƒŠ¶Š“ŠµŠ½ŠøŠµ суŠ“ŠµŠ±Š½Š¾Š¹ ŠæрŠ°ŠŗтŠøŠŗŠø ŠæŠ¾ Š²Ń‹Š±Š¾Ń€Š°Š¼ 14 сŠµŠ½Ń‚яŠ±Ń€Ń 2014 Š³Š¾Š“Š° Š² Š”Š°Š½Šŗт-ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Šµ',
757 'description': '',
758 'upload_date': '20150404',
759 'uploader_id': 'spbelect',
760 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
761 },
762 'params': {
763 'skip_download': 'requires avconv',
764 },
765 'skip': 'This live event has ended.',
766 },
767 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
768 {
769 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
770 'info_dict': {
771 'id': 'FIl7x6_3R5Y',
772 'ext': 'webm',
773 'title': 'md5:7b81415841e02ecd4313668cde88737a',
774 'description': 'md5:116377fd2963b81ec4ce64b542173306',
775 'duration': 220,
776 'upload_date': '20150625',
777 'uploader_id': 'dorappi2000',
778 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
779 'uploader': 'dorappi2000',
780 'formats': 'mincount:31',
781 },
782 'skip': 'not actual anymore',
783 },
784 # DASH manifest with segment_list
785 {
786 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
787 'md5': '8ce563a1d667b599d21064e982ab9e31',
788 'info_dict': {
789 'id': 'CsmdDsKjzN8',
790 'ext': 'mp4',
791 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
792 'uploader': 'Airtek',
793 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
794 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
795 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
796 },
797 'params': {
798 'youtube_include_dash_manifest': True,
799 'format': '135', # bestvideo
800 },
801 'skip': 'This live event has ended.',
802 },
803 {
804 # Multifeed videos (multiple cameras), URL is for Main Camera
805 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
806 'info_dict': {
807 'id': 'jqWvoWXjCVs',
808 'title': 'teamPGP: Rocket League Noob Stream',
809 'description': 'md5:dc7872fb300e143831327f1bae3af010',
810 },
811 'playlist': [{
812 'info_dict': {
813 'id': 'jqWvoWXjCVs',
814 'ext': 'mp4',
815 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
816 'description': 'md5:dc7872fb300e143831327f1bae3af010',
817 'duration': 7335,
818 'upload_date': '20150721',
819 'uploader': 'Beer Games Beer',
820 'uploader_id': 'beergamesbeer',
821 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
822 'license': 'Standard YouTube License',
823 },
824 }, {
825 'info_dict': {
826 'id': '6h8e8xoXJzg',
827 'ext': 'mp4',
828 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
829 'description': 'md5:dc7872fb300e143831327f1bae3af010',
830 'duration': 7337,
831 'upload_date': '20150721',
832 'uploader': 'Beer Games Beer',
833 'uploader_id': 'beergamesbeer',
834 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
835 'license': 'Standard YouTube License',
836 },
837 }, {
838 'info_dict': {
839 'id': 'PUOgX5z9xZw',
840 'ext': 'mp4',
841 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
842 'description': 'md5:dc7872fb300e143831327f1bae3af010',
843 'duration': 7337,
844 'upload_date': '20150721',
845 'uploader': 'Beer Games Beer',
846 'uploader_id': 'beergamesbeer',
847 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
848 'license': 'Standard YouTube License',
849 },
850 }, {
851 'info_dict': {
852 'id': 'teuwxikvS5k',
853 'ext': 'mp4',
854 'title': 'teamPGP: Rocket League Noob Stream (zim)',
855 'description': 'md5:dc7872fb300e143831327f1bae3af010',
856 'duration': 7334,
857 'upload_date': '20150721',
858 'uploader': 'Beer Games Beer',
859 'uploader_id': 'beergamesbeer',
860 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
861 'license': 'Standard YouTube License',
862 },
863 }],
864 'params': {
865 'skip_download': True,
866 },
867 'skip': 'This video is not available.',
868 },
869 {
870 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
871 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
872 'info_dict': {
873 'id': 'gVfLd0zydlo',
874 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
875 },
876 'playlist_count': 2,
877 'skip': 'Not multifeed anymore',
878 },
879 {
880 'url': 'https://vid.plus/FlRa-iH7PGw',
881 'only_matching': True,
882 },
883 {
884 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
885 'only_matching': True,
886 },
887 {
888 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
889 # Also tests cut-off URL expansion in video description (see
890 # https://github.com/rg3/youtube-dl/issues/1892,
891 # https://github.com/rg3/youtube-dl/issues/8164)
892 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
893 'info_dict': {
894 'id': 'lsguqyKfVQg',
895 'ext': 'mp4',
896 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
897 'alt_title': 'Dark Walk - Position Music',
898 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
899 'duration': 133,
900 'upload_date': '20151119',
901 'uploader_id': 'IronSoulElf',
902 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
903 'uploader': 'IronSoulElf',
904 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
905 'track': 'Dark Walk - Position Music',
906 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
907 },
908 'params': {
909 'skip_download': True,
910 },
911 },
912 {
913 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
914 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
915 'only_matching': True,
916 },
917 {
918 # Video with yt:stretch=17:0
919 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
920 'info_dict': {
921 'id': 'Q39EVAstoRM',
922 'ext': 'mp4',
923 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
924 'description': 'md5:ee18a25c350637c8faff806845bddee9',
925 'upload_date': '20151107',
926 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
927 'uploader': 'CH GAMER DROID',
928 },
929 'params': {
930 'skip_download': True,
931 },
932 'skip': 'This video does not exist.',
933 },
934 {
935 # Video licensed under Creative Commons
936 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
937 'info_dict': {
938 'id': 'M4gD1WSo5mA',
939 'ext': 'mp4',
940 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
941 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
942 'duration': 721,
943 'upload_date': '20150127',
944 'uploader_id': 'BerkmanCenter',
945 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
946 'uploader': 'The Berkman Klein Center for Internet & Society',
947 'license': 'Creative Commons Attribution license (reuse allowed)',
948 },
949 'params': {
950 'skip_download': True,
951 },
952 },
953 {
954 # Channel-like uploader_url
955 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
956 'info_dict': {
957 'id': 'eQcmzGIKrzg',
958 'ext': 'mp4',
959 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
960 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
961 'duration': 4060,
962 'upload_date': '20151119',
963 'uploader': 'Bernie Sanders',
964 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
965 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
966 'license': 'Creative Commons Attribution license (reuse allowed)',
967 },
968 'params': {
969 'skip_download': True,
970 },
971 },
972 {
973 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
974 'only_matching': True,
975 },
976 {
977 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
978 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
979 'only_matching': True,
980 },
981 {
982 # Rental video preview
983 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
984 'info_dict': {
985 'id': 'uGpuVWrhIzE',
986 'ext': 'mp4',
987 'title': 'Piku - Trailer',
988 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
989 'upload_date': '20150811',
990 'uploader': 'FlixMatrix',
991 'uploader_id': 'FlixMatrixKaravan',
992 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
993 'license': 'Standard YouTube License',
994 },
995 'params': {
996 'skip_download': True,
997 },
998 'skip': 'This video is not available.',
999 },
1000 {
1001 # YouTube Red video with episode data
1002 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1003 'info_dict': {
1004 'id': 'iqKdEhx-dD4',
1005 'ext': 'mp4',
1006 'title': 'Isolation - Mind Field (Ep 1)',
1007 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1008 'duration': 2085,
1009 'upload_date': '20170118',
1010 'uploader': 'Vsauce',
1011 'uploader_id': 'Vsauce',
1012 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1013 'series': 'Mind Field',
1014 'season_number': 1,
1015 'episode_number': 1,
1016 },
1017 'params': {
1018 'skip_download': True,
1019 },
1020 'expected_warnings': [
1021 'Skipping DASH manifest',
1022 ],
1023 },
1024 {
1025 # The following content has been identified by the YouTube community
1026 # as inappropriate or offensive to some audiences.
1027 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1028 'info_dict': {
1029 'id': '6SJNVb0GnPI',
1030 'ext': 'mp4',
1031 'title': 'Race Differences in Intelligence',
1032 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1033 'duration': 965,
1034 'upload_date': '20140124',
1035 'uploader': 'New Century Foundation',
1036 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1037 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1038 },
1039 'params': {
1040 'skip_download': True,
1041 },
1042 },
1043 {
1044 # itag 212
1045 'url': '1t24XAntNCY',
1046 'only_matching': True,
1047 },
1048 {
1049 # geo restricted to JP
1050 'url': 'sJL6WA-aGkQ',
1051 'only_matching': True,
1052 },
1053 {
1054 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1055 'only_matching': True,
1056 },
1057 {
1058 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1059 'only_matching': True,
1060 },
1061 {
1062 # DRM protected
1063 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1064 'only_matching': True,
1065 },
1066 {
1067 # Video with unsupported adaptive stream type formats
1068 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1069 'info_dict': {
1070 'id': 'Z4Vy8R84T1U',
1071 'ext': 'mp4',
1072 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1073 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1074 'duration': 433,
1075 'upload_date': '20130923',
1076 'uploader': 'Amelia Putri Harwita',
1077 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1078 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1079 'formats': 'maxcount:10',
1080 },
1081 'params': {
1082 'skip_download': True,
1083 'youtube_include_dash_manifest': False,
1084 },
1085 }
1086 ]
1087
1088 def __init__(self, *args, **kwargs):
1089 super(YoutubeIE, self).__init__(*args, **kwargs)
1090 self._player_cache = {}
1091
1092 def report_video_info_webpage_download(self, video_id):
1093 """Report attempt to download video info webpage."""
1094 self.to_screen('%s: Downloading video info webpage' % video_id)
1095
1096 def report_information_extraction(self, video_id):
1097 """Report attempt to extract video information."""
1098 self.to_screen('%s: Extracting video information' % video_id)
1099
1100 def report_unavailable_format(self, video_id, format):
1101 """Report extracted video URL."""
1102 self.to_screen('%s: Format %s not available' % (video_id, format))
1103
1104 def report_rtmp_download(self):
1105 """Indicate the download will use the RTMP protocol."""
1106 self.to_screen('RTMP download detected')
1107
1108 def _signature_cache_id(self, example_sig):
1109 """ Return a string representation of a signature """
1110 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1111
1112 def _extract_signature_function(self, video_id, player_url, example_sig):
1113 id_m = re.match(
1114 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1115 player_url)
1116 if not id_m:
1117 raise ExtractorError('Cannot identify player %r' % player_url)
1118 player_type = id_m.group('ext')
1119 player_id = id_m.group('id')
1120
1121 # Read from filesystem cache
1122 func_id = '%s_%s_%s' % (
1123 player_type, player_id, self._signature_cache_id(example_sig))
1124 assert os.path.basename(func_id) == func_id
1125
1126 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1127 if cache_spec is not None:
1128 return lambda s: ''.join(s[i] for i in cache_spec)
1129
1130 download_note = (
1131 'Downloading player %s' % player_url
1132 if self._downloader.params.get('verbose') else
1133 'Downloading %s player %s' % (player_type, player_id)
1134 )
1135 if player_type == 'js':
1136 code = self._download_webpage(
1137 player_url, video_id,
1138 note=download_note,
1139 errnote='Download of %s failed' % player_url)
1140 res = self._parse_sig_js(code)
1141 elif player_type == 'swf':
1142 urlh = self._request_webpage(
1143 player_url, video_id,
1144 note=download_note,
1145 errnote='Download of %s failed' % player_url)
1146 code = urlh.read()
1147 res = self._parse_sig_swf(code)
1148 else:
1149 assert False, 'Invalid player type %r' % player_type
1150
1151 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1152 cache_res = res(test_string)
1153 cache_spec = [ord(c) for c in cache_res]
1154
1155 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1156 return res
1157
1158 def _print_sig_code(self, func, example_sig):
1159 def gen_sig_code(idxs):
1160 def _genslice(start, end, step):
1161 starts = '' if start == 0 else str(start)
1162 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1163 steps = '' if step == 1 else (':%d' % step)
1164 return 's[%s%s%s]' % (starts, ends, steps)
1165
1166 step = None
1167 # Quelch pyflakes warnings - start will be set when step is set
1168 start = '(Never used)'
1169 for i, prev in zip(idxs[1:], idxs[:-1]):
1170 if step is not None:
1171 if i - prev == step:
1172 continue
1173 yield _genslice(start, prev, step)
1174 step = None
1175 continue
1176 if i - prev in [-1, 1]:
1177 step = i - prev
1178 start = prev
1179 continue
1180 else:
1181 yield 's[%d]' % prev
1182 if step is None:
1183 yield 's[%d]' % i
1184 else:
1185 yield _genslice(start, i, step)
1186
1187 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1188 cache_res = func(test_string)
1189 cache_spec = [ord(c) for c in cache_res]
1190 expr_code = ' + '.join(gen_sig_code(cache_spec))
1191 signature_id_tuple = '(%s)' % (
1192 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1193 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1194 ' return %s\n') % (signature_id_tuple, expr_code)
1195 self.to_screen('Extracted signature function:\n' + code)
1196
1197 def _parse_sig_js(self, jscode):
1198 funcname = self._search_regex(
1199 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1200 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1201 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1202 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1203 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1204 jscode, 'Initial JS player signature function name', group='sig')
1205
1206 jsi = JSInterpreter(jscode)
1207 initial_function = jsi.extract_function(funcname)
1208 return lambda s: initial_function([s])
1209
1210 def _parse_sig_swf(self, file_contents):
1211 swfi = SWFInterpreter(file_contents)
1212 TARGET_CLASSNAME = 'SignatureDecipher'
1213 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1214 initial_function = swfi.extract_function(searched_class, 'decipher')
1215 return lambda s: initial_function([s])
1216
1217 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1218 """Turn the encrypted s field into a working signature"""
1219
1220 if player_url is None:
1221 raise ExtractorError('Cannot decrypt signature without player_url')
1222
1223 if player_url.startswith('//'):
1224 player_url = 'https:' + player_url
1225 elif not re.match(r'https?://', player_url):
1226 player_url = compat_urlparse.urljoin(
1227 'https://www.youtube.com', player_url)
1228 try:
1229 player_id = (player_url, self._signature_cache_id(s))
1230 if player_id not in self._player_cache:
1231 func = self._extract_signature_function(
1232 video_id, player_url, s
1233 )
1234 self._player_cache[player_id] = func
1235 func = self._player_cache[player_id]
1236 if self._downloader.params.get('youtube_print_sig_code'):
1237 self._print_sig_code(func, s)
1238 return func(s)
1239 except Exception as e:
1240 tb = traceback.format_exc()
1241 raise ExtractorError(
1242 'Signature extraction failed: ' + tb, cause=e)
1243
1244 def _get_subtitles(self, video_id, webpage):
1245 try:
1246 subs_doc = self._download_xml(
1247 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1248 video_id, note=False)
1249 except ExtractorError as err:
1250 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1251 return {}
1252
1253 sub_lang_list = {}
1254 for track in subs_doc.findall('track'):
1255 lang = track.attrib['lang_code']
1256 if lang in sub_lang_list:
1257 continue
1258 sub_formats = []
1259 for ext in self._SUBTITLE_FORMATS:
1260 params = compat_urllib_parse_urlencode({
1261 'lang': lang,
1262 'v': video_id,
1263 'fmt': ext,
1264 'name': track.attrib['name'].encode('utf-8'),
1265 })
1266 sub_formats.append({
1267 'url': 'https://www.youtube.com/api/timedtext?' + params,
1268 'ext': ext,
1269 })
1270 sub_lang_list[lang] = sub_formats
1271 if not sub_lang_list:
1272 self._downloader.report_warning('video doesn\'t have subtitles')
1273 return {}
1274 return sub_lang_list
1275
1276 def _get_ytplayer_config(self, video_id, webpage):
1277 patterns = (
1278 # User data may contain arbitrary character sequences that may affect
1279 # JSON extraction with regex, e.g. when '};' is contained the second
1280 # regex won't capture the whole JSON. Yet working around by trying more
1281 # concrete regex first keeping in mind proper quoted string handling
1282 # to be implemented in future that will replace this workaround (see
1283 # https://github.com/rg3/youtube-dl/issues/7468,
1284 # https://github.com/rg3/youtube-dl/pull/7599)
1285 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1286 r';ytplayer\.config\s*=\s*({.+?});',
1287 )
1288 config = self._search_regex(
1289 patterns, webpage, 'ytplayer.config', default=None)
1290 if config:
1291 return self._parse_json(
1292 uppercase_escape(config), video_id, fatal=False)
1293
1294 def _get_automatic_captions(self, video_id, webpage):
1295 """We need the webpage for getting the captions url, pass it as an
1296 argument to speed up the process."""
1297 self.to_screen('%s: Looking for automatic captions' % video_id)
1298 player_config = self._get_ytplayer_config(video_id, webpage)
1299 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1300 if not player_config:
1301 self._downloader.report_warning(err_msg)
1302 return {}
1303 try:
1304 args = player_config['args']
1305 caption_url = args.get('ttsurl')
1306 if caption_url:
1307 timestamp = args['timestamp']
1308 # We get the available subtitles
1309 list_params = compat_urllib_parse_urlencode({
1310 'type': 'list',
1311 'tlangs': 1,
1312 'asrs': 1,
1313 })
1314 list_url = caption_url + '&' + list_params
1315 caption_list = self._download_xml(list_url, video_id)
1316 original_lang_node = caption_list.find('track')
1317 if original_lang_node is None:
1318 self._downloader.report_warning('Video doesn\'t have automatic captions')
1319 return {}
1320 original_lang = original_lang_node.attrib['lang_code']
1321 caption_kind = original_lang_node.attrib.get('kind', '')
1322
1323 sub_lang_list = {}
1324 for lang_node in caption_list.findall('target'):
1325 sub_lang = lang_node.attrib['lang_code']
1326 sub_formats = []
1327 for ext in self._SUBTITLE_FORMATS:
1328 params = compat_urllib_parse_urlencode({
1329 'lang': original_lang,
1330 'tlang': sub_lang,
1331 'fmt': ext,
1332 'ts': timestamp,
1333 'kind': caption_kind,
1334 })
1335 sub_formats.append({
1336 'url': caption_url + '&' + params,
1337 'ext': ext,
1338 })
1339 sub_lang_list[sub_lang] = sub_formats
1340 return sub_lang_list
1341
1342 def make_captions(sub_url, sub_langs):
1343 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1344 caption_qs = compat_parse_qs(parsed_sub_url.query)
1345 captions = {}
1346 for sub_lang in sub_langs:
1347 sub_formats = []
1348 for ext in self._SUBTITLE_FORMATS:
1349 caption_qs.update({
1350 'tlang': [sub_lang],
1351 'fmt': [ext],
1352 })
1353 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1354 query=compat_urllib_parse_urlencode(caption_qs, True)))
1355 sub_formats.append({
1356 'url': sub_url,
1357 'ext': ext,
1358 })
1359 captions[sub_lang] = sub_formats
1360 return captions
1361
1362 # New captions format as of 22.06.2017
1363 player_response = args.get('player_response')
1364 if player_response and isinstance(player_response, compat_str):
1365 player_response = self._parse_json(
1366 player_response, video_id, fatal=False)
1367 if player_response:
1368 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1369 base_url = renderer['captionTracks'][0]['baseUrl']
1370 sub_lang_list = []
1371 for lang in renderer['translationLanguages']:
1372 lang_code = lang.get('languageCode')
1373 if lang_code:
1374 sub_lang_list.append(lang_code)
1375 return make_captions(base_url, sub_lang_list)
1376
1377 # Some videos don't provide ttsurl but rather caption_tracks and
1378 # caption_translation_languages (e.g. 20LmZk1hakA)
1379 # Does not used anymore as of 22.06.2017
1380 caption_tracks = args['caption_tracks']
1381 caption_translation_languages = args['caption_translation_languages']
1382 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1383 sub_lang_list = []
1384 for lang in caption_translation_languages.split(','):
1385 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1386 sub_lang = lang_qs.get('lc', [None])[0]
1387 if sub_lang:
1388 sub_lang_list.append(sub_lang)
1389 return make_captions(caption_url, sub_lang_list)
1390 # An extractor error can be raise by the download process if there are
1391 # no automatic captions but there are subtitles
1392 except (KeyError, IndexError, ExtractorError):
1393 self._downloader.report_warning(err_msg)
1394 return {}
1395
1396 def _mark_watched(self, video_id, video_info, player_response):
1397 playback_url = url_or_none(try_get(
1398 player_response,
1399 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1400 video_info, lambda x: x['videostats_playback_base_url'][0]))
1401 if not playback_url:
1402 return
1403 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1404 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1405
1406 # cpn generation algorithm is reverse engineered from base.js.
1407 # In fact it works even with dummy cpn.
1408 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1409 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1410
1411 qs.update({
1412 'ver': ['2'],
1413 'cpn': [cpn],
1414 })
1415 playback_url = compat_urlparse.urlunparse(
1416 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1417
1418 self._download_webpage(
1419 playback_url, video_id, 'Marking watched',
1420 'Unable to mark watched', fatal=False)
1421
1422 @staticmethod
1423 def _extract_urls(webpage):
1424 # Embedded YouTube player
1425 entries = [
1426 unescapeHTML(mobj.group('url'))
1427 for mobj in re.finditer(r'''(?x)
1428 (?:
1429 <iframe[^>]+?src=|
1430 data-video-url=|
1431 <embed[^>]+?src=|
1432 embedSWF\(?:\s*|
1433 <object[^>]+data=|
1434 new\s+SWFObject\(
1435 )
1436 (["\'])
1437 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1438 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1439 \1''', webpage)]
1440
1441 # lazyYT YouTube embed
1442 entries.extend(list(map(
1443 unescapeHTML,
1444 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1445
1446 # Wordpress "YouTube Video Importer" plugin
1447 matches = re.findall(r'''(?x)<div[^>]+
1448 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1449 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1450 entries.extend(m[-1] for m in matches)
1451
1452 return entries
1453
1454 @staticmethod
1455 def _extract_url(webpage):
1456 urls = YoutubeIE._extract_urls(webpage)
1457 return urls[0] if urls else None
1458
1459 @classmethod
1460 def extract_id(cls, url):
1461 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1462 if mobj is None:
1463 raise ExtractorError('Invalid URL: %s' % url)
1464 video_id = mobj.group(2)
1465 return video_id
1466
1467 def _extract_annotations(self, video_id):
1468 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1469 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1470
1471 @staticmethod
1472 def _extract_chapters(description, duration):
1473 if not description:
1474 return None
1475 chapter_lines = re.findall(
1476 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1477 description)
1478 if not chapter_lines:
1479 return None
1480 chapters = []
1481 for next_num, (chapter_line, time_point) in enumerate(
1482 chapter_lines, start=1):
1483 start_time = parse_duration(time_point)
1484 if start_time is None:
1485 continue
1486 if start_time > duration:
1487 break
1488 end_time = (duration if next_num == len(chapter_lines)
1489 else parse_duration(chapter_lines[next_num][1]))
1490 if end_time is None:
1491 continue
1492 if end_time > duration:
1493 end_time = duration
1494 if start_time > end_time:
1495 break
1496 chapter_title = re.sub(
1497 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1498 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1499 chapters.append({
1500 'start_time': start_time,
1501 'end_time': end_time,
1502 'title': chapter_title,
1503 })
1504 return chapters
1505
1506 def _real_extract(self, url):
1507 url, smuggled_data = unsmuggle_url(url, {})
1508
1509 proto = (
1510 'http' if self._downloader.params.get('prefer_insecure', False)
1511 else 'https')
1512
1513 start_time = None
1514 end_time = None
1515 parsed_url = compat_urllib_parse_urlparse(url)
1516 for component in [parsed_url.fragment, parsed_url.query]:
1517 query = compat_parse_qs(component)
1518 if start_time is None and 't' in query:
1519 start_time = parse_duration(query['t'][0])
1520 if start_time is None and 'start' in query:
1521 start_time = parse_duration(query['start'][0])
1522 if end_time is None and 'end' in query:
1523 end_time = parse_duration(query['end'][0])
1524
1525 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1526 mobj = re.search(self._NEXT_URL_RE, url)
1527 if mobj:
1528 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1529 video_id = self.extract_id(url)
1530
1531 # Get video webpage
1532 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1533 video_webpage = self._download_webpage(url, video_id)
1534
1535 # Attempt to extract SWF player URL
1536 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1537 if mobj is not None:
1538 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1539 else:
1540 player_url = None
1541
1542 dash_mpds = []
1543
1544 def add_dash_mpd(video_info):
1545 dash_mpd = video_info.get('dashmpd')
1546 if dash_mpd and dash_mpd[0] not in dash_mpds:
1547 dash_mpds.append(dash_mpd[0])
1548
1549 def add_dash_mpd_pr(pl_response):
1550 dash_mpd = url_or_none(try_get(
1551 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1552 compat_str))
1553 if dash_mpd and dash_mpd not in dash_mpds:
1554 dash_mpds.append(dash_mpd)
1555
1556 is_live = None
1557 view_count = None
1558
1559 def extract_view_count(v_info):
1560 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1561
1562 player_response = {}
1563
1564 # Get video info
1565 embed_webpage = None
1566 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1567 age_gate = True
1568 # We simulate the access to the video from www.youtube.com/v/{video_id}
1569 # this can be viewed without login into Youtube
1570 url = proto + '://www.youtube.com/embed/%s' % video_id
1571 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1572 data = compat_urllib_parse_urlencode({
1573 'video_id': video_id,
1574 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1575 'sts': self._search_regex(
1576 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1577 })
1578 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1579 video_info_webpage = self._download_webpage(
1580 video_info_url, video_id,
1581 note='Refetching age-gated info webpage',
1582 errnote='unable to download video info webpage')
1583 video_info = compat_parse_qs(video_info_webpage)
1584 add_dash_mpd(video_info)
1585 else:
1586 age_gate = False
1587 video_info = None
1588 sts = None
1589 # Try looking directly into the video webpage
1590 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1591 if ytplayer_config:
1592 args = ytplayer_config['args']
1593 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1594 # Convert to the same format returned by compat_parse_qs
1595 video_info = dict((k, [v]) for k, v in args.items())
1596 add_dash_mpd(video_info)
1597 # Rental video is not rented but preview is available (e.g.
1598 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1599 # https://github.com/rg3/youtube-dl/issues/10532)
1600 if not video_info and args.get('ypc_vid'):
1601 return self.url_result(
1602 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1603 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1604 is_live = True
1605 sts = ytplayer_config.get('sts')
1606 if not player_response:
1607 pl_response = str_or_none(args.get('player_response'))
1608 if pl_response:
1609 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1610 if isinstance(pl_response, dict):
1611 player_response = pl_response
1612 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1613 add_dash_mpd_pr(player_response)
1614 # We also try looking in get_video_info since it may contain different dashmpd
1615 # URL that points to a DASH manifest with possibly different itag set (some itags
1616 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1617 # manifest pointed by get_video_info's dashmpd).
1618 # The general idea is to take a union of itags of both DASH manifests (for example
1619 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1620 self.report_video_info_webpage_download(video_id)
1621 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1622 query = {
1623 'video_id': video_id,
1624 'ps': 'default',
1625 'eurl': '',
1626 'gl': 'US',
1627 'hl': 'en',
1628 }
1629 if el:
1630 query['el'] = el
1631 if sts:
1632 query['sts'] = sts
1633 video_info_webpage = self._download_webpage(
1634 '%s://www.youtube.com/get_video_info' % proto,
1635 video_id, note=False,
1636 errnote='unable to download video info webpage',
1637 fatal=False, query=query)
1638 if not video_info_webpage:
1639 continue
1640 get_video_info = compat_parse_qs(video_info_webpage)
1641 if not player_response:
1642 pl_response = get_video_info.get('player_response', [None])[0]
1643 if isinstance(pl_response, dict):
1644 player_response = pl_response
1645 add_dash_mpd_pr(player_response)
1646 add_dash_mpd(get_video_info)
1647 if view_count is None:
1648 view_count = extract_view_count(get_video_info)
1649 if not video_info:
1650 video_info = get_video_info
1651 if 'token' in get_video_info:
1652 # Different get_video_info requests may report different results, e.g.
1653 # some may report video unavailability, but some may serve it without
1654 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1655 # the original webpage as well as el=info and el=embedded get_video_info
1656 # requests report video unavailability due to geo restriction while
1657 # el=detailpage succeeds and returns valid data). This is probably
1658 # due to YouTube measures against IP ranges of hosting providers.
1659 # Working around by preferring the first succeeded video_info containing
1660 # the token if no such video_info yet was found.
1661 if 'token' not in video_info:
1662 video_info = get_video_info
1663 break
1664
1665 def extract_unavailable_message():
1666 return self._html_search_regex(
1667 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1668 video_webpage, 'unavailable message', default=None)
1669
1670 if 'token' not in video_info:
1671 if 'reason' in video_info:
1672 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1673 regions_allowed = self._html_search_meta(
1674 'regionsAllowed', video_webpage, default=None)
1675 countries = regions_allowed.split(',') if regions_allowed else None
1676 self.raise_geo_restricted(
1677 msg=video_info['reason'][0], countries=countries)
1678 reason = video_info['reason'][0]
1679 if 'Invalid parameters' in reason:
1680 unavailable_message = extract_unavailable_message()
1681 if unavailable_message:
1682 reason = unavailable_message
1683 raise ExtractorError(
1684 'YouTube said: %s' % reason,
1685 expected=True, video_id=video_id)
1686 else:
1687 raise ExtractorError(
1688 '"token" parameter not in video info for unknown reason',
1689 video_id=video_id)
1690
1691 if video_info.get('license_info'):
1692 raise ExtractorError('This video is DRM protected.', expected=True)
1693
1694 video_details = try_get(
1695 player_response, lambda x: x['videoDetails'], dict) or {}
1696
1697 # title
1698 if 'title' in video_info:
1699 video_title = video_info['title'][0]
1700 elif 'title' in player_response:
1701 video_title = video_details['title']
1702 else:
1703 self._downloader.report_warning('Unable to extract video title')
1704 video_title = '_'
1705
1706 # description
1707 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1708 if video_description:
1709
1710 def replace_url(m):
1711 redir_url = compat_urlparse.urljoin(url, m.group(1))
1712 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1713 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1714 qs = compat_parse_qs(parsed_redir_url.query)
1715 q = qs.get('q')
1716 if q and q[0]:
1717 return q[0]
1718 return redir_url
1719
1720 description_original = video_description = re.sub(r'''(?x)
1721 <a\s+
1722 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1723 (?:title|href)="([^"]+)"\s+
1724 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1725 class="[^"]*"[^>]*>
1726 [^<]+\.{3}\s*
1727 </a>
1728 ''', replace_url, video_description)
1729 video_description = clean_html(video_description)
1730 else:
1731 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1732 if fd_mobj:
1733 video_description = unescapeHTML(fd_mobj.group(1))
1734 else:
1735 video_description = ''
1736
1737 if not smuggled_data.get('force_singlefeed', False):
1738 if not self._downloader.params.get('noplaylist'):
1739 multifeed_metadata_list = try_get(
1740 player_response,
1741 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1742 compat_str) or try_get(
1743 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1744 if multifeed_metadata_list:
1745 entries = []
1746 feed_ids = []
1747 for feed in multifeed_metadata_list.split(','):
1748 # Unquote should take place before split on comma (,) since textual
1749 # fields may contain comma as well (see
1750 # https://github.com/rg3/youtube-dl/issues/8536)
1751 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1752 entries.append({
1753 '_type': 'url_transparent',
1754 'ie_key': 'Youtube',
1755 'url': smuggle_url(
1756 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1757 {'force_singlefeed': True}),
1758 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1759 })
1760 feed_ids.append(feed_data['id'][0])
1761 self.to_screen(
1762 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1763 % (', '.join(feed_ids), video_id))
1764 return self.playlist_result(entries, video_id, video_title, video_description)
1765 else:
1766 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1767
1768 if view_count is None:
1769 view_count = extract_view_count(video_info)
1770 if view_count is None and video_details:
1771 view_count = int_or_none(video_details.get('viewCount'))
1772
1773 # Check for "rental" videos
1774 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1775 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1776
1777 def _extract_filesize(media_url):
1778 return int_or_none(self._search_regex(
1779 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1780
1781 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1782 self.report_rtmp_download()
1783 formats = [{
1784 'format_id': '_rtmp',
1785 'protocol': 'rtmp',
1786 'url': video_info['conn'][0],
1787 'player_url': player_url,
1788 }]
1789 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):
1790 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1791 if 'rtmpe%3Dyes' in encoded_url_map:
1792 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1793 formats_spec = {}
1794 fmt_list = video_info.get('fmt_list', [''])[0]
1795 if fmt_list:
1796 for fmt in fmt_list.split(','):
1797 spec = fmt.split('/')
1798 if len(spec) > 1:
1799 width_height = spec[1].split('x')
1800 if len(width_height) == 2:
1801 formats_spec[spec[0]] = {
1802 'resolution': spec[1],
1803 'width': int_or_none(width_height[0]),
1804 'height': int_or_none(width_height[1]),
1805 }
1806 q = qualities(['small', 'medium', 'hd720'])
1807 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1808 if streaming_formats:
1809 for fmt in streaming_formats:
1810 itag = str_or_none(fmt.get('itag'))
1811 if not itag:
1812 continue
1813 quality = fmt.get('quality')
1814 quality_label = fmt.get('qualityLabel') or quality
1815 formats_spec[itag] = {
1816 'asr': int_or_none(fmt.get('audioSampleRate')),
1817 'filesize': int_or_none(fmt.get('contentLength')),
1818 'format_note': quality_label,
1819 'fps': int_or_none(fmt.get('fps')),
1820 'height': int_or_none(fmt.get('height')),
1821 'quality': q(quality),
1822 # bitrate for itag 43 is always 2147483647
1823 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1824 'width': int_or_none(fmt.get('width')),
1825 }
1826 formats = []
1827 for url_data_str in encoded_url_map.split(','):
1828 url_data = compat_parse_qs(url_data_str)
1829 if 'itag' not in url_data or 'url' not in url_data:
1830 continue
1831 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1832 # Unsupported FORMAT_STREAM_TYPE_OTF
1833 if stream_type == 3:
1834 continue
1835 format_id = url_data['itag'][0]
1836 url = url_data['url'][0]
1837
1838 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1839 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1840 jsplayer_url_json = self._search_regex(
1841 ASSETS_RE,
1842 embed_webpage if age_gate else video_webpage,
1843 'JS player URL (1)', default=None)
1844 if not jsplayer_url_json and not age_gate:
1845 # We need the embed website after all
1846 if embed_webpage is None:
1847 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1848 embed_webpage = self._download_webpage(
1849 embed_url, video_id, 'Downloading embed webpage')
1850 jsplayer_url_json = self._search_regex(
1851 ASSETS_RE, embed_webpage, 'JS player URL')
1852
1853 player_url = json.loads(jsplayer_url_json)
1854 if player_url is None:
1855 player_url_json = self._search_regex(
1856 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1857 video_webpage, 'age gate player URL')
1858 player_url = json.loads(player_url_json)
1859
1860 if 'sig' in url_data:
1861 url += '&signature=' + url_data['sig'][0]
1862 elif 's' in url_data:
1863 encrypted_sig = url_data['s'][0]
1864
1865 if self._downloader.params.get('verbose'):
1866 if player_url is None:
1867 player_version = 'unknown'
1868 player_desc = 'unknown'
1869 else:
1870 if player_url.endswith('swf'):
1871 player_version = self._search_regex(
1872 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1873 'flash player', fatal=False)
1874 player_desc = 'flash player %s' % player_version
1875 else:
1876 player_version = self._search_regex(
1877 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1878 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1879 player_url,
1880 'html5 player', fatal=False)
1881 player_desc = 'html5 player %s' % player_version
1882
1883 parts_sizes = self._signature_cache_id(encrypted_sig)
1884 self.to_screen('{%s} signature length %s, %s' %
1885 (format_id, parts_sizes, player_desc))
1886
1887 signature = self._decrypt_signature(
1888 encrypted_sig, video_id, player_url, age_gate)
1889 url += '&signature=' + signature
1890 if 'ratebypass' not in url:
1891 url += '&ratebypass=yes'
1892
1893 dct = {
1894 'format_id': format_id,
1895 'url': url,
1896 'player_url': player_url,
1897 }
1898 if format_id in self._formats:
1899 dct.update(self._formats[format_id])
1900 if format_id in formats_spec:
1901 dct.update(formats_spec[format_id])
1902
1903 # Some itags are not included in DASH manifest thus corresponding formats will
1904 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1905 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1906 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1907 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1908
1909 filesize = int_or_none(url_data.get(
1910 'clen', [None])[0]) or _extract_filesize(url)
1911
1912 quality = url_data.get('quality', [None])[0]
1913
1914 more_fields = {
1915 'filesize': filesize,
1916 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1917 'width': width,
1918 'height': height,
1919 'fps': int_or_none(url_data.get('fps', [None])[0]),
1920 'format_note': url_data.get('quality_label', [None])[0] or quality,
1921 'quality': q(quality),
1922 }
1923 for key, value in more_fields.items():
1924 if value:
1925 dct[key] = value
1926 type_ = url_data.get('type', [None])[0]
1927 if type_:
1928 type_split = type_.split(';')
1929 kind_ext = type_split[0].split('/')
1930 if len(kind_ext) == 2:
1931 kind, _ = kind_ext
1932 dct['ext'] = mimetype2ext(type_split[0])
1933 if kind in ('audio', 'video'):
1934 codecs = None
1935 for mobj in re.finditer(
1936 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1937 if mobj.group('key') == 'codecs':
1938 codecs = mobj.group('val')
1939 break
1940 if codecs:
1941 dct.update(parse_codecs(codecs))
1942 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1943 dct['downloader_options'] = {
1944 # Youtube throttles chunks >~10M
1945 'http_chunk_size': 10485760,
1946 }
1947 formats.append(dct)
1948 else:
1949 manifest_url = (
1950 url_or_none(try_get(
1951 player_response,
1952 lambda x: x['streamingData']['hlsManifestUrl'],
1953 compat_str)) or
1954 url_or_none(try_get(
1955 video_info, lambda x: x['hlsvp'][0], compat_str)))
1956 if manifest_url:
1957 formats = []
1958 m3u8_formats = self._extract_m3u8_formats(
1959 manifest_url, video_id, 'mp4', fatal=False)
1960 for a_format in m3u8_formats:
1961 itag = self._search_regex(
1962 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1963 if itag:
1964 a_format['format_id'] = itag
1965 if itag in self._formats:
1966 dct = self._formats[itag].copy()
1967 dct.update(a_format)
1968 a_format = dct
1969 a_format['player_url'] = player_url
1970 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1971 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1972 formats.append(a_format)
1973 else:
1974 error_message = clean_html(video_info.get('reason', [None])[0])
1975 if not error_message:
1976 error_message = extract_unavailable_message()
1977 if error_message:
1978 raise ExtractorError(error_message, expected=True)
1979 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
1980
1981 # uploader
1982 video_uploader = try_get(
1983 video_info, lambda x: x['author'][0],
1984 compat_str) or str_or_none(video_details.get('author'))
1985 if video_uploader:
1986 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1987 else:
1988 self._downloader.report_warning('unable to extract uploader name')
1989
1990 # uploader_id
1991 video_uploader_id = None
1992 video_uploader_url = None
1993 mobj = re.search(
1994 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1995 video_webpage)
1996 if mobj is not None:
1997 video_uploader_id = mobj.group('uploader_id')
1998 video_uploader_url = mobj.group('uploader_url')
1999 else:
2000 self._downloader.report_warning('unable to extract uploader nickname')
2001
2002 channel_id = self._html_search_meta(
2003 'channelId', video_webpage, 'channel id')
2004 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2005
2006 # thumbnail image
2007 # We try first to get a high quality image:
2008 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2009 video_webpage, re.DOTALL)
2010 if m_thumb is not None:
2011 video_thumbnail = m_thumb.group(1)
2012 elif 'thumbnail_url' not in video_info:
2013 self._downloader.report_warning('unable to extract video thumbnail')
2014 video_thumbnail = None
2015 else: # don't panic if we can't find it
2016 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2017
2018 # upload date
2019 upload_date = self._html_search_meta(
2020 'datePublished', video_webpage, 'upload date', default=None)
2021 if not upload_date:
2022 upload_date = self._search_regex(
2023 [r'(?s)id="eow-date.*?>(.*?)</span>',
2024 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2025 video_webpage, 'upload date', default=None)
2026 upload_date = unified_strdate(upload_date)
2027
2028 video_license = self._html_search_regex(
2029 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2030 video_webpage, 'license', default=None)
2031
2032 m_music = re.search(
2033 r'''(?x)
2034 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2035 <ul[^>]*>\s*
2036 <li>(?P<title>.+?)
2037 by (?P<creator>.+?)
2038 (?:
2039 \(.+?\)|
2040 <a[^>]*
2041 (?:
2042 \bhref=["\']/red[^>]*>| # drop possible
2043 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2044 )
2045 .*?
2046 )?</li
2047 ''',
2048 video_webpage)
2049 if m_music:
2050 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2051 video_creator = clean_html(m_music.group('creator'))
2052 else:
2053 video_alt_title = video_creator = None
2054
2055 def extract_meta(field):
2056 return self._html_search_regex(
2057 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2058 video_webpage, field, default=None)
2059
2060 track = extract_meta('Song')
2061 artist = extract_meta('Artist')
2062
2063 m_episode = re.search(
2064 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2065 video_webpage)
2066 if m_episode:
2067 series = unescapeHTML(m_episode.group('series'))
2068 season_number = int(m_episode.group('season'))
2069 episode_number = int(m_episode.group('episode'))
2070 else:
2071 series = season_number = episode_number = None
2072
2073 m_cat_container = self._search_regex(
2074 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2075 video_webpage, 'categories', default=None)
2076 if m_cat_container:
2077 category = self._html_search_regex(
2078 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2079 default=None)
2080 video_categories = None if category is None else [category]
2081 else:
2082 video_categories = None
2083
2084 video_tags = [
2085 unescapeHTML(m.group('content'))
2086 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2087
2088 def _extract_count(count_name):
2089 return str_to_int(self._search_regex(
2090 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2091 % re.escape(count_name),
2092 video_webpage, count_name, default=None))
2093
2094 like_count = _extract_count('like')
2095 dislike_count = _extract_count('dislike')
2096
2097 if view_count is None:
2098 view_count = str_to_int(self._search_regex(
2099 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2100 'view count', default=None))
2101
2102 # subtitles
2103 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2104 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2105
2106 video_duration = try_get(
2107 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2108 if not video_duration:
2109 video_duration = int_or_none(video_details.get('lengthSeconds'))
2110 if not video_duration:
2111 video_duration = parse_duration(self._html_search_meta(
2112 'duration', video_webpage, 'video duration'))
2113
2114 # annotations
2115 video_annotations = None
2116 if self._downloader.params.get('writeannotations', False):
2117 video_annotations = self._extract_annotations(video_id)
2118
2119 chapters = self._extract_chapters(description_original, video_duration)
2120
2121 # Look for the DASH manifest
2122 if self._downloader.params.get('youtube_include_dash_manifest', True):
2123 dash_mpd_fatal = True
2124 for mpd_url in dash_mpds:
2125 dash_formats = {}
2126 try:
2127 def decrypt_sig(mobj):
2128 s = mobj.group(1)
2129 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2130 return '/signature/%s' % dec_s
2131
2132 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2133
2134 for df in self._extract_mpd_formats(
2135 mpd_url, video_id, fatal=dash_mpd_fatal,
2136 formats_dict=self._formats):
2137 if not df.get('filesize'):
2138 df['filesize'] = _extract_filesize(df['url'])
2139 # Do not overwrite DASH format found in some previous DASH manifest
2140 if df['format_id'] not in dash_formats:
2141 dash_formats[df['format_id']] = df
2142 # Additional DASH manifests may end up in HTTP Error 403 therefore
2143 # allow them to fail without bug report message if we already have
2144 # some DASH manifest succeeded. This is temporary workaround to reduce
2145 # burst of bug reports until we figure out the reason and whether it
2146 # can be fixed at all.
2147 dash_mpd_fatal = False
2148 except (ExtractorError, KeyError) as e:
2149 self.report_warning(
2150 'Skipping DASH manifest: %r' % e, video_id)
2151 if dash_formats:
2152 # Remove the formats we found through non-DASH, they
2153 # contain less info and it can be wrong, because we use
2154 # fixed values (for example the resolution). See
2155 # https://github.com/rg3/youtube-dl/issues/5774 for an
2156 # example.
2157 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2158 formats.extend(dash_formats.values())
2159
2160 # Check for malformed aspect ratio
2161 stretched_m = re.search(
2162 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2163 video_webpage)
2164 if stretched_m:
2165 w = float(stretched_m.group('w'))
2166 h = float(stretched_m.group('h'))
2167 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2168 # We will only process correct ratios.
2169 if w > 0 and h > 0:
2170 ratio = w / h
2171 for f in formats:
2172 if f.get('vcodec') != 'none':
2173 f['stretched_ratio'] = ratio
2174
2175 self._sort_formats(formats)
2176
2177 self.mark_watched(video_id, video_info, player_response)
2178
2179 return {
2180 'id': video_id,
2181 'uploader': video_uploader,
2182 'uploader_id': video_uploader_id,
2183 'uploader_url': video_uploader_url,
2184 'channel_id': channel_id,
2185 'channel_url': channel_url,
2186 'upload_date': upload_date,
2187 'license': video_license,
2188 'creator': video_creator or artist,
2189 'title': video_title,
2190 'alt_title': video_alt_title or track,
2191 'thumbnail': video_thumbnail,
2192 'description': video_description,
2193 'categories': video_categories,
2194 'tags': video_tags,
2195 'subtitles': video_subtitles,
2196 'automatic_captions': automatic_captions,
2197 'duration': video_duration,
2198 'age_limit': 18 if age_gate else 0,
2199 'annotations': video_annotations,
2200 'chapters': chapters,
2201 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2202 'view_count': view_count,
2203 'like_count': like_count,
2204 'dislike_count': dislike_count,
2205 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2206 'formats': formats,
2207 'is_live': is_live,
2208 'start_time': start_time,
2209 'end_time': end_time,
2210 'series': series,
2211 'season_number': season_number,
2212 'episode_number': episode_number,
2213 'track': track,
2214 'artist': artist,
2215 }
2216
2217
2218 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2219 IE_DESC = 'YouTube.com playlists'
2220 _VALID_URL = r"""(?x)(?:
2221 (?:https?://)?
2222 (?:\w+\.)?
2223 (?:
2224 (?:
2225 youtube\.com|
2226 invidio\.us
2227 )
2228 /
2229 (?:
2230 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2231 \? (?:.*?[&;])*? (?:p|a|list)=
2232 | p/
2233 )|
2234 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2235 )
2236 (
2237 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2238 # Top tracks, they can also include dots
2239 |(?:MC)[\w\.]*
2240 )
2241 .*
2242 |
2243 (%(playlist_id)s)
2244 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2245 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2246 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2247 IE_NAME = 'youtube:playlist'
2248 _TESTS = [{
2249 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2250 'info_dict': {
2251 'title': 'ytdl test PL',
2252 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2253 },
2254 'playlist_count': 3,
2255 }, {
2256 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2257 'info_dict': {
2258 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2259 'title': 'YDL_Empty_List',
2260 },
2261 'playlist_count': 0,
2262 'skip': 'This playlist is private',
2263 }, {
2264 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2265 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2266 'info_dict': {
2267 'title': '29C3: Not my department',
2268 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2269 },
2270 'playlist_count': 95,
2271 }, {
2272 'note': 'issue #673',
2273 'url': 'PLBB231211A4F62143',
2274 'info_dict': {
2275 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2276 'id': 'PLBB231211A4F62143',
2277 },
2278 'playlist_mincount': 26,
2279 }, {
2280 'note': 'Large playlist',
2281 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2282 'info_dict': {
2283 'title': 'Uploads from Cauchemar',
2284 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2285 },
2286 'playlist_mincount': 799,
2287 }, {
2288 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2289 'info_dict': {
2290 'title': 'YDL_safe_search',
2291 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2292 },
2293 'playlist_count': 2,
2294 'skip': 'This playlist is private',
2295 }, {
2296 'note': 'embedded',
2297 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2298 'playlist_count': 4,
2299 'info_dict': {
2300 'title': 'JODA15',
2301 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2302 }
2303 }, {
2304 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2305 'playlist_mincount': 485,
2306 'info_dict': {
2307 'title': '2017 čÆčŖžęœ€ę–°å–®ę›² (2/24ꛓꖰ)',
2308 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2309 }
2310 }, {
2311 'note': 'Embedded SWF player',
2312 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2313 'playlist_count': 4,
2314 'info_dict': {
2315 'title': 'JODA7',
2316 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2317 }
2318 }, {
2319 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2320 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2321 'info_dict': {
2322 'title': 'Uploads from Interstellar Movie',
2323 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2324 },
2325 'playlist_mincount': 21,
2326 }, {
2327 # Playlist URL that does not actually serve a playlist
2328 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2329 'info_dict': {
2330 'id': 'FqZTN594JQw',
2331 'ext': 'webm',
2332 'title': "Smiley's People 01 detective, Adventure Series, Action",
2333 'uploader': 'STREEM',
2334 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2335 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2336 'upload_date': '20150526',
2337 'license': 'Standard YouTube License',
2338 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2339 'categories': ['People & Blogs'],
2340 'tags': list,
2341 'view_count': int,
2342 'like_count': int,
2343 'dislike_count': int,
2344 },
2345 'params': {
2346 'skip_download': True,
2347 },
2348 'add_ie': [YoutubeIE.ie_key()],
2349 }, {
2350 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2351 'info_dict': {
2352 'id': 'yeWKywCrFtk',
2353 'ext': 'mp4',
2354 'title': 'Small Scale Baler and Braiding Rugs',
2355 'uploader': 'Backus-Page House Museum',
2356 'uploader_id': 'backuspagemuseum',
2357 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2358 'upload_date': '20161008',
2359 'license': 'Standard YouTube License',
2360 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2361 'categories': ['Nonprofits & Activism'],
2362 'tags': list,
2363 'like_count': int,
2364 'dislike_count': int,
2365 },
2366 'params': {
2367 'noplaylist': True,
2368 'skip_download': True,
2369 },
2370 }, {
2371 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2372 'only_matching': True,
2373 }, {
2374 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2375 'only_matching': True,
2376 }, {
2377 # music album playlist
2378 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2379 'only_matching': True,
2380 }, {
2381 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2382 'only_matching': True,
2383 }]
2384
2385 def _real_initialize(self):
2386 self._login()
2387
2388 def _extract_mix(self, playlist_id):
2389 # The mixes are generated from a single video
2390 # the id of the playlist is just 'RD' + video_id
2391 ids = []
2392 last_id = playlist_id[-11:]
2393 for n in itertools.count(1):
2394 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2395 webpage = self._download_webpage(
2396 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2397 new_ids = orderedSet(re.findall(
2398 r'''(?xs)data-video-username=".*?".*?
2399 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2400 webpage))
2401 # Fetch new pages until all the videos are repeated, it seems that
2402 # there are always 51 unique videos.
2403 new_ids = [_id for _id in new_ids if _id not in ids]
2404 if not new_ids:
2405 break
2406 ids.extend(new_ids)
2407 last_id = ids[-1]
2408
2409 url_results = self._ids_to_results(ids)
2410
2411 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2412 title_span = (
2413 search_title('playlist-title') or
2414 search_title('title long-title') or
2415 search_title('title'))
2416 title = clean_html(title_span)
2417
2418 return self.playlist_result(url_results, playlist_id, title)
2419
2420 def _extract_playlist(self, playlist_id):
2421 url = self._TEMPLATE_URL % playlist_id
2422 page = self._download_webpage(url, playlist_id)
2423
2424 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2425 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2426 match = match.strip()
2427 # Check if the playlist exists or is private
2428 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2429 if mobj:
2430 reason = mobj.group('reason')
2431 message = 'This playlist %s' % reason
2432 if 'private' in reason:
2433 message += ', use --username or --netrc to access it'
2434 message += '.'
2435 raise ExtractorError(message, expected=True)
2436 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2437 raise ExtractorError(
2438 'Invalid parameters. Maybe URL is incorrect.',
2439 expected=True)
2440 elif re.match(r'[^<]*Choose your language[^<]*', match):
2441 continue
2442 else:
2443 self.report_warning('Youtube gives an alert message: ' + match)
2444
2445 playlist_title = self._html_search_regex(
2446 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2447 page, 'title', default=None)
2448
2449 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2450 uploader = self._search_regex(
2451 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2452 page, 'uploader', default=None)
2453 mobj = re.search(
2454 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2455 page)
2456 if mobj:
2457 uploader_id = mobj.group('uploader_id')
2458 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2459 else:
2460 uploader_id = uploader_url = None
2461
2462 has_videos = True
2463
2464 if not playlist_title:
2465 try:
2466 # Some playlist URLs don't actually serve a playlist (e.g.
2467 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2468 next(self._entries(page, playlist_id))
2469 except StopIteration:
2470 has_videos = False
2471
2472 playlist = self.playlist_result(
2473 self._entries(page, playlist_id), playlist_id, playlist_title)
2474 playlist.update({
2475 'uploader': uploader,
2476 'uploader_id': uploader_id,
2477 'uploader_url': uploader_url,
2478 })
2479
2480 return has_videos, playlist
2481
2482 def _check_download_just_video(self, url, playlist_id):
2483 # Check if it's a video-specific URL
2484 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2485 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2486 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2487 'video id', default=None)
2488 if video_id:
2489 if self._downloader.params.get('noplaylist'):
2490 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2491 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2492 else:
2493 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2494 return video_id, None
2495 return None, None
2496
2497 def _real_extract(self, url):
2498 # Extract playlist id
2499 mobj = re.match(self._VALID_URL, url)
2500 if mobj is None:
2501 raise ExtractorError('Invalid URL: %s' % url)
2502 playlist_id = mobj.group(1) or mobj.group(2)
2503
2504 video_id, video = self._check_download_just_video(url, playlist_id)
2505 if video:
2506 return video
2507
2508 if playlist_id.startswith(('RD', 'UL', 'PU')):
2509 # Mixes require a custom extraction process
2510 return self._extract_mix(playlist_id)
2511
2512 has_videos, playlist = self._extract_playlist(playlist_id)
2513 if has_videos or not video_id:
2514 return playlist
2515
2516 # Some playlist URLs don't actually serve a playlist (see
2517 # https://github.com/rg3/youtube-dl/issues/10537).
2518 # Fallback to plain video extraction if there is a video id
2519 # along with playlist id.
2520 return self.url_result(video_id, 'Youtube', video_id=video_id)
2521
2522
2523 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2524 IE_DESC = 'YouTube.com channels'
2525 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2526 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2527 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2528 IE_NAME = 'youtube:channel'
2529 _TESTS = [{
2530 'note': 'paginated channel',
2531 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2532 'playlist_mincount': 91,
2533 'info_dict': {
2534 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2535 'title': 'Uploads from lex will',
2536 }
2537 }, {
2538 'note': 'Age restricted channel',
2539 # from https://www.youtube.com/user/DeusExOfficial
2540 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2541 'playlist_mincount': 64,
2542 'info_dict': {
2543 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2544 'title': 'Uploads from Deus Ex',
2545 },
2546 }, {
2547 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2548 'only_matching': True,
2549 }]
2550
2551 @classmethod
2552 def suitable(cls, url):
2553 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2554 else super(YoutubeChannelIE, cls).suitable(url))
2555
2556 def _build_template_url(self, url, channel_id):
2557 return self._TEMPLATE_URL % channel_id
2558
2559 def _real_extract(self, url):
2560 channel_id = self._match_id(url)
2561
2562 url = self._build_template_url(url, channel_id)
2563
2564 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2565 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2566 # otherwise fallback on channel by page extraction
2567 channel_page = self._download_webpage(
2568 url + '?view=57', channel_id,
2569 'Downloading channel page', fatal=False)
2570 if channel_page is False:
2571 channel_playlist_id = False
2572 else:
2573 channel_playlist_id = self._html_search_meta(
2574 'channelId', channel_page, 'channel id', default=None)
2575 if not channel_playlist_id:
2576 channel_url = self._html_search_meta(
2577 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2578 channel_page, 'channel url', default=None)
2579 if channel_url:
2580 channel_playlist_id = self._search_regex(
2581 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2582 channel_url, 'channel id', default=None)
2583 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2584 playlist_id = 'UU' + channel_playlist_id[2:]
2585 return self.url_result(
2586 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2587
2588 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2589 autogenerated = re.search(r'''(?x)
2590 class="[^"]*?(?:
2591 channel-header-autogenerated-label|
2592 yt-channel-title-autogenerated
2593 )[^"]*"''', channel_page) is not None
2594
2595 if autogenerated:
2596 # The videos are contained in a single page
2597 # the ajax pages can't be used, they are empty
2598 entries = [
2599 self.url_result(
2600 video_id, 'Youtube', video_id=video_id,
2601 video_title=video_title)
2602 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2603 return self.playlist_result(entries, channel_id)
2604
2605 try:
2606 next(self._entries(channel_page, channel_id))
2607 except StopIteration:
2608 alert_message = self._html_search_regex(
2609 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2610 channel_page, 'alert', default=None, group='alert')
2611 if alert_message:
2612 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2613
2614 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2615
2616
2617 class YoutubeUserIE(YoutubeChannelIE):
2618 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2619 _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_-]+)'
2620 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2621 IE_NAME = 'youtube:user'
2622
2623 _TESTS = [{
2624 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2625 'playlist_mincount': 320,
2626 'info_dict': {
2627 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2628 'title': 'Uploads from The Linux Foundation',
2629 }
2630 }, {
2631 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2632 # but not https://www.youtube.com/user/12minuteathlete/videos
2633 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2634 'playlist_mincount': 249,
2635 'info_dict': {
2636 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2637 'title': 'Uploads from 12 Minute Athlete',
2638 }
2639 }, {
2640 'url': 'ytuser:phihag',
2641 'only_matching': True,
2642 }, {
2643 'url': 'https://www.youtube.com/c/gametrailers',
2644 'only_matching': True,
2645 }, {
2646 'url': 'https://www.youtube.com/gametrailers',
2647 'only_matching': True,
2648 }, {
2649 # This channel is not available, geo restricted to JP
2650 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2651 'only_matching': True,
2652 }]
2653
2654 @classmethod
2655 def suitable(cls, url):
2656 # Don't return True if the url can be extracted with other youtube
2657 # extractor, the regex would is too permissive and it would match.
2658 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2659 if any(ie.suitable(url) for ie in other_yt_ies):
2660 return False
2661 else:
2662 return super(YoutubeUserIE, cls).suitable(url)
2663
2664 def _build_template_url(self, url, channel_id):
2665 mobj = re.match(self._VALID_URL, url)
2666 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2667
2668
2669 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2670 IE_DESC = 'YouTube.com live streams'
2671 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2672 IE_NAME = 'youtube:live'
2673
2674 _TESTS = [{
2675 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2676 'info_dict': {
2677 'id': 'a48o2S1cPoo',
2678 'ext': 'mp4',
2679 'title': 'The Young Turks - Live Main Show',
2680 'uploader': 'The Young Turks',
2681 'uploader_id': 'TheYoungTurks',
2682 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2683 'upload_date': '20150715',
2684 'license': 'Standard YouTube License',
2685 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2686 'categories': ['News & Politics'],
2687 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2688 'like_count': int,
2689 'dislike_count': int,
2690 },
2691 'params': {
2692 'skip_download': True,
2693 },
2694 }, {
2695 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2696 'only_matching': True,
2697 }, {
2698 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2699 'only_matching': True,
2700 }, {
2701 'url': 'https://www.youtube.com/TheYoungTurks/live',
2702 'only_matching': True,
2703 }]
2704
2705 def _real_extract(self, url):
2706 mobj = re.match(self._VALID_URL, url)
2707 channel_id = mobj.group('id')
2708 base_url = mobj.group('base_url')
2709 webpage = self._download_webpage(url, channel_id, fatal=False)
2710 if webpage:
2711 page_type = self._og_search_property(
2712 'type', webpage, 'page type', default='')
2713 video_id = self._html_search_meta(
2714 'videoId', webpage, 'video id', default=None)
2715 if page_type.startswith('video') and video_id and re.match(
2716 r'^[0-9A-Za-z_-]{11}$', video_id):
2717 return self.url_result(video_id, YoutubeIE.ie_key())
2718 return self.url_result(base_url)
2719
2720
2721 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2722 IE_DESC = 'YouTube.com user/channel playlists'
2723 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2724 IE_NAME = 'youtube:playlists'
2725
2726 _TESTS = [{
2727 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2728 'playlist_mincount': 4,
2729 'info_dict': {
2730 'id': 'ThirstForScience',
2731 'title': 'Thirst for Science',
2732 },
2733 }, {
2734 # with "Load more" button
2735 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2736 'playlist_mincount': 70,
2737 'info_dict': {
2738 'id': 'igorkle1',
2739 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
2740 },
2741 }, {
2742 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2743 'playlist_mincount': 17,
2744 'info_dict': {
2745 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2746 'title': 'Chem Player',
2747 },
2748 }]
2749
2750
2751 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2752 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2753
2754
2755 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2756 IE_DESC = 'YouTube.com searches'
2757 # there doesn't appear to be a real limit, for example if you search for
2758 # 'python' you get more than 8.000.000 results
2759 _MAX_RESULTS = float('inf')
2760 IE_NAME = 'youtube:search'
2761 _SEARCH_KEY = 'ytsearch'
2762 _EXTRA_QUERY_ARGS = {}
2763 _TESTS = []
2764
2765 def _get_n_results(self, query, n):
2766 """Get a specified number of results for a query"""
2767
2768 videos = []
2769 limit = n
2770
2771 url_query = {
2772 'search_query': query.encode('utf-8'),
2773 }
2774 url_query.update(self._EXTRA_QUERY_ARGS)
2775 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2776
2777 for pagenum in itertools.count(1):
2778 data = self._download_json(
2779 result_url, video_id='query "%s"' % query,
2780 note='Downloading page %s' % pagenum,
2781 errnote='Unable to download API page',
2782 query={'spf': 'navigate'})
2783 html_content = data[1]['body']['content']
2784
2785 if 'class="search-message' in html_content:
2786 raise ExtractorError(
2787 '[youtube] No video results', expected=True)
2788
2789 new_videos = list(self._process_page(html_content))
2790 videos += new_videos
2791 if not new_videos or len(videos) > limit:
2792 break
2793 next_link = self._html_search_regex(
2794 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2795 html_content, 'next link', default=None)
2796 if next_link is None:
2797 break
2798 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2799
2800 if len(videos) > n:
2801 videos = videos[:n]
2802 return self.playlist_result(videos, query)
2803
2804
2805 class YoutubeSearchDateIE(YoutubeSearchIE):
2806 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2807 _SEARCH_KEY = 'ytsearchdate'
2808 IE_DESC = 'YouTube.com searches, newest videos first'
2809 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2810
2811
2812 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2813 IE_DESC = 'YouTube.com search URLs'
2814 IE_NAME = 'youtube:search_url'
2815 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2816 _TESTS = [{
2817 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2818 'playlist_mincount': 5,
2819 'info_dict': {
2820 'title': 'youtube-dl test video',
2821 }
2822 }, {
2823 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2824 'only_matching': True,
2825 }]
2826
2827 def _real_extract(self, url):
2828 mobj = re.match(self._VALID_URL, url)
2829 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2830 webpage = self._download_webpage(url, query)
2831 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2832
2833
2834 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2835 IE_DESC = 'YouTube.com (multi-season) shows'
2836 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2837 IE_NAME = 'youtube:show'
2838 _TESTS = [{
2839 'url': 'https://www.youtube.com/show/airdisasters',
2840 'playlist_mincount': 5,
2841 'info_dict': {
2842 'id': 'airdisasters',
2843 'title': 'Air Disasters',
2844 }
2845 }]
2846
2847 def _real_extract(self, url):
2848 playlist_id = self._match_id(url)
2849 return super(YoutubeShowIE, self)._real_extract(
2850 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2851
2852
2853 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2854 """
2855 Base class for feed extractors
2856 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2857 """
2858 _LOGIN_REQUIRED = True
2859
2860 @property
2861 def IE_NAME(self):
2862 return 'youtube:%s' % self._FEED_NAME
2863
2864 def _real_initialize(self):
2865 self._login()
2866
2867 def _entries(self, page):
2868 # The extraction process is the same as for playlists, but the regex
2869 # for the video ids doesn't contain an index
2870 ids = []
2871 more_widget_html = content_html = page
2872 for page_num in itertools.count(1):
2873 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2874
2875 # 'recommended' feed has infinite 'load more' and each new portion spins
2876 # the same videos in (sometimes) slightly different order, so we'll check
2877 # for unicity and break when portion has no new videos
2878 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2879 if not new_ids:
2880 break
2881
2882 ids.extend(new_ids)
2883
2884 for entry in self._ids_to_results(new_ids):
2885 yield entry
2886
2887 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2888 if not mobj:
2889 break
2890
2891 more = self._download_json(
2892 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2893 'Downloading page #%s' % page_num,
2894 transform_source=uppercase_escape)
2895 content_html = more['content_html']
2896 more_widget_html = more['load_more_widget_html']
2897
2898 def _real_extract(self, url):
2899 page = self._download_webpage(
2900 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2901 self._PLAYLIST_TITLE)
2902 return self.playlist_result(
2903 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2904
2905
2906 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2907 IE_NAME = 'youtube:watchlater'
2908 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2909 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2910
2911 _TESTS = [{
2912 'url': 'https://www.youtube.com/playlist?list=WL',
2913 'only_matching': True,
2914 }, {
2915 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2916 'only_matching': True,
2917 }]
2918
2919 def _real_extract(self, url):
2920 _, video = self._check_download_just_video(url, 'WL')
2921 if video:
2922 return video
2923 _, playlist = self._extract_playlist('WL')
2924 return playlist
2925
2926
2927 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2928 IE_NAME = 'youtube:favorites'
2929 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2930 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2931 _LOGIN_REQUIRED = True
2932
2933 def _real_extract(self, url):
2934 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2935 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2936 return self.url_result(playlist_id, 'YoutubePlaylist')
2937
2938
2939 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2940 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2941 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2942 _FEED_NAME = 'recommended'
2943 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2944
2945
2946 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2947 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2948 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2949 _FEED_NAME = 'subscriptions'
2950 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2951
2952
2953 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2954 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2955 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2956 _FEED_NAME = 'history'
2957 _PLAYLIST_TITLE = 'Youtube History'
2958
2959
2960 class YoutubeTruncatedURLIE(InfoExtractor):
2961 IE_NAME = 'youtube:truncated_url'
2962 IE_DESC = False # Do not list
2963 _VALID_URL = r'''(?x)
2964 (?:https?://)?
2965 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2966 (?:watch\?(?:
2967 feature=[a-z_]+|
2968 annotation_id=annotation_[^&]+|
2969 x-yt-cl=[0-9]+|
2970 hl=[^&]*|
2971 t=[0-9]+
2972 )?
2973 |
2974 attribution_link\?a=[^&]+
2975 )
2976 $
2977 '''
2978
2979 _TESTS = [{
2980 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2981 'only_matching': True,
2982 }, {
2983 'url': 'https://www.youtube.com/watch?',
2984 'only_matching': True,
2985 }, {
2986 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2987 'only_matching': True,
2988 }, {
2989 'url': 'https://www.youtube.com/watch?feature=foo',
2990 'only_matching': True,
2991 }, {
2992 'url': 'https://www.youtube.com/watch?hl=en-GB',
2993 'only_matching': True,
2994 }, {
2995 'url': 'https://www.youtube.com/watch?t=2372',
2996 'only_matching': True,
2997 }]
2998
2999 def _real_extract(self, url):
3000 raise ExtractorError(
3001 'Did you forget to quote the URL? Remember that & is a meta '
3002 'character in most shells, so you want to put the URL in quotes, '
3003 'like youtube-dl '
3004 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3005 ' or simply youtube-dl BaW_jenozKc .',
3006 expected=True)
3007
3008
3009 class YoutubeTruncatedIDIE(InfoExtractor):
3010 IE_NAME = 'youtube:truncated_id'
3011 IE_DESC = False # Do not list
3012 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3013
3014 _TESTS = [{
3015 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3016 'only_matching': True,
3017 }]
3018
3019 def _real_extract(self, url):
3020 video_id = self._match_id(url)
3021 raise ExtractorError(
3022 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3023 expected=True)