]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
Fix W504 and disable W503 (closes #20863)
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_HTTPError,
20 compat_kwargs,
21 compat_parse_qs,
22 compat_urllib_parse_unquote,
23 compat_urllib_parse_unquote_plus,
24 compat_urllib_parse_urlencode,
25 compat_urllib_parse_urlparse,
26 compat_urlparse,
27 compat_str,
28 )
29 from ..utils import (
30 clean_html,
31 dict_get,
32 error_to_compat_str,
33 ExtractorError,
34 float_or_none,
35 get_element_by_attribute,
36 get_element_by_id,
37 int_or_none,
38 mimetype2ext,
39 orderedSet,
40 parse_codecs,
41 parse_duration,
42 qualities,
43 remove_quotes,
44 remove_start,
45 smuggle_url,
46 str_or_none,
47 str_to_int,
48 try_get,
49 unescapeHTML,
50 unified_strdate,
51 unsmuggle_url,
52 uppercase_escape,
53 url_or_none,
54 urlencode_postdata,
55 )
56
57
58 class YoutubeBaseInfoExtractor(InfoExtractor):
59 """Provide base functions for Youtube extractors"""
60 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
61 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
62
63 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
64 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
65 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
66
67 _NETRC_MACHINE = 'youtube'
68 # If True it will raise an error if no login info is provided
69 _LOGIN_REQUIRED = False
70
71 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
72
73 def _set_language(self):
74 self._set_cookie(
75 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
76 # YouTube sets the expire time to about two months
77 expire_time=time.time() + 2 * 30 * 24 * 3600)
78
79 def _ids_to_results(self, ids):
80 return [
81 self.url_result(vid_id, 'Youtube', video_id=vid_id)
82 for vid_id in ids]
83
84 def _login(self):
85 """
86 Attempt to log in to YouTube.
87 True is returned if successful or skipped.
88 False is returned if login failed.
89
90 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
91 """
92 username, password = self._get_login_info()
93 # No authentication to be performed
94 if username is None:
95 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
96 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
97 return True
98
99 login_page = self._download_webpage(
100 self._LOGIN_URL, None,
101 note='Downloading login page',
102 errnote='unable to fetch login page', fatal=False)
103 if login_page is False:
104 return
105
106 login_form = self._hidden_inputs(login_page)
107
108 def req(url, f_req, note, errnote):
109 data = login_form.copy()
110 data.update({
111 'pstMsg': 1,
112 'checkConnection': 'youtube',
113 'checkedDomains': 'youtube',
114 'hl': 'en',
115 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
116 'f.req': json.dumps(f_req),
117 'flowName': 'GlifWebSignIn',
118 'flowEntry': 'ServiceLogin',
119 })
120 return self._download_json(
121 url, None, note=note, errnote=errnote,
122 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
123 fatal=False,
124 data=urlencode_postdata(data), headers={
125 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
126 'Google-Accounts-XSRF': 1,
127 })
128
129 def warn(message):
130 self._downloader.report_warning(message)
131
132 lookup_req = [
133 username,
134 None, [], None, 'US', None, None, 2, False, True,
135 [
136 None, None,
137 [2, 1, None, 1,
138 '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',
139 None, [], 4],
140 1, [None, None, []], None, None, None, True
141 ],
142 username,
143 ]
144
145 lookup_results = req(
146 self._LOOKUP_URL, lookup_req,
147 'Looking up account info', 'Unable to look up account info')
148
149 if lookup_results is False:
150 return False
151
152 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
153 if not user_hash:
154 warn('Unable to extract user hash')
155 return False
156
157 challenge_req = [
158 user_hash,
159 None, 1, None, [1, None, None, None, [password, None, True]],
160 [
161 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],
162 1, [None, None, []], None, None, None, True
163 ]]
164
165 challenge_results = req(
166 self._CHALLENGE_URL, challenge_req,
167 'Logging in', 'Unable to log in')
168
169 if challenge_results is False:
170 return
171
172 login_res = try_get(challenge_results, lambda x: x[0][5], list)
173 if login_res:
174 login_msg = try_get(login_res, lambda x: x[5], compat_str)
175 warn(
176 'Unable to login: %s' % 'Invalid password'
177 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
178 return False
179
180 res = try_get(challenge_results, lambda x: x[0][-1], list)
181 if not res:
182 warn('Unable to extract result entry')
183 return False
184
185 login_challenge = try_get(res, lambda x: x[0][0], list)
186 if login_challenge:
187 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
188 if challenge_str == 'TWO_STEP_VERIFICATION':
189 # SEND_SUCCESS - TFA code has been successfully sent to phone
190 # QUOTA_EXCEEDED - reached the limit of TFA codes
191 status = try_get(login_challenge, lambda x: x[5], compat_str)
192 if status == 'QUOTA_EXCEEDED':
193 warn('Exceeded the limit of TFA codes, try later')
194 return False
195
196 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
197 if not tl:
198 warn('Unable to extract TL')
199 return False
200
201 tfa_code = self._get_tfa_info('2-step verification code')
202
203 if not tfa_code:
204 warn(
205 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
206 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
207 return False
208
209 tfa_code = remove_start(tfa_code, 'G-')
210
211 tfa_req = [
212 user_hash, None, 2, None,
213 [
214 9, None, None, None, None, None, None, None,
215 [None, tfa_code, True, 2]
216 ]]
217
218 tfa_results = req(
219 self._TFA_URL.format(tl), tfa_req,
220 'Submitting TFA code', 'Unable to submit TFA code')
221
222 if tfa_results is False:
223 return False
224
225 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
226 if tfa_res:
227 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
228 warn(
229 'Unable to finish TFA: %s' % 'Invalid TFA code'
230 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
231 return False
232
233 check_cookie_url = try_get(
234 tfa_results, lambda x: x[0][-1][2], compat_str)
235 else:
236 CHALLENGES = {
237 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
238 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
239 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
240 }
241 challenge = CHALLENGES.get(
242 challenge_str,
243 '%s returned error %s.' % (self.IE_NAME, challenge_str))
244 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
245 return False
246 else:
247 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
248
249 if not check_cookie_url:
250 warn('Unable to extract CheckCookie URL')
251 return False
252
253 check_cookie_results = self._download_webpage(
254 check_cookie_url, None, 'Checking cookie', fatal=False)
255
256 if check_cookie_results is False:
257 return False
258
259 if 'https://myaccount.google.com/' not in check_cookie_results:
260 warn('Unable to log in')
261 return False
262
263 return True
264
265 def _download_webpage_handle(self, *args, **kwargs):
266 query = kwargs.get('query', {}).copy()
267 query['disable_polymer'] = 'true'
268 kwargs['query'] = query
269 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
270 *args, **compat_kwargs(kwargs))
271
272 def _real_initialize(self):
273 if self._downloader is None:
274 return
275 self._set_language()
276 if not self._login():
277 return
278
279
280 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
281 # Extract entries from page with "Load more" button
282 def _entries(self, page, playlist_id):
283 more_widget_html = content_html = page
284 for page_num in itertools.count(1):
285 for entry in self._process_page(content_html):
286 yield entry
287
288 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
289 if not mobj:
290 break
291
292 count = 0
293 retries = 3
294 while count <= retries:
295 try:
296 # Downloading page may result in intermittent 5xx HTTP error
297 # that is usually worked around with a retry
298 more = self._download_json(
299 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
300 'Downloading page #%s%s'
301 % (page_num, ' (retry #%d)' % count if count else ''),
302 transform_source=uppercase_escape)
303 break
304 except ExtractorError as e:
305 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
306 count += 1
307 if count <= retries:
308 continue
309 raise
310
311 content_html = more['content_html']
312 if not content_html.strip():
313 # Some webpages show a "Load more" button but they don't
314 # have more videos
315 break
316 more_widget_html = more['load_more_widget_html']
317
318
319 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
320 def _process_page(self, content):
321 for video_id, video_title in self.extract_videos_from_page(content):
322 yield self.url_result(video_id, 'Youtube', video_id, video_title)
323
324 def extract_videos_from_page(self, page):
325 ids_in_page = []
326 titles_in_page = []
327 for mobj in re.finditer(self._VIDEO_RE, page):
328 # The link with index 0 is not the first video of the playlist (not sure if still actual)
329 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
330 continue
331 video_id = mobj.group('id')
332 video_title = unescapeHTML(mobj.group('title'))
333 if video_title:
334 video_title = video_title.strip()
335 try:
336 idx = ids_in_page.index(video_id)
337 if video_title and not titles_in_page[idx]:
338 titles_in_page[idx] = video_title
339 except ValueError:
340 ids_in_page.append(video_id)
341 titles_in_page.append(video_title)
342 return zip(ids_in_page, titles_in_page)
343
344
345 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
346 def _process_page(self, content):
347 for playlist_id in orderedSet(re.findall(
348 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
349 content)):
350 yield self.url_result(
351 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
352
353 def _real_extract(self, url):
354 playlist_id = self._match_id(url)
355 webpage = self._download_webpage(url, playlist_id)
356 title = self._og_search_title(webpage, fatal=False)
357 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
358
359
360 class YoutubeIE(YoutubeBaseInfoExtractor):
361 IE_DESC = 'YouTube.com'
362 _VALID_URL = r"""(?x)^
363 (
364 (?:https?://|//) # http(s):// or protocol-independent URL
365 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
366 (?:www\.)?deturl\.com/www\.youtube\.com/|
367 (?:www\.)?pwnyoutube\.com/|
368 (?:www\.)?hooktube\.com/|
369 (?:www\.)?yourepeat\.com/|
370 tube\.majestyc\.net/|
371 (?:(?:www|dev)\.)?invidio\.us/|
372 (?:www\.)?invidiou\.sh/|
373 (?:www\.)?invidious\.snopyta\.org/|
374 (?:www\.)?invidious\.kabi\.tk/|
375 (?:www\.)?vid\.wxzm\.sx/|
376 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
377 (?:.*?\#/)? # handle anchor (#/) redirect urls
378 (?: # the various things that can precede the ID:
379 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
380 |(?: # or the v= param in all its forms
381 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
382 (?:\?|\#!?) # the params delimiter ? or # or #!
383 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
384 v=
385 )
386 ))
387 |(?:
388 youtu\.be| # just youtu.be/xxxx
389 vid\.plus| # or vid.plus/xxxx
390 zwearz\.com/watch| # or zwearz.com/watch/xxxx
391 )/
392 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
393 )
394 )? # all until now is optional -> you can pass the naked ID
395 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
396 (?!.*?\blist=
397 (?:
398 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
399 WL # WL are handled by the watch later IE
400 )
401 )
402 (?(1).+)? # if we found the ID, everything can follow
403 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
404 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
405 _formats = {
406 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
407 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
408 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
409 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
410 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
411 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
412 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
413 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
414 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
415 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
416 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
417 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
418 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
419 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
420 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
421 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
422 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
423 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
424
425
426 # 3D videos
427 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
428 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
429 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
430 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
431 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
432 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
433 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
434
435 # Apple HTTP Live Streaming
436 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
437 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
438 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
439 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
440 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
441 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
442 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
443 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
444
445 # DASH mp4 video
446 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
447 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
448 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
449 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
450 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
451 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
452 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
453 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
454 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
455 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
456 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
457 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
458
459 # Dash mp4 audio
460 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
461 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
462 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
463 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
464 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
465 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
466 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
467
468 # Dash webm
469 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
470 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
471 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
472 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
473 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
474 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
475 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
476 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
477 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
478 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
479 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
480 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
481 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
482 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
483 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
484 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
485 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
486 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
487 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
488 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
489 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
490 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
491
492 # Dash webm audio
493 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
494 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
495
496 # Dash webm audio with opus inside
497 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
498 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
499 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
500
501 # RTMP (unnamed)
502 '_rtmp': {'protocol': 'rtmp'},
503 }
504 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
505
506 _GEO_BYPASS = False
507
508 IE_NAME = 'youtube'
509 _TESTS = [
510 {
511 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
512 'info_dict': {
513 'id': 'BaW_jenozKc',
514 'ext': 'mp4',
515 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
516 'uploader': 'Philipp Hagemeister',
517 'uploader_id': 'phihag',
518 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
519 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
520 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
521 'upload_date': '20121002',
522 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
523 'categories': ['Science & Technology'],
524 'tags': ['youtube-dl'],
525 'duration': 10,
526 'view_count': int,
527 'like_count': int,
528 'dislike_count': int,
529 'start_time': 1,
530 'end_time': 9,
531 }
532 },
533 {
534 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
535 'note': 'Test generic use_cipher_signature video (#897)',
536 'info_dict': {
537 'id': 'UxxajLWwzqY',
538 'ext': 'mp4',
539 'upload_date': '20120506',
540 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
541 'alt_title': 'I Love It (feat. Charli XCX)',
542 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
543 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
544 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
545 'iconic ep', 'iconic', 'love', 'it'],
546 'duration': 180,
547 'uploader': 'Icona Pop',
548 'uploader_id': 'IconaPop',
549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
550 'creator': 'Icona Pop',
551 'track': 'I Love It (feat. Charli XCX)',
552 'artist': 'Icona Pop',
553 }
554 },
555 {
556 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
557 'note': 'Test VEVO video with age protection (#956)',
558 'info_dict': {
559 'id': '07FYdnEawAQ',
560 'ext': 'mp4',
561 'upload_date': '20130703',
562 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
563 'alt_title': 'Tunnel Vision',
564 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
565 'duration': 419,
566 'uploader': 'justintimberlakeVEVO',
567 'uploader_id': 'justintimberlakeVEVO',
568 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
569 'creator': 'Justin Timberlake',
570 'track': 'Tunnel Vision',
571 'artist': 'Justin Timberlake',
572 'age_limit': 18,
573 }
574 },
575 {
576 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
577 'note': 'Embed-only video (#1746)',
578 'info_dict': {
579 'id': 'yZIXLfi8CZQ',
580 'ext': 'mp4',
581 'upload_date': '20120608',
582 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
583 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
584 'uploader': 'SET India',
585 'uploader_id': 'setindia',
586 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
587 'age_limit': 18,
588 }
589 },
590 {
591 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
592 'note': 'Use the first video ID in the URL',
593 'info_dict': {
594 'id': 'BaW_jenozKc',
595 'ext': 'mp4',
596 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
597 'uploader': 'Philipp Hagemeister',
598 'uploader_id': 'phihag',
599 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
600 'upload_date': '20121002',
601 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
602 'categories': ['Science & Technology'],
603 'tags': ['youtube-dl'],
604 'duration': 10,
605 'view_count': int,
606 'like_count': int,
607 'dislike_count': int,
608 },
609 'params': {
610 'skip_download': True,
611 },
612 },
613 {
614 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
615 'note': '256k DASH audio (format 141) via DASH manifest',
616 'info_dict': {
617 'id': 'a9LDPn-MO4I',
618 'ext': 'm4a',
619 'upload_date': '20121002',
620 'uploader_id': '8KVIDEO',
621 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
622 'description': '',
623 'uploader': '8KVIDEO',
624 'title': 'UHDTV TEST 8K VIDEO.mp4'
625 },
626 'params': {
627 'youtube_include_dash_manifest': True,
628 'format': '141',
629 },
630 'skip': 'format 141 not served anymore',
631 },
632 # DASH manifest with encrypted signature
633 {
634 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
635 'info_dict': {
636 'id': 'IB3lcPjvWLA',
637 'ext': 'm4a',
638 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
639 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
640 'duration': 244,
641 'uploader': 'AfrojackVEVO',
642 'uploader_id': 'AfrojackVEVO',
643 'upload_date': '20131011',
644 },
645 'params': {
646 'youtube_include_dash_manifest': True,
647 'format': '141/bestaudio[ext=m4a]',
648 },
649 },
650 # JS player signature function name containing $
651 {
652 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
653 'info_dict': {
654 'id': 'nfWlot6h_JM',
655 'ext': 'm4a',
656 'title': 'Taylor Swift - Shake It Off',
657 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
658 'duration': 242,
659 'uploader': 'TaylorSwiftVEVO',
660 'uploader_id': 'TaylorSwiftVEVO',
661 'upload_date': '20140818',
662 'creator': 'Taylor Swift',
663 },
664 'params': {
665 'youtube_include_dash_manifest': True,
666 'format': '141/bestaudio[ext=m4a]',
667 },
668 },
669 # Controversy video
670 {
671 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
672 'info_dict': {
673 'id': 'T4XJQO3qol8',
674 'ext': 'mp4',
675 'duration': 219,
676 'upload_date': '20100909',
677 'uploader': 'Amazing Atheist',
678 'uploader_id': 'TheAmazingAtheist',
679 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
680 'title': 'Burning Everyone\'s Koran',
681 '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',
682 }
683 },
684 # Normal age-gate video (No vevo, embed allowed)
685 {
686 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
687 'info_dict': {
688 'id': 'HtVdAasjOgU',
689 'ext': 'mp4',
690 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
691 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
692 'duration': 142,
693 'uploader': 'The Witcher',
694 'uploader_id': 'WitcherGame',
695 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
696 'upload_date': '20140605',
697 'age_limit': 18,
698 },
699 },
700 # Age-gate video with encrypted signature
701 {
702 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
703 'info_dict': {
704 'id': '6kLq3WMV1nU',
705 'ext': 'mp4',
706 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
707 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
708 'duration': 246,
709 'uploader': 'LloydVEVO',
710 'uploader_id': 'LloydVEVO',
711 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
712 'upload_date': '20110629',
713 'age_limit': 18,
714 },
715 },
716 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
717 # YouTube Red ad is not captured for creator
718 {
719 'url': '__2ABJjxzNo',
720 'info_dict': {
721 'id': '__2ABJjxzNo',
722 'ext': 'mp4',
723 'duration': 266,
724 'upload_date': '20100430',
725 'uploader_id': 'deadmau5',
726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
727 'creator': 'deadmau5',
728 'description': 'md5:12c56784b8032162bb936a5f76d55360',
729 'uploader': 'deadmau5',
730 'title': 'Deadmau5 - Some Chords (HD)',
731 'alt_title': 'Some Chords',
732 },
733 'expected_warnings': [
734 'DASH manifest missing',
735 ]
736 },
737 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
738 {
739 'url': 'lqQg6PlCWgI',
740 'info_dict': {
741 'id': 'lqQg6PlCWgI',
742 'ext': 'mp4',
743 'duration': 6085,
744 'upload_date': '20150827',
745 'uploader_id': 'olympic',
746 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
747 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
748 'uploader': 'Olympic',
749 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
750 },
751 'params': {
752 'skip_download': 'requires avconv',
753 }
754 },
755 # Non-square pixels
756 {
757 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
758 'info_dict': {
759 'id': '_b-2C3KPAM0',
760 'ext': 'mp4',
761 'stretched_ratio': 16 / 9.,
762 'duration': 85,
763 'upload_date': '20110310',
764 'uploader_id': 'AllenMeow',
765 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
766 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
767 'uploader': '孫ᄋᄅ',
768 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
769 },
770 },
771 # url_encoded_fmt_stream_map is empty string
772 {
773 'url': 'qEJwOuvDf7I',
774 'info_dict': {
775 'id': 'qEJwOuvDf7I',
776 'ext': 'webm',
777 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
778 'description': '',
779 'upload_date': '20150404',
780 'uploader_id': 'spbelect',
781 'uploader': 'Наблюдатели Петербурга',
782 },
783 'params': {
784 'skip_download': 'requires avconv',
785 },
786 'skip': 'This live event has ended.',
787 },
788 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
789 {
790 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
791 'info_dict': {
792 'id': 'FIl7x6_3R5Y',
793 'ext': 'webm',
794 'title': 'md5:7b81415841e02ecd4313668cde88737a',
795 'description': 'md5:116377fd2963b81ec4ce64b542173306',
796 'duration': 220,
797 'upload_date': '20150625',
798 'uploader_id': 'dorappi2000',
799 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
800 'uploader': 'dorappi2000',
801 'formats': 'mincount:31',
802 },
803 'skip': 'not actual anymore',
804 },
805 # DASH manifest with segment_list
806 {
807 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
808 'md5': '8ce563a1d667b599d21064e982ab9e31',
809 'info_dict': {
810 'id': 'CsmdDsKjzN8',
811 'ext': 'mp4',
812 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
813 'uploader': 'Airtek',
814 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
815 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
816 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
817 },
818 'params': {
819 'youtube_include_dash_manifest': True,
820 'format': '135', # bestvideo
821 },
822 'skip': 'This live event has ended.',
823 },
824 {
825 # Multifeed videos (multiple cameras), URL is for Main Camera
826 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
827 'info_dict': {
828 'id': 'jqWvoWXjCVs',
829 'title': 'teamPGP: Rocket League Noob Stream',
830 'description': 'md5:dc7872fb300e143831327f1bae3af010',
831 },
832 'playlist': [{
833 'info_dict': {
834 'id': 'jqWvoWXjCVs',
835 'ext': 'mp4',
836 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
837 'description': 'md5:dc7872fb300e143831327f1bae3af010',
838 'duration': 7335,
839 'upload_date': '20150721',
840 'uploader': 'Beer Games Beer',
841 'uploader_id': 'beergamesbeer',
842 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
843 'license': 'Standard YouTube License',
844 },
845 }, {
846 'info_dict': {
847 'id': '6h8e8xoXJzg',
848 'ext': 'mp4',
849 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
850 'description': 'md5:dc7872fb300e143831327f1bae3af010',
851 'duration': 7337,
852 'upload_date': '20150721',
853 'uploader': 'Beer Games Beer',
854 'uploader_id': 'beergamesbeer',
855 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
856 'license': 'Standard YouTube License',
857 },
858 }, {
859 'info_dict': {
860 'id': 'PUOgX5z9xZw',
861 'ext': 'mp4',
862 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
863 'description': 'md5:dc7872fb300e143831327f1bae3af010',
864 'duration': 7337,
865 'upload_date': '20150721',
866 'uploader': 'Beer Games Beer',
867 'uploader_id': 'beergamesbeer',
868 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
869 'license': 'Standard YouTube License',
870 },
871 }, {
872 'info_dict': {
873 'id': 'teuwxikvS5k',
874 'ext': 'mp4',
875 'title': 'teamPGP: Rocket League Noob Stream (zim)',
876 'description': 'md5:dc7872fb300e143831327f1bae3af010',
877 'duration': 7334,
878 'upload_date': '20150721',
879 'uploader': 'Beer Games Beer',
880 'uploader_id': 'beergamesbeer',
881 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
882 'license': 'Standard YouTube License',
883 },
884 }],
885 'params': {
886 'skip_download': True,
887 },
888 'skip': 'This video is not available.',
889 },
890 {
891 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
892 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
893 'info_dict': {
894 'id': 'gVfLd0zydlo',
895 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
896 },
897 'playlist_count': 2,
898 'skip': 'Not multifeed anymore',
899 },
900 {
901 'url': 'https://vid.plus/FlRa-iH7PGw',
902 'only_matching': True,
903 },
904 {
905 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
906 'only_matching': True,
907 },
908 {
909 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
910 # Also tests cut-off URL expansion in video description (see
911 # https://github.com/ytdl-org/youtube-dl/issues/1892,
912 # https://github.com/ytdl-org/youtube-dl/issues/8164)
913 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
914 'info_dict': {
915 'id': 'lsguqyKfVQg',
916 'ext': 'mp4',
917 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
918 'alt_title': 'Dark Walk - Position Music',
919 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
920 'duration': 133,
921 'upload_date': '20151119',
922 'uploader_id': 'IronSoulElf',
923 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
924 'uploader': 'IronSoulElf',
925 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
926 'track': 'Dark Walk - Position Music',
927 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
928 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
929 },
930 'params': {
931 'skip_download': True,
932 },
933 },
934 {
935 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
936 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
937 'only_matching': True,
938 },
939 {
940 # Video with yt:stretch=17:0
941 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
942 'info_dict': {
943 'id': 'Q39EVAstoRM',
944 'ext': 'mp4',
945 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
946 'description': 'md5:ee18a25c350637c8faff806845bddee9',
947 'upload_date': '20151107',
948 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
949 'uploader': 'CH GAMER DROID',
950 },
951 'params': {
952 'skip_download': True,
953 },
954 'skip': 'This video does not exist.',
955 },
956 {
957 # Video licensed under Creative Commons
958 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
959 'info_dict': {
960 'id': 'M4gD1WSo5mA',
961 'ext': 'mp4',
962 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
963 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
964 'duration': 721,
965 'upload_date': '20150127',
966 'uploader_id': 'BerkmanCenter',
967 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
968 'uploader': 'The Berkman Klein Center for Internet & Society',
969 'license': 'Creative Commons Attribution license (reuse allowed)',
970 },
971 'params': {
972 'skip_download': True,
973 },
974 },
975 {
976 # Channel-like uploader_url
977 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
978 'info_dict': {
979 'id': 'eQcmzGIKrzg',
980 'ext': 'mp4',
981 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
982 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
983 'duration': 4060,
984 'upload_date': '20151119',
985 'uploader': 'Bernie Sanders',
986 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
987 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
988 'license': 'Creative Commons Attribution license (reuse allowed)',
989 },
990 'params': {
991 'skip_download': True,
992 },
993 },
994 {
995 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
996 'only_matching': True,
997 },
998 {
999 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1000 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1001 'only_matching': True,
1002 },
1003 {
1004 # Rental video preview
1005 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1006 'info_dict': {
1007 'id': 'uGpuVWrhIzE',
1008 'ext': 'mp4',
1009 'title': 'Piku - Trailer',
1010 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1011 'upload_date': '20150811',
1012 'uploader': 'FlixMatrix',
1013 'uploader_id': 'FlixMatrixKaravan',
1014 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1015 'license': 'Standard YouTube License',
1016 },
1017 'params': {
1018 'skip_download': True,
1019 },
1020 'skip': 'This video is not available.',
1021 },
1022 {
1023 # YouTube Red video with episode data
1024 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1025 'info_dict': {
1026 'id': 'iqKdEhx-dD4',
1027 'ext': 'mp4',
1028 'title': 'Isolation - Mind Field (Ep 1)',
1029 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1030 'duration': 2085,
1031 'upload_date': '20170118',
1032 'uploader': 'Vsauce',
1033 'uploader_id': 'Vsauce',
1034 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1035 'series': 'Mind Field',
1036 'season_number': 1,
1037 'episode_number': 1,
1038 },
1039 'params': {
1040 'skip_download': True,
1041 },
1042 'expected_warnings': [
1043 'Skipping DASH manifest',
1044 ],
1045 },
1046 {
1047 # The following content has been identified by the YouTube community
1048 # as inappropriate or offensive to some audiences.
1049 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1050 'info_dict': {
1051 'id': '6SJNVb0GnPI',
1052 'ext': 'mp4',
1053 'title': 'Race Differences in Intelligence',
1054 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1055 'duration': 965,
1056 'upload_date': '20140124',
1057 'uploader': 'New Century Foundation',
1058 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1059 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1060 },
1061 'params': {
1062 'skip_download': True,
1063 },
1064 },
1065 {
1066 # itag 212
1067 'url': '1t24XAntNCY',
1068 'only_matching': True,
1069 },
1070 {
1071 # geo restricted to JP
1072 'url': 'sJL6WA-aGkQ',
1073 'only_matching': True,
1074 },
1075 {
1076 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1077 'only_matching': True,
1078 },
1079 {
1080 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1081 'only_matching': True,
1082 },
1083 {
1084 # DRM protected
1085 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1086 'only_matching': True,
1087 },
1088 {
1089 # Video with unsupported adaptive stream type formats
1090 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1091 'info_dict': {
1092 'id': 'Z4Vy8R84T1U',
1093 'ext': 'mp4',
1094 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1095 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1096 'duration': 433,
1097 'upload_date': '20130923',
1098 'uploader': 'Amelia Putri Harwita',
1099 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1100 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1101 'formats': 'maxcount:10',
1102 },
1103 'params': {
1104 'skip_download': True,
1105 'youtube_include_dash_manifest': False,
1106 },
1107 },
1108 {
1109 # Youtube Music Auto-generated description
1110 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1111 'info_dict': {
1112 'id': 'MgNrAu2pzNs',
1113 'ext': 'mp4',
1114 'title': 'Voyeur Girl',
1115 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1116 'upload_date': '20190312',
1117 'uploader': 'Various Artists - Topic',
1118 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1119 'artist': 'Stephen',
1120 'track': 'Voyeur Girl',
1121 'album': 'it\'s too much love to know my dear',
1122 'release_date': '20190313',
1123 'release_year': 2019,
1124 },
1125 'params': {
1126 'skip_download': True,
1127 },
1128 },
1129 {
1130 # Youtube Music Auto-generated description
1131 # Retrieve 'artist' field from 'Artist:' in video description
1132 # when it is present on youtube music video
1133 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1134 'info_dict': {
1135 'id': 'k0jLE7tTwjY',
1136 'ext': 'mp4',
1137 'title': 'Latch Feat. Sam Smith',
1138 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1139 'upload_date': '20150110',
1140 'uploader': 'Various Artists - Topic',
1141 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1142 'artist': 'Disclosure',
1143 'track': 'Latch Feat. Sam Smith',
1144 'album': 'Latch Featuring Sam Smith',
1145 'release_date': '20121008',
1146 'release_year': 2012,
1147 },
1148 'params': {
1149 'skip_download': True,
1150 },
1151 },
1152 {
1153 # Youtube Music Auto-generated description
1154 # handle multiple artists on youtube music video
1155 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1156 'info_dict': {
1157 'id': '74qn0eJSjpA',
1158 'ext': 'mp4',
1159 'title': 'Eastside',
1160 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1161 'upload_date': '20180710',
1162 'uploader': 'Benny Blanco - Topic',
1163 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1164 'artist': 'benny blanco, Halsey, Khalid',
1165 'track': 'Eastside',
1166 'album': 'Eastside',
1167 'release_date': '20180713',
1168 'release_year': 2018,
1169 },
1170 'params': {
1171 'skip_download': True,
1172 },
1173 },
1174 {
1175 # Youtube Music Auto-generated description
1176 # handle youtube music video with release_year and no release_date
1177 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1178 'info_dict': {
1179 'id': '-hcAI0g-f5M',
1180 'ext': 'mp4',
1181 'title': 'Put It On Me',
1182 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1183 'upload_date': '20180426',
1184 'uploader': 'Matt Maeson - Topic',
1185 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1186 'artist': 'Matt Maeson',
1187 'track': 'Put It On Me',
1188 'album': 'The Hearse',
1189 'release_date': None,
1190 'release_year': 2018,
1191 },
1192 'params': {
1193 'skip_download': True,
1194 },
1195 },
1196 ]
1197
1198 def __init__(self, *args, **kwargs):
1199 super(YoutubeIE, self).__init__(*args, **kwargs)
1200 self._player_cache = {}
1201
1202 def report_video_info_webpage_download(self, video_id):
1203 """Report attempt to download video info webpage."""
1204 self.to_screen('%s: Downloading video info webpage' % video_id)
1205
1206 def report_information_extraction(self, video_id):
1207 """Report attempt to extract video information."""
1208 self.to_screen('%s: Extracting video information' % video_id)
1209
1210 def report_unavailable_format(self, video_id, format):
1211 """Report extracted video URL."""
1212 self.to_screen('%s: Format %s not available' % (video_id, format))
1213
1214 def report_rtmp_download(self):
1215 """Indicate the download will use the RTMP protocol."""
1216 self.to_screen('RTMP download detected')
1217
1218 def _signature_cache_id(self, example_sig):
1219 """ Return a string representation of a signature """
1220 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1221
1222 def _extract_signature_function(self, video_id, player_url, example_sig):
1223 id_m = re.match(
1224 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1225 player_url)
1226 if not id_m:
1227 raise ExtractorError('Cannot identify player %r' % player_url)
1228 player_type = id_m.group('ext')
1229 player_id = id_m.group('id')
1230
1231 # Read from filesystem cache
1232 func_id = '%s_%s_%s' % (
1233 player_type, player_id, self._signature_cache_id(example_sig))
1234 assert os.path.basename(func_id) == func_id
1235
1236 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1237 if cache_spec is not None:
1238 return lambda s: ''.join(s[i] for i in cache_spec)
1239
1240 download_note = (
1241 'Downloading player %s' % player_url
1242 if self._downloader.params.get('verbose') else
1243 'Downloading %s player %s' % (player_type, player_id)
1244 )
1245 if player_type == 'js':
1246 code = self._download_webpage(
1247 player_url, video_id,
1248 note=download_note,
1249 errnote='Download of %s failed' % player_url)
1250 res = self._parse_sig_js(code)
1251 elif player_type == 'swf':
1252 urlh = self._request_webpage(
1253 player_url, video_id,
1254 note=download_note,
1255 errnote='Download of %s failed' % player_url)
1256 code = urlh.read()
1257 res = self._parse_sig_swf(code)
1258 else:
1259 assert False, 'Invalid player type %r' % player_type
1260
1261 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1262 cache_res = res(test_string)
1263 cache_spec = [ord(c) for c in cache_res]
1264
1265 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1266 return res
1267
1268 def _print_sig_code(self, func, example_sig):
1269 def gen_sig_code(idxs):
1270 def _genslice(start, end, step):
1271 starts = '' if start == 0 else str(start)
1272 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1273 steps = '' if step == 1 else (':%d' % step)
1274 return 's[%s%s%s]' % (starts, ends, steps)
1275
1276 step = None
1277 # Quelch pyflakes warnings - start will be set when step is set
1278 start = '(Never used)'
1279 for i, prev in zip(idxs[1:], idxs[:-1]):
1280 if step is not None:
1281 if i - prev == step:
1282 continue
1283 yield _genslice(start, prev, step)
1284 step = None
1285 continue
1286 if i - prev in [-1, 1]:
1287 step = i - prev
1288 start = prev
1289 continue
1290 else:
1291 yield 's[%d]' % prev
1292 if step is None:
1293 yield 's[%d]' % i
1294 else:
1295 yield _genslice(start, i, step)
1296
1297 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1298 cache_res = func(test_string)
1299 cache_spec = [ord(c) for c in cache_res]
1300 expr_code = ' + '.join(gen_sig_code(cache_spec))
1301 signature_id_tuple = '(%s)' % (
1302 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1303 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1304 ' return %s\n') % (signature_id_tuple, expr_code)
1305 self.to_screen('Extracted signature function:\n' + code)
1306
1307 def _parse_sig_js(self, jscode):
1308 funcname = self._search_regex(
1309 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1310 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1311 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
1312 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1313 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1314 jscode, 'Initial JS player signature function name', group='sig')
1315
1316 jsi = JSInterpreter(jscode)
1317 initial_function = jsi.extract_function(funcname)
1318 return lambda s: initial_function([s])
1319
1320 def _parse_sig_swf(self, file_contents):
1321 swfi = SWFInterpreter(file_contents)
1322 TARGET_CLASSNAME = 'SignatureDecipher'
1323 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1324 initial_function = swfi.extract_function(searched_class, 'decipher')
1325 return lambda s: initial_function([s])
1326
1327 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1328 """Turn the encrypted s field into a working signature"""
1329
1330 if player_url is None:
1331 raise ExtractorError('Cannot decrypt signature without player_url')
1332
1333 if player_url.startswith('//'):
1334 player_url = 'https:' + player_url
1335 elif not re.match(r'https?://', player_url):
1336 player_url = compat_urlparse.urljoin(
1337 'https://www.youtube.com', player_url)
1338 try:
1339 player_id = (player_url, self._signature_cache_id(s))
1340 if player_id not in self._player_cache:
1341 func = self._extract_signature_function(
1342 video_id, player_url, s
1343 )
1344 self._player_cache[player_id] = func
1345 func = self._player_cache[player_id]
1346 if self._downloader.params.get('youtube_print_sig_code'):
1347 self._print_sig_code(func, s)
1348 return func(s)
1349 except Exception as e:
1350 tb = traceback.format_exc()
1351 raise ExtractorError(
1352 'Signature extraction failed: ' + tb, cause=e)
1353
1354 def _get_subtitles(self, video_id, webpage):
1355 try:
1356 subs_doc = self._download_xml(
1357 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1358 video_id, note=False)
1359 except ExtractorError as err:
1360 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1361 return {}
1362
1363 sub_lang_list = {}
1364 for track in subs_doc.findall('track'):
1365 lang = track.attrib['lang_code']
1366 if lang in sub_lang_list:
1367 continue
1368 sub_formats = []
1369 for ext in self._SUBTITLE_FORMATS:
1370 params = compat_urllib_parse_urlencode({
1371 'lang': lang,
1372 'v': video_id,
1373 'fmt': ext,
1374 'name': track.attrib['name'].encode('utf-8'),
1375 })
1376 sub_formats.append({
1377 'url': 'https://www.youtube.com/api/timedtext?' + params,
1378 'ext': ext,
1379 })
1380 sub_lang_list[lang] = sub_formats
1381 if not sub_lang_list:
1382 self._downloader.report_warning('video doesn\'t have subtitles')
1383 return {}
1384 return sub_lang_list
1385
1386 def _get_ytplayer_config(self, video_id, webpage):
1387 patterns = (
1388 # User data may contain arbitrary character sequences that may affect
1389 # JSON extraction with regex, e.g. when '};' is contained the second
1390 # regex won't capture the whole JSON. Yet working around by trying more
1391 # concrete regex first keeping in mind proper quoted string handling
1392 # to be implemented in future that will replace this workaround (see
1393 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1394 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1395 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1396 r';ytplayer\.config\s*=\s*({.+?});',
1397 )
1398 config = self._search_regex(
1399 patterns, webpage, 'ytplayer.config', default=None)
1400 if config:
1401 return self._parse_json(
1402 uppercase_escape(config), video_id, fatal=False)
1403
1404 def _get_automatic_captions(self, video_id, webpage):
1405 """We need the webpage for getting the captions url, pass it as an
1406 argument to speed up the process."""
1407 self.to_screen('%s: Looking for automatic captions' % video_id)
1408 player_config = self._get_ytplayer_config(video_id, webpage)
1409 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1410 if not player_config:
1411 self._downloader.report_warning(err_msg)
1412 return {}
1413 try:
1414 args = player_config['args']
1415 caption_url = args.get('ttsurl')
1416 if caption_url:
1417 timestamp = args['timestamp']
1418 # We get the available subtitles
1419 list_params = compat_urllib_parse_urlencode({
1420 'type': 'list',
1421 'tlangs': 1,
1422 'asrs': 1,
1423 })
1424 list_url = caption_url + '&' + list_params
1425 caption_list = self._download_xml(list_url, video_id)
1426 original_lang_node = caption_list.find('track')
1427 if original_lang_node is None:
1428 self._downloader.report_warning('Video doesn\'t have automatic captions')
1429 return {}
1430 original_lang = original_lang_node.attrib['lang_code']
1431 caption_kind = original_lang_node.attrib.get('kind', '')
1432
1433 sub_lang_list = {}
1434 for lang_node in caption_list.findall('target'):
1435 sub_lang = lang_node.attrib['lang_code']
1436 sub_formats = []
1437 for ext in self._SUBTITLE_FORMATS:
1438 params = compat_urllib_parse_urlencode({
1439 'lang': original_lang,
1440 'tlang': sub_lang,
1441 'fmt': ext,
1442 'ts': timestamp,
1443 'kind': caption_kind,
1444 })
1445 sub_formats.append({
1446 'url': caption_url + '&' + params,
1447 'ext': ext,
1448 })
1449 sub_lang_list[sub_lang] = sub_formats
1450 return sub_lang_list
1451
1452 def make_captions(sub_url, sub_langs):
1453 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1454 caption_qs = compat_parse_qs(parsed_sub_url.query)
1455 captions = {}
1456 for sub_lang in sub_langs:
1457 sub_formats = []
1458 for ext in self._SUBTITLE_FORMATS:
1459 caption_qs.update({
1460 'tlang': [sub_lang],
1461 'fmt': [ext],
1462 })
1463 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1464 query=compat_urllib_parse_urlencode(caption_qs, True)))
1465 sub_formats.append({
1466 'url': sub_url,
1467 'ext': ext,
1468 })
1469 captions[sub_lang] = sub_formats
1470 return captions
1471
1472 # New captions format as of 22.06.2017
1473 player_response = args.get('player_response')
1474 if player_response and isinstance(player_response, compat_str):
1475 player_response = self._parse_json(
1476 player_response, video_id, fatal=False)
1477 if player_response:
1478 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1479 base_url = renderer['captionTracks'][0]['baseUrl']
1480 sub_lang_list = []
1481 for lang in renderer['translationLanguages']:
1482 lang_code = lang.get('languageCode')
1483 if lang_code:
1484 sub_lang_list.append(lang_code)
1485 return make_captions(base_url, sub_lang_list)
1486
1487 # Some videos don't provide ttsurl but rather caption_tracks and
1488 # caption_translation_languages (e.g. 20LmZk1hakA)
1489 # Does not used anymore as of 22.06.2017
1490 caption_tracks = args['caption_tracks']
1491 caption_translation_languages = args['caption_translation_languages']
1492 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1493 sub_lang_list = []
1494 for lang in caption_translation_languages.split(','):
1495 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1496 sub_lang = lang_qs.get('lc', [None])[0]
1497 if sub_lang:
1498 sub_lang_list.append(sub_lang)
1499 return make_captions(caption_url, sub_lang_list)
1500 # An extractor error can be raise by the download process if there are
1501 # no automatic captions but there are subtitles
1502 except (KeyError, IndexError, ExtractorError):
1503 self._downloader.report_warning(err_msg)
1504 return {}
1505
1506 def _mark_watched(self, video_id, video_info, player_response):
1507 playback_url = url_or_none(try_get(
1508 player_response,
1509 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1510 video_info, lambda x: x['videostats_playback_base_url'][0]))
1511 if not playback_url:
1512 return
1513 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1514 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1515
1516 # cpn generation algorithm is reverse engineered from base.js.
1517 # In fact it works even with dummy cpn.
1518 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1519 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1520
1521 qs.update({
1522 'ver': ['2'],
1523 'cpn': [cpn],
1524 })
1525 playback_url = compat_urlparse.urlunparse(
1526 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1527
1528 self._download_webpage(
1529 playback_url, video_id, 'Marking watched',
1530 'Unable to mark watched', fatal=False)
1531
1532 @staticmethod
1533 def _extract_urls(webpage):
1534 # Embedded YouTube player
1535 entries = [
1536 unescapeHTML(mobj.group('url'))
1537 for mobj in re.finditer(r'''(?x)
1538 (?:
1539 <iframe[^>]+?src=|
1540 data-video-url=|
1541 <embed[^>]+?src=|
1542 embedSWF\(?:\s*|
1543 <object[^>]+data=|
1544 new\s+SWFObject\(
1545 )
1546 (["\'])
1547 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1548 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1549 \1''', webpage)]
1550
1551 # lazyYT YouTube embed
1552 entries.extend(list(map(
1553 unescapeHTML,
1554 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1555
1556 # Wordpress "YouTube Video Importer" plugin
1557 matches = re.findall(r'''(?x)<div[^>]+
1558 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1559 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1560 entries.extend(m[-1] for m in matches)
1561
1562 return entries
1563
1564 @staticmethod
1565 def _extract_url(webpage):
1566 urls = YoutubeIE._extract_urls(webpage)
1567 return urls[0] if urls else None
1568
1569 @classmethod
1570 def extract_id(cls, url):
1571 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1572 if mobj is None:
1573 raise ExtractorError('Invalid URL: %s' % url)
1574 video_id = mobj.group(2)
1575 return video_id
1576
1577 def _extract_annotations(self, video_id):
1578 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1579 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1580
1581 @staticmethod
1582 def _extract_chapters(description, duration):
1583 if not description:
1584 return None
1585 chapter_lines = re.findall(
1586 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1587 description)
1588 if not chapter_lines:
1589 return None
1590 chapters = []
1591 for next_num, (chapter_line, time_point) in enumerate(
1592 chapter_lines, start=1):
1593 start_time = parse_duration(time_point)
1594 if start_time is None:
1595 continue
1596 if start_time > duration:
1597 break
1598 end_time = (duration if next_num == len(chapter_lines)
1599 else parse_duration(chapter_lines[next_num][1]))
1600 if end_time is None:
1601 continue
1602 if end_time > duration:
1603 end_time = duration
1604 if start_time > end_time:
1605 break
1606 chapter_title = re.sub(
1607 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1608 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1609 chapters.append({
1610 'start_time': start_time,
1611 'end_time': end_time,
1612 'title': chapter_title,
1613 })
1614 return chapters
1615
1616 def _real_extract(self, url):
1617 url, smuggled_data = unsmuggle_url(url, {})
1618
1619 proto = (
1620 'http' if self._downloader.params.get('prefer_insecure', False)
1621 else 'https')
1622
1623 start_time = None
1624 end_time = None
1625 parsed_url = compat_urllib_parse_urlparse(url)
1626 for component in [parsed_url.fragment, parsed_url.query]:
1627 query = compat_parse_qs(component)
1628 if start_time is None and 't' in query:
1629 start_time = parse_duration(query['t'][0])
1630 if start_time is None and 'start' in query:
1631 start_time = parse_duration(query['start'][0])
1632 if end_time is None and 'end' in query:
1633 end_time = parse_duration(query['end'][0])
1634
1635 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1636 mobj = re.search(self._NEXT_URL_RE, url)
1637 if mobj:
1638 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1639 video_id = self.extract_id(url)
1640
1641 # Get video webpage
1642 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1643 video_webpage = self._download_webpage(url, video_id)
1644
1645 # Attempt to extract SWF player URL
1646 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1647 if mobj is not None:
1648 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1649 else:
1650 player_url = None
1651
1652 dash_mpds = []
1653
1654 def add_dash_mpd(video_info):
1655 dash_mpd = video_info.get('dashmpd')
1656 if dash_mpd and dash_mpd[0] not in dash_mpds:
1657 dash_mpds.append(dash_mpd[0])
1658
1659 def add_dash_mpd_pr(pl_response):
1660 dash_mpd = url_or_none(try_get(
1661 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1662 compat_str))
1663 if dash_mpd and dash_mpd not in dash_mpds:
1664 dash_mpds.append(dash_mpd)
1665
1666 is_live = None
1667 view_count = None
1668
1669 def extract_view_count(v_info):
1670 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1671
1672 def extract_token(v_info):
1673 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1674
1675 player_response = {}
1676
1677 # Get video info
1678 embed_webpage = None
1679 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1680 age_gate = True
1681 # We simulate the access to the video from www.youtube.com/v/{video_id}
1682 # this can be viewed without login into Youtube
1683 url = proto + '://www.youtube.com/embed/%s' % video_id
1684 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1685 data = compat_urllib_parse_urlencode({
1686 'video_id': video_id,
1687 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1688 'sts': self._search_regex(
1689 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1690 })
1691 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1692 video_info_webpage = self._download_webpage(
1693 video_info_url, video_id,
1694 note='Refetching age-gated info webpage',
1695 errnote='unable to download video info webpage')
1696 video_info = compat_parse_qs(video_info_webpage)
1697 add_dash_mpd(video_info)
1698 else:
1699 age_gate = False
1700 video_info = None
1701 sts = None
1702 # Try looking directly into the video webpage
1703 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1704 if ytplayer_config:
1705 args = ytplayer_config['args']
1706 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1707 # Convert to the same format returned by compat_parse_qs
1708 video_info = dict((k, [v]) for k, v in args.items())
1709 add_dash_mpd(video_info)
1710 # Rental video is not rented but preview is available (e.g.
1711 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1712 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1713 if not video_info and args.get('ypc_vid'):
1714 return self.url_result(
1715 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1716 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1717 is_live = True
1718 sts = ytplayer_config.get('sts')
1719 if not player_response:
1720 pl_response = str_or_none(args.get('player_response'))
1721 if pl_response:
1722 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1723 if isinstance(pl_response, dict):
1724 player_response = pl_response
1725 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1726 add_dash_mpd_pr(player_response)
1727 # We also try looking in get_video_info since it may contain different dashmpd
1728 # URL that points to a DASH manifest with possibly different itag set (some itags
1729 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1730 # manifest pointed by get_video_info's dashmpd).
1731 # The general idea is to take a union of itags of both DASH manifests (for example
1732 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1733 self.report_video_info_webpage_download(video_id)
1734 for el in ('embedded', 'detailpage', 'vevo', ''):
1735 query = {
1736 'video_id': video_id,
1737 'ps': 'default',
1738 'eurl': '',
1739 'gl': 'US',
1740 'hl': 'en',
1741 }
1742 if el:
1743 query['el'] = el
1744 if sts:
1745 query['sts'] = sts
1746 video_info_webpage = self._download_webpage(
1747 '%s://www.youtube.com/get_video_info' % proto,
1748 video_id, note=False,
1749 errnote='unable to download video info webpage',
1750 fatal=False, query=query)
1751 if not video_info_webpage:
1752 continue
1753 get_video_info = compat_parse_qs(video_info_webpage)
1754 if not player_response:
1755 pl_response = get_video_info.get('player_response', [None])[0]
1756 if isinstance(pl_response, dict):
1757 player_response = pl_response
1758 add_dash_mpd_pr(player_response)
1759 add_dash_mpd(get_video_info)
1760 if view_count is None:
1761 view_count = extract_view_count(get_video_info)
1762 if not video_info:
1763 video_info = get_video_info
1764 get_token = extract_token(get_video_info)
1765 if get_token:
1766 # Different get_video_info requests may report different results, e.g.
1767 # some may report video unavailability, but some may serve it without
1768 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1769 # the original webpage as well as el=info and el=embedded get_video_info
1770 # requests report video unavailability due to geo restriction while
1771 # el=detailpage succeeds and returns valid data). This is probably
1772 # due to YouTube measures against IP ranges of hosting providers.
1773 # Working around by preferring the first succeeded video_info containing
1774 # the token if no such video_info yet was found.
1775 token = extract_token(video_info)
1776 if not token:
1777 video_info = get_video_info
1778 break
1779
1780 def extract_unavailable_message():
1781 return self._html_search_regex(
1782 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1783 video_webpage, 'unavailable message', default=None)
1784
1785 if not video_info:
1786 unavailable_message = extract_unavailable_message()
1787 if not unavailable_message:
1788 unavailable_message = 'Unable to extract video data'
1789 raise ExtractorError(
1790 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1791
1792 if video_info.get('license_info'):
1793 raise ExtractorError('This video is DRM protected.', expected=True)
1794
1795 video_details = try_get(
1796 player_response, lambda x: x['videoDetails'], dict) or {}
1797
1798 # title
1799 if 'title' in video_info:
1800 video_title = video_info['title'][0]
1801 elif 'title' in player_response:
1802 video_title = video_details['title']
1803 else:
1804 self._downloader.report_warning('Unable to extract video title')
1805 video_title = '_'
1806
1807 # description
1808 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1809 if video_description:
1810
1811 def replace_url(m):
1812 redir_url = compat_urlparse.urljoin(url, m.group(1))
1813 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1814 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1815 qs = compat_parse_qs(parsed_redir_url.query)
1816 q = qs.get('q')
1817 if q and q[0]:
1818 return q[0]
1819 return redir_url
1820
1821 description_original = video_description = re.sub(r'''(?x)
1822 <a\s+
1823 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1824 (?:title|href)="([^"]+)"\s+
1825 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1826 class="[^"]*"[^>]*>
1827 [^<]+\.{3}\s*
1828 </a>
1829 ''', replace_url, video_description)
1830 video_description = clean_html(video_description)
1831 else:
1832 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1833 if fd_mobj:
1834 video_description = unescapeHTML(fd_mobj.group(1))
1835 else:
1836 video_description = ''
1837
1838 if not smuggled_data.get('force_singlefeed', False):
1839 if not self._downloader.params.get('noplaylist'):
1840 multifeed_metadata_list = try_get(
1841 player_response,
1842 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1843 compat_str) or try_get(
1844 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1845 if multifeed_metadata_list:
1846 entries = []
1847 feed_ids = []
1848 for feed in multifeed_metadata_list.split(','):
1849 # Unquote should take place before split on comma (,) since textual
1850 # fields may contain comma as well (see
1851 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1852 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1853 entries.append({
1854 '_type': 'url_transparent',
1855 'ie_key': 'Youtube',
1856 'url': smuggle_url(
1857 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1858 {'force_singlefeed': True}),
1859 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1860 })
1861 feed_ids.append(feed_data['id'][0])
1862 self.to_screen(
1863 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1864 % (', '.join(feed_ids), video_id))
1865 return self.playlist_result(entries, video_id, video_title, video_description)
1866 else:
1867 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1868
1869 if view_count is None:
1870 view_count = extract_view_count(video_info)
1871 if view_count is None and video_details:
1872 view_count = int_or_none(video_details.get('viewCount'))
1873
1874 # Check for "rental" videos
1875 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1876 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1877
1878 def _extract_filesize(media_url):
1879 return int_or_none(self._search_regex(
1880 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1881
1882 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1883 self.report_rtmp_download()
1884 formats = [{
1885 'format_id': '_rtmp',
1886 'protocol': 'rtmp',
1887 'url': video_info['conn'][0],
1888 'player_url': player_url,
1889 }]
1890 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):
1891 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1892 if 'rtmpe%3Dyes' in encoded_url_map:
1893 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1894 formats_spec = {}
1895 fmt_list = video_info.get('fmt_list', [''])[0]
1896 if fmt_list:
1897 for fmt in fmt_list.split(','):
1898 spec = fmt.split('/')
1899 if len(spec) > 1:
1900 width_height = spec[1].split('x')
1901 if len(width_height) == 2:
1902 formats_spec[spec[0]] = {
1903 'resolution': spec[1],
1904 'width': int_or_none(width_height[0]),
1905 'height': int_or_none(width_height[1]),
1906 }
1907 q = qualities(['small', 'medium', 'hd720'])
1908 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1909 if streaming_formats:
1910 for fmt in streaming_formats:
1911 itag = str_or_none(fmt.get('itag'))
1912 if not itag:
1913 continue
1914 quality = fmt.get('quality')
1915 quality_label = fmt.get('qualityLabel') or quality
1916 formats_spec[itag] = {
1917 'asr': int_or_none(fmt.get('audioSampleRate')),
1918 'filesize': int_or_none(fmt.get('contentLength')),
1919 'format_note': quality_label,
1920 'fps': int_or_none(fmt.get('fps')),
1921 'height': int_or_none(fmt.get('height')),
1922 'quality': q(quality),
1923 # bitrate for itag 43 is always 2147483647
1924 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1925 'width': int_or_none(fmt.get('width')),
1926 }
1927 formats = []
1928 for url_data_str in encoded_url_map.split(','):
1929 url_data = compat_parse_qs(url_data_str)
1930 if 'itag' not in url_data or 'url' not in url_data:
1931 continue
1932 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1933 # Unsupported FORMAT_STREAM_TYPE_OTF
1934 if stream_type == 3:
1935 continue
1936 format_id = url_data['itag'][0]
1937 url = url_data['url'][0]
1938
1939 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1940 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1941 jsplayer_url_json = self._search_regex(
1942 ASSETS_RE,
1943 embed_webpage if age_gate else video_webpage,
1944 'JS player URL (1)', default=None)
1945 if not jsplayer_url_json and not age_gate:
1946 # We need the embed website after all
1947 if embed_webpage is None:
1948 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1949 embed_webpage = self._download_webpage(
1950 embed_url, video_id, 'Downloading embed webpage')
1951 jsplayer_url_json = self._search_regex(
1952 ASSETS_RE, embed_webpage, 'JS player URL')
1953
1954 player_url = json.loads(jsplayer_url_json)
1955 if player_url is None:
1956 player_url_json = self._search_regex(
1957 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1958 video_webpage, 'age gate player URL')
1959 player_url = json.loads(player_url_json)
1960
1961 if 'sig' in url_data:
1962 url += '&signature=' + url_data['sig'][0]
1963 elif 's' in url_data:
1964 encrypted_sig = url_data['s'][0]
1965
1966 if self._downloader.params.get('verbose'):
1967 if player_url is None:
1968 player_version = 'unknown'
1969 player_desc = 'unknown'
1970 else:
1971 if player_url.endswith('swf'):
1972 player_version = self._search_regex(
1973 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1974 'flash player', fatal=False)
1975 player_desc = 'flash player %s' % player_version
1976 else:
1977 player_version = self._search_regex(
1978 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1979 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1980 player_url,
1981 'html5 player', fatal=False)
1982 player_desc = 'html5 player %s' % player_version
1983
1984 parts_sizes = self._signature_cache_id(encrypted_sig)
1985 self.to_screen('{%s} signature length %s, %s' %
1986 (format_id, parts_sizes, player_desc))
1987
1988 signature = self._decrypt_signature(
1989 encrypted_sig, video_id, player_url, age_gate)
1990 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
1991 url += '&%s=%s' % (sp, signature)
1992 if 'ratebypass' not in url:
1993 url += '&ratebypass=yes'
1994
1995 dct = {
1996 'format_id': format_id,
1997 'url': url,
1998 'player_url': player_url,
1999 }
2000 if format_id in self._formats:
2001 dct.update(self._formats[format_id])
2002 if format_id in formats_spec:
2003 dct.update(formats_spec[format_id])
2004
2005 # Some itags are not included in DASH manifest thus corresponding formats will
2006 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2007 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2008 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2009 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2010
2011 filesize = int_or_none(url_data.get(
2012 'clen', [None])[0]) or _extract_filesize(url)
2013
2014 quality = url_data.get('quality', [None])[0]
2015
2016 more_fields = {
2017 'filesize': filesize,
2018 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2019 'width': width,
2020 'height': height,
2021 'fps': int_or_none(url_data.get('fps', [None])[0]),
2022 'format_note': url_data.get('quality_label', [None])[0] or quality,
2023 'quality': q(quality),
2024 }
2025 for key, value in more_fields.items():
2026 if value:
2027 dct[key] = value
2028 type_ = url_data.get('type', [None])[0]
2029 if type_:
2030 type_split = type_.split(';')
2031 kind_ext = type_split[0].split('/')
2032 if len(kind_ext) == 2:
2033 kind, _ = kind_ext
2034 dct['ext'] = mimetype2ext(type_split[0])
2035 if kind in ('audio', 'video'):
2036 codecs = None
2037 for mobj in re.finditer(
2038 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2039 if mobj.group('key') == 'codecs':
2040 codecs = mobj.group('val')
2041 break
2042 if codecs:
2043 dct.update(parse_codecs(codecs))
2044 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2045 dct['downloader_options'] = {
2046 # Youtube throttles chunks >~10M
2047 'http_chunk_size': 10485760,
2048 }
2049 formats.append(dct)
2050 else:
2051 manifest_url = (
2052 url_or_none(try_get(
2053 player_response,
2054 lambda x: x['streamingData']['hlsManifestUrl'],
2055 compat_str))
2056 or url_or_none(try_get(
2057 video_info, lambda x: x['hlsvp'][0], compat_str)))
2058 if manifest_url:
2059 formats = []
2060 m3u8_formats = self._extract_m3u8_formats(
2061 manifest_url, video_id, 'mp4', fatal=False)
2062 for a_format in m3u8_formats:
2063 itag = self._search_regex(
2064 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2065 if itag:
2066 a_format['format_id'] = itag
2067 if itag in self._formats:
2068 dct = self._formats[itag].copy()
2069 dct.update(a_format)
2070 a_format = dct
2071 a_format['player_url'] = player_url
2072 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2073 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2074 formats.append(a_format)
2075 else:
2076 error_message = clean_html(video_info.get('reason', [None])[0])
2077 if not error_message:
2078 error_message = extract_unavailable_message()
2079 if error_message:
2080 raise ExtractorError(error_message, expected=True)
2081 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2082
2083 # uploader
2084 video_uploader = try_get(
2085 video_info, lambda x: x['author'][0],
2086 compat_str) or str_or_none(video_details.get('author'))
2087 if video_uploader:
2088 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2089 else:
2090 self._downloader.report_warning('unable to extract uploader name')
2091
2092 # uploader_id
2093 video_uploader_id = None
2094 video_uploader_url = None
2095 mobj = re.search(
2096 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2097 video_webpage)
2098 if mobj is not None:
2099 video_uploader_id = mobj.group('uploader_id')
2100 video_uploader_url = mobj.group('uploader_url')
2101 else:
2102 self._downloader.report_warning('unable to extract uploader nickname')
2103
2104 channel_id = (
2105 str_or_none(video_details.get('channelId'))
2106 or self._html_search_meta(
2107 'channelId', video_webpage, 'channel id', default=None)
2108 or self._search_regex(
2109 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2110 video_webpage, 'channel id', default=None, group='id'))
2111 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2112
2113 # thumbnail image
2114 # We try first to get a high quality image:
2115 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2116 video_webpage, re.DOTALL)
2117 if m_thumb is not None:
2118 video_thumbnail = m_thumb.group(1)
2119 elif 'thumbnail_url' not in video_info:
2120 self._downloader.report_warning('unable to extract video thumbnail')
2121 video_thumbnail = None
2122 else: # don't panic if we can't find it
2123 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2124
2125 # upload date
2126 upload_date = self._html_search_meta(
2127 'datePublished', video_webpage, 'upload date', default=None)
2128 if not upload_date:
2129 upload_date = self._search_regex(
2130 [r'(?s)id="eow-date.*?>(.*?)</span>',
2131 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2132 video_webpage, 'upload date', default=None)
2133 upload_date = unified_strdate(upload_date)
2134
2135 video_license = self._html_search_regex(
2136 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2137 video_webpage, 'license', default=None)
2138
2139 m_music = re.search(
2140 r'''(?x)
2141 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2142 <ul[^>]*>\s*
2143 <li>(?P<title>.+?)
2144 by (?P<creator>.+?)
2145 (?:
2146 \(.+?\)|
2147 <a[^>]*
2148 (?:
2149 \bhref=["\']/red[^>]*>| # drop possible
2150 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2151 )
2152 .*?
2153 )?</li
2154 ''',
2155 video_webpage)
2156 if m_music:
2157 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2158 video_creator = clean_html(m_music.group('creator'))
2159 else:
2160 video_alt_title = video_creator = None
2161
2162 def extract_meta(field):
2163 return self._html_search_regex(
2164 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2165 video_webpage, field, default=None)
2166
2167 track = extract_meta('Song')
2168 artist = extract_meta('Artist')
2169 album = extract_meta('Album')
2170
2171 # Youtube Music Auto-generated description
2172 release_date = release_year = None
2173 if video_description:
2174 mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?', video_description)
2175 if mobj:
2176 if not track:
2177 track = mobj.group('track').strip()
2178 if not artist:
2179 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2180 if not album:
2181 album = mobj.group('album'.strip())
2182 release_year = mobj.group('release_year')
2183 release_date = mobj.group('release_date')
2184 if release_date:
2185 release_date = release_date.replace('-', '')
2186 if not release_year:
2187 release_year = int(release_date[:4])
2188 if release_year:
2189 release_year = int(release_year)
2190
2191 m_episode = re.search(
2192 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2193 video_webpage)
2194 if m_episode:
2195 series = unescapeHTML(m_episode.group('series'))
2196 season_number = int(m_episode.group('season'))
2197 episode_number = int(m_episode.group('episode'))
2198 else:
2199 series = season_number = episode_number = None
2200
2201 m_cat_container = self._search_regex(
2202 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2203 video_webpage, 'categories', default=None)
2204 if m_cat_container:
2205 category = self._html_search_regex(
2206 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2207 default=None)
2208 video_categories = None if category is None else [category]
2209 else:
2210 video_categories = None
2211
2212 video_tags = [
2213 unescapeHTML(m.group('content'))
2214 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2215
2216 def _extract_count(count_name):
2217 return str_to_int(self._search_regex(
2218 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2219 % re.escape(count_name),
2220 video_webpage, count_name, default=None))
2221
2222 like_count = _extract_count('like')
2223 dislike_count = _extract_count('dislike')
2224
2225 if view_count is None:
2226 view_count = str_to_int(self._search_regex(
2227 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2228 'view count', default=None))
2229
2230 # subtitles
2231 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2232 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2233
2234 video_duration = try_get(
2235 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2236 if not video_duration:
2237 video_duration = int_or_none(video_details.get('lengthSeconds'))
2238 if not video_duration:
2239 video_duration = parse_duration(self._html_search_meta(
2240 'duration', video_webpage, 'video duration'))
2241
2242 # annotations
2243 video_annotations = None
2244 if self._downloader.params.get('writeannotations', False):
2245 video_annotations = self._extract_annotations(video_id)
2246
2247 chapters = self._extract_chapters(description_original, video_duration)
2248
2249 # Look for the DASH manifest
2250 if self._downloader.params.get('youtube_include_dash_manifest', True):
2251 dash_mpd_fatal = True
2252 for mpd_url in dash_mpds:
2253 dash_formats = {}
2254 try:
2255 def decrypt_sig(mobj):
2256 s = mobj.group(1)
2257 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2258 return '/signature/%s' % dec_s
2259
2260 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2261
2262 for df in self._extract_mpd_formats(
2263 mpd_url, video_id, fatal=dash_mpd_fatal,
2264 formats_dict=self._formats):
2265 if not df.get('filesize'):
2266 df['filesize'] = _extract_filesize(df['url'])
2267 # Do not overwrite DASH format found in some previous DASH manifest
2268 if df['format_id'] not in dash_formats:
2269 dash_formats[df['format_id']] = df
2270 # Additional DASH manifests may end up in HTTP Error 403 therefore
2271 # allow them to fail without bug report message if we already have
2272 # some DASH manifest succeeded. This is temporary workaround to reduce
2273 # burst of bug reports until we figure out the reason and whether it
2274 # can be fixed at all.
2275 dash_mpd_fatal = False
2276 except (ExtractorError, KeyError) as e:
2277 self.report_warning(
2278 'Skipping DASH manifest: %r' % e, video_id)
2279 if dash_formats:
2280 # Remove the formats we found through non-DASH, they
2281 # contain less info and it can be wrong, because we use
2282 # fixed values (for example the resolution). See
2283 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2284 # example.
2285 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2286 formats.extend(dash_formats.values())
2287
2288 # Check for malformed aspect ratio
2289 stretched_m = re.search(
2290 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2291 video_webpage)
2292 if stretched_m:
2293 w = float(stretched_m.group('w'))
2294 h = float(stretched_m.group('h'))
2295 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2296 # We will only process correct ratios.
2297 if w > 0 and h > 0:
2298 ratio = w / h
2299 for f in formats:
2300 if f.get('vcodec') != 'none':
2301 f['stretched_ratio'] = ratio
2302
2303 if not formats:
2304 token = extract_token(video_info)
2305 if not token:
2306 if 'reason' in video_info:
2307 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2308 regions_allowed = self._html_search_meta(
2309 'regionsAllowed', video_webpage, default=None)
2310 countries = regions_allowed.split(',') if regions_allowed else None
2311 self.raise_geo_restricted(
2312 msg=video_info['reason'][0], countries=countries)
2313 reason = video_info['reason'][0]
2314 if 'Invalid parameters' in reason:
2315 unavailable_message = extract_unavailable_message()
2316 if unavailable_message:
2317 reason = unavailable_message
2318 raise ExtractorError(
2319 'YouTube said: %s' % reason,
2320 expected=True, video_id=video_id)
2321 else:
2322 raise ExtractorError(
2323 '"token" parameter not in video info for unknown reason',
2324 video_id=video_id)
2325
2326 self._sort_formats(formats)
2327
2328 self.mark_watched(video_id, video_info, player_response)
2329
2330 return {
2331 'id': video_id,
2332 'uploader': video_uploader,
2333 'uploader_id': video_uploader_id,
2334 'uploader_url': video_uploader_url,
2335 'channel_id': channel_id,
2336 'channel_url': channel_url,
2337 'upload_date': upload_date,
2338 'license': video_license,
2339 'creator': video_creator or artist,
2340 'title': video_title,
2341 'alt_title': video_alt_title or track,
2342 'thumbnail': video_thumbnail,
2343 'description': video_description,
2344 'categories': video_categories,
2345 'tags': video_tags,
2346 'subtitles': video_subtitles,
2347 'automatic_captions': automatic_captions,
2348 'duration': video_duration,
2349 'age_limit': 18 if age_gate else 0,
2350 'annotations': video_annotations,
2351 'chapters': chapters,
2352 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2353 'view_count': view_count,
2354 'like_count': like_count,
2355 'dislike_count': dislike_count,
2356 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2357 'formats': formats,
2358 'is_live': is_live,
2359 'start_time': start_time,
2360 'end_time': end_time,
2361 'series': series,
2362 'season_number': season_number,
2363 'episode_number': episode_number,
2364 'track': track,
2365 'artist': artist,
2366 'album': album,
2367 'release_date': release_date,
2368 'release_year': release_year,
2369 }
2370
2371
2372 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2373 IE_DESC = 'YouTube.com playlists'
2374 _VALID_URL = r"""(?x)(?:
2375 (?:https?://)?
2376 (?:\w+\.)?
2377 (?:
2378 (?:
2379 youtube\.com|
2380 invidio\.us
2381 )
2382 /
2383 (?:
2384 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2385 \? (?:.*?[&;])*? (?:p|a|list)=
2386 | p/
2387 )|
2388 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2389 )
2390 (
2391 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2392 # Top tracks, they can also include dots
2393 |(?:MC)[\w\.]*
2394 )
2395 .*
2396 |
2397 (%(playlist_id)s)
2398 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2399 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2400 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2401 IE_NAME = 'youtube:playlist'
2402 _TESTS = [{
2403 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2404 'info_dict': {
2405 'title': 'ytdl test PL',
2406 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2407 },
2408 'playlist_count': 3,
2409 }, {
2410 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2411 'info_dict': {
2412 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2413 'title': 'YDL_Empty_List',
2414 },
2415 'playlist_count': 0,
2416 'skip': 'This playlist is private',
2417 }, {
2418 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2419 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2420 'info_dict': {
2421 'title': '29C3: Not my department',
2422 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2423 },
2424 'playlist_count': 95,
2425 }, {
2426 'note': 'issue #673',
2427 'url': 'PLBB231211A4F62143',
2428 'info_dict': {
2429 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2430 'id': 'PLBB231211A4F62143',
2431 },
2432 'playlist_mincount': 26,
2433 }, {
2434 'note': 'Large playlist',
2435 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2436 'info_dict': {
2437 'title': 'Uploads from Cauchemar',
2438 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2439 },
2440 'playlist_mincount': 799,
2441 }, {
2442 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2443 'info_dict': {
2444 'title': 'YDL_safe_search',
2445 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2446 },
2447 'playlist_count': 2,
2448 'skip': 'This playlist is private',
2449 }, {
2450 'note': 'embedded',
2451 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2452 'playlist_count': 4,
2453 'info_dict': {
2454 'title': 'JODA15',
2455 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2456 }
2457 }, {
2458 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2459 'playlist_mincount': 485,
2460 'info_dict': {
2461 'title': '2017 華語最新單曲 (2/24更新)',
2462 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2463 }
2464 }, {
2465 'note': 'Embedded SWF player',
2466 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2467 'playlist_count': 4,
2468 'info_dict': {
2469 'title': 'JODA7',
2470 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2471 }
2472 }, {
2473 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2474 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2475 'info_dict': {
2476 'title': 'Uploads from Interstellar Movie',
2477 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2478 },
2479 'playlist_mincount': 21,
2480 }, {
2481 # Playlist URL that does not actually serve a playlist
2482 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2483 'info_dict': {
2484 'id': 'FqZTN594JQw',
2485 'ext': 'webm',
2486 'title': "Smiley's People 01 detective, Adventure Series, Action",
2487 'uploader': 'STREEM',
2488 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2489 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2490 'upload_date': '20150526',
2491 'license': 'Standard YouTube License',
2492 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2493 'categories': ['People & Blogs'],
2494 'tags': list,
2495 'view_count': int,
2496 'like_count': int,
2497 'dislike_count': int,
2498 },
2499 'params': {
2500 'skip_download': True,
2501 },
2502 'add_ie': [YoutubeIE.ie_key()],
2503 }, {
2504 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2505 'info_dict': {
2506 'id': 'yeWKywCrFtk',
2507 'ext': 'mp4',
2508 'title': 'Small Scale Baler and Braiding Rugs',
2509 'uploader': 'Backus-Page House Museum',
2510 'uploader_id': 'backuspagemuseum',
2511 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2512 'upload_date': '20161008',
2513 'license': 'Standard YouTube License',
2514 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2515 'categories': ['Nonprofits & Activism'],
2516 'tags': list,
2517 'like_count': int,
2518 'dislike_count': int,
2519 },
2520 'params': {
2521 'noplaylist': True,
2522 'skip_download': True,
2523 },
2524 }, {
2525 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2526 'only_matching': True,
2527 }, {
2528 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2529 'only_matching': True,
2530 }, {
2531 # music album playlist
2532 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2533 'only_matching': True,
2534 }, {
2535 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2536 'only_matching': True,
2537 }]
2538
2539 def _real_initialize(self):
2540 self._login()
2541
2542 def _extract_mix(self, playlist_id):
2543 # The mixes are generated from a single video
2544 # the id of the playlist is just 'RD' + video_id
2545 ids = []
2546 last_id = playlist_id[-11:]
2547 for n in itertools.count(1):
2548 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2549 webpage = self._download_webpage(
2550 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2551 new_ids = orderedSet(re.findall(
2552 r'''(?xs)data-video-username=".*?".*?
2553 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2554 webpage))
2555 # Fetch new pages until all the videos are repeated, it seems that
2556 # there are always 51 unique videos.
2557 new_ids = [_id for _id in new_ids if _id not in ids]
2558 if not new_ids:
2559 break
2560 ids.extend(new_ids)
2561 last_id = ids[-1]
2562
2563 url_results = self._ids_to_results(ids)
2564
2565 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2566 title_span = (
2567 search_title('playlist-title')
2568 or search_title('title long-title')
2569 or search_title('title'))
2570 title = clean_html(title_span)
2571
2572 return self.playlist_result(url_results, playlist_id, title)
2573
2574 def _extract_playlist(self, playlist_id):
2575 url = self._TEMPLATE_URL % playlist_id
2576 page = self._download_webpage(url, playlist_id)
2577
2578 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2579 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2580 match = match.strip()
2581 # Check if the playlist exists or is private
2582 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2583 if mobj:
2584 reason = mobj.group('reason')
2585 message = 'This playlist %s' % reason
2586 if 'private' in reason:
2587 message += ', use --username or --netrc to access it'
2588 message += '.'
2589 raise ExtractorError(message, expected=True)
2590 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2591 raise ExtractorError(
2592 'Invalid parameters. Maybe URL is incorrect.',
2593 expected=True)
2594 elif re.match(r'[^<]*Choose your language[^<]*', match):
2595 continue
2596 else:
2597 self.report_warning('Youtube gives an alert message: ' + match)
2598
2599 playlist_title = self._html_search_regex(
2600 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2601 page, 'title', default=None)
2602
2603 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2604 uploader = self._search_regex(
2605 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2606 page, 'uploader', default=None)
2607 mobj = re.search(
2608 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2609 page)
2610 if mobj:
2611 uploader_id = mobj.group('uploader_id')
2612 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2613 else:
2614 uploader_id = uploader_url = None
2615
2616 has_videos = True
2617
2618 if not playlist_title:
2619 try:
2620 # Some playlist URLs don't actually serve a playlist (e.g.
2621 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2622 next(self._entries(page, playlist_id))
2623 except StopIteration:
2624 has_videos = False
2625
2626 playlist = self.playlist_result(
2627 self._entries(page, playlist_id), playlist_id, playlist_title)
2628 playlist.update({
2629 'uploader': uploader,
2630 'uploader_id': uploader_id,
2631 'uploader_url': uploader_url,
2632 })
2633
2634 return has_videos, playlist
2635
2636 def _check_download_just_video(self, url, playlist_id):
2637 # Check if it's a video-specific URL
2638 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2639 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2640 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2641 'video id', default=None)
2642 if video_id:
2643 if self._downloader.params.get('noplaylist'):
2644 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2645 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2646 else:
2647 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2648 return video_id, None
2649 return None, None
2650
2651 def _real_extract(self, url):
2652 # Extract playlist id
2653 mobj = re.match(self._VALID_URL, url)
2654 if mobj is None:
2655 raise ExtractorError('Invalid URL: %s' % url)
2656 playlist_id = mobj.group(1) or mobj.group(2)
2657
2658 video_id, video = self._check_download_just_video(url, playlist_id)
2659 if video:
2660 return video
2661
2662 if playlist_id.startswith(('RD', 'UL', 'PU')):
2663 # Mixes require a custom extraction process
2664 return self._extract_mix(playlist_id)
2665
2666 has_videos, playlist = self._extract_playlist(playlist_id)
2667 if has_videos or not video_id:
2668 return playlist
2669
2670 # Some playlist URLs don't actually serve a playlist (see
2671 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2672 # Fallback to plain video extraction if there is a video id
2673 # along with playlist id.
2674 return self.url_result(video_id, 'Youtube', video_id=video_id)
2675
2676
2677 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2678 IE_DESC = 'YouTube.com channels'
2679 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2680 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2681 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2682 IE_NAME = 'youtube:channel'
2683 _TESTS = [{
2684 'note': 'paginated channel',
2685 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2686 'playlist_mincount': 91,
2687 'info_dict': {
2688 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2689 'title': 'Uploads from lex will',
2690 }
2691 }, {
2692 'note': 'Age restricted channel',
2693 # from https://www.youtube.com/user/DeusExOfficial
2694 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2695 'playlist_mincount': 64,
2696 'info_dict': {
2697 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2698 'title': 'Uploads from Deus Ex',
2699 },
2700 }, {
2701 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2702 'only_matching': True,
2703 }]
2704
2705 @classmethod
2706 def suitable(cls, url):
2707 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2708 else super(YoutubeChannelIE, cls).suitable(url))
2709
2710 def _build_template_url(self, url, channel_id):
2711 return self._TEMPLATE_URL % channel_id
2712
2713 def _real_extract(self, url):
2714 channel_id = self._match_id(url)
2715
2716 url = self._build_template_url(url, channel_id)
2717
2718 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2719 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2720 # otherwise fallback on channel by page extraction
2721 channel_page = self._download_webpage(
2722 url + '?view=57', channel_id,
2723 'Downloading channel page', fatal=False)
2724 if channel_page is False:
2725 channel_playlist_id = False
2726 else:
2727 channel_playlist_id = self._html_search_meta(
2728 'channelId', channel_page, 'channel id', default=None)
2729 if not channel_playlist_id:
2730 channel_url = self._html_search_meta(
2731 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2732 channel_page, 'channel url', default=None)
2733 if channel_url:
2734 channel_playlist_id = self._search_regex(
2735 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2736 channel_url, 'channel id', default=None)
2737 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2738 playlist_id = 'UU' + channel_playlist_id[2:]
2739 return self.url_result(
2740 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2741
2742 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2743 autogenerated = re.search(r'''(?x)
2744 class="[^"]*?(?:
2745 channel-header-autogenerated-label|
2746 yt-channel-title-autogenerated
2747 )[^"]*"''', channel_page) is not None
2748
2749 if autogenerated:
2750 # The videos are contained in a single page
2751 # the ajax pages can't be used, they are empty
2752 entries = [
2753 self.url_result(
2754 video_id, 'Youtube', video_id=video_id,
2755 video_title=video_title)
2756 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2757 return self.playlist_result(entries, channel_id)
2758
2759 try:
2760 next(self._entries(channel_page, channel_id))
2761 except StopIteration:
2762 alert_message = self._html_search_regex(
2763 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2764 channel_page, 'alert', default=None, group='alert')
2765 if alert_message:
2766 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2767
2768 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2769
2770
2771 class YoutubeUserIE(YoutubeChannelIE):
2772 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2773 _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_-]+)'
2774 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2775 IE_NAME = 'youtube:user'
2776
2777 _TESTS = [{
2778 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2779 'playlist_mincount': 320,
2780 'info_dict': {
2781 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2782 'title': 'Uploads from The Linux Foundation',
2783 }
2784 }, {
2785 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2786 # but not https://www.youtube.com/user/12minuteathlete/videos
2787 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2788 'playlist_mincount': 249,
2789 'info_dict': {
2790 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2791 'title': 'Uploads from 12 Minute Athlete',
2792 }
2793 }, {
2794 'url': 'ytuser:phihag',
2795 'only_matching': True,
2796 }, {
2797 'url': 'https://www.youtube.com/c/gametrailers',
2798 'only_matching': True,
2799 }, {
2800 'url': 'https://www.youtube.com/gametrailers',
2801 'only_matching': True,
2802 }, {
2803 # This channel is not available, geo restricted to JP
2804 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2805 'only_matching': True,
2806 }]
2807
2808 @classmethod
2809 def suitable(cls, url):
2810 # Don't return True if the url can be extracted with other youtube
2811 # extractor, the regex would is too permissive and it would match.
2812 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2813 if any(ie.suitable(url) for ie in other_yt_ies):
2814 return False
2815 else:
2816 return super(YoutubeUserIE, cls).suitable(url)
2817
2818 def _build_template_url(self, url, channel_id):
2819 mobj = re.match(self._VALID_URL, url)
2820 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2821
2822
2823 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2824 IE_DESC = 'YouTube.com live streams'
2825 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2826 IE_NAME = 'youtube:live'
2827
2828 _TESTS = [{
2829 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2830 'info_dict': {
2831 'id': 'a48o2S1cPoo',
2832 'ext': 'mp4',
2833 'title': 'The Young Turks - Live Main Show',
2834 'uploader': 'The Young Turks',
2835 'uploader_id': 'TheYoungTurks',
2836 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2837 'upload_date': '20150715',
2838 'license': 'Standard YouTube License',
2839 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2840 'categories': ['News & Politics'],
2841 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2842 'like_count': int,
2843 'dislike_count': int,
2844 },
2845 'params': {
2846 'skip_download': True,
2847 },
2848 }, {
2849 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2850 'only_matching': True,
2851 }, {
2852 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2853 'only_matching': True,
2854 }, {
2855 'url': 'https://www.youtube.com/TheYoungTurks/live',
2856 'only_matching': True,
2857 }]
2858
2859 def _real_extract(self, url):
2860 mobj = re.match(self._VALID_URL, url)
2861 channel_id = mobj.group('id')
2862 base_url = mobj.group('base_url')
2863 webpage = self._download_webpage(url, channel_id, fatal=False)
2864 if webpage:
2865 page_type = self._og_search_property(
2866 'type', webpage, 'page type', default='')
2867 video_id = self._html_search_meta(
2868 'videoId', webpage, 'video id', default=None)
2869 if page_type.startswith('video') and video_id and re.match(
2870 r'^[0-9A-Za-z_-]{11}$', video_id):
2871 return self.url_result(video_id, YoutubeIE.ie_key())
2872 return self.url_result(base_url)
2873
2874
2875 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2876 IE_DESC = 'YouTube.com user/channel playlists'
2877 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2878 IE_NAME = 'youtube:playlists'
2879
2880 _TESTS = [{
2881 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2882 'playlist_mincount': 4,
2883 'info_dict': {
2884 'id': 'ThirstForScience',
2885 'title': 'Thirst for Science',
2886 },
2887 }, {
2888 # with "Load more" button
2889 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2890 'playlist_mincount': 70,
2891 'info_dict': {
2892 'id': 'igorkle1',
2893 'title': 'Игорь Клейнер',
2894 },
2895 }, {
2896 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2897 'playlist_mincount': 17,
2898 'info_dict': {
2899 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2900 'title': 'Chem Player',
2901 },
2902 }]
2903
2904
2905 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2906 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2907
2908
2909 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2910 IE_DESC = 'YouTube.com searches'
2911 # there doesn't appear to be a real limit, for example if you search for
2912 # 'python' you get more than 8.000.000 results
2913 _MAX_RESULTS = float('inf')
2914 IE_NAME = 'youtube:search'
2915 _SEARCH_KEY = 'ytsearch'
2916 _EXTRA_QUERY_ARGS = {}
2917 _TESTS = []
2918
2919 def _get_n_results(self, query, n):
2920 """Get a specified number of results for a query"""
2921
2922 videos = []
2923 limit = n
2924
2925 url_query = {
2926 'search_query': query.encode('utf-8'),
2927 }
2928 url_query.update(self._EXTRA_QUERY_ARGS)
2929 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2930
2931 for pagenum in itertools.count(1):
2932 data = self._download_json(
2933 result_url, video_id='query "%s"' % query,
2934 note='Downloading page %s' % pagenum,
2935 errnote='Unable to download API page',
2936 query={'spf': 'navigate'})
2937 html_content = data[1]['body']['content']
2938
2939 if 'class="search-message' in html_content:
2940 raise ExtractorError(
2941 '[youtube] No video results', expected=True)
2942
2943 new_videos = list(self._process_page(html_content))
2944 videos += new_videos
2945 if not new_videos or len(videos) > limit:
2946 break
2947 next_link = self._html_search_regex(
2948 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2949 html_content, 'next link', default=None)
2950 if next_link is None:
2951 break
2952 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2953
2954 if len(videos) > n:
2955 videos = videos[:n]
2956 return self.playlist_result(videos, query)
2957
2958
2959 class YoutubeSearchDateIE(YoutubeSearchIE):
2960 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2961 _SEARCH_KEY = 'ytsearchdate'
2962 IE_DESC = 'YouTube.com searches, newest videos first'
2963 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2964
2965
2966 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2967 IE_DESC = 'YouTube.com search URLs'
2968 IE_NAME = 'youtube:search_url'
2969 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2970 _TESTS = [{
2971 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2972 'playlist_mincount': 5,
2973 'info_dict': {
2974 'title': 'youtube-dl test video',
2975 }
2976 }, {
2977 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2978 'only_matching': True,
2979 }]
2980
2981 def _real_extract(self, url):
2982 mobj = re.match(self._VALID_URL, url)
2983 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2984 webpage = self._download_webpage(url, query)
2985 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2986
2987
2988 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2989 IE_DESC = 'YouTube.com (multi-season) shows'
2990 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2991 IE_NAME = 'youtube:show'
2992 _TESTS = [{
2993 'url': 'https://www.youtube.com/show/airdisasters',
2994 'playlist_mincount': 5,
2995 'info_dict': {
2996 'id': 'airdisasters',
2997 'title': 'Air Disasters',
2998 }
2999 }]
3000
3001 def _real_extract(self, url):
3002 playlist_id = self._match_id(url)
3003 return super(YoutubeShowIE, self)._real_extract(
3004 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3005
3006
3007 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3008 """
3009 Base class for feed extractors
3010 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3011 """
3012 _LOGIN_REQUIRED = True
3013
3014 @property
3015 def IE_NAME(self):
3016 return 'youtube:%s' % self._FEED_NAME
3017
3018 def _real_initialize(self):
3019 self._login()
3020
3021 def _entries(self, page):
3022 # The extraction process is the same as for playlists, but the regex
3023 # for the video ids doesn't contain an index
3024 ids = []
3025 more_widget_html = content_html = page
3026 for page_num in itertools.count(1):
3027 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3028
3029 # 'recommended' feed has infinite 'load more' and each new portion spins
3030 # the same videos in (sometimes) slightly different order, so we'll check
3031 # for unicity and break when portion has no new videos
3032 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3033 if not new_ids:
3034 break
3035
3036 ids.extend(new_ids)
3037
3038 for entry in self._ids_to_results(new_ids):
3039 yield entry
3040
3041 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3042 if not mobj:
3043 break
3044
3045 more = self._download_json(
3046 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3047 'Downloading page #%s' % page_num,
3048 transform_source=uppercase_escape)
3049 content_html = more['content_html']
3050 more_widget_html = more['load_more_widget_html']
3051
3052 def _real_extract(self, url):
3053 page = self._download_webpage(
3054 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3055 self._PLAYLIST_TITLE)
3056 return self.playlist_result(
3057 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3058
3059
3060 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3061 IE_NAME = 'youtube:watchlater'
3062 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3063 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3064
3065 _TESTS = [{
3066 'url': 'https://www.youtube.com/playlist?list=WL',
3067 'only_matching': True,
3068 }, {
3069 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3070 'only_matching': True,
3071 }]
3072
3073 def _real_extract(self, url):
3074 _, video = self._check_download_just_video(url, 'WL')
3075 if video:
3076 return video
3077 _, playlist = self._extract_playlist('WL')
3078 return playlist
3079
3080
3081 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3082 IE_NAME = 'youtube:favorites'
3083 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3084 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3085 _LOGIN_REQUIRED = True
3086
3087 def _real_extract(self, url):
3088 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3089 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3090 return self.url_result(playlist_id, 'YoutubePlaylist')
3091
3092
3093 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3094 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3095 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3096 _FEED_NAME = 'recommended'
3097 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3098
3099
3100 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3101 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3102 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3103 _FEED_NAME = 'subscriptions'
3104 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3105
3106
3107 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3108 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3109 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3110 _FEED_NAME = 'history'
3111 _PLAYLIST_TITLE = 'Youtube History'
3112
3113
3114 class YoutubeTruncatedURLIE(InfoExtractor):
3115 IE_NAME = 'youtube:truncated_url'
3116 IE_DESC = False # Do not list
3117 _VALID_URL = r'''(?x)
3118 (?:https?://)?
3119 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3120 (?:watch\?(?:
3121 feature=[a-z_]+|
3122 annotation_id=annotation_[^&]+|
3123 x-yt-cl=[0-9]+|
3124 hl=[^&]*|
3125 t=[0-9]+
3126 )?
3127 |
3128 attribution_link\?a=[^&]+
3129 )
3130 $
3131 '''
3132
3133 _TESTS = [{
3134 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3135 'only_matching': True,
3136 }, {
3137 'url': 'https://www.youtube.com/watch?',
3138 'only_matching': True,
3139 }, {
3140 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3141 'only_matching': True,
3142 }, {
3143 'url': 'https://www.youtube.com/watch?feature=foo',
3144 'only_matching': True,
3145 }, {
3146 'url': 'https://www.youtube.com/watch?hl=en-GB',
3147 'only_matching': True,
3148 }, {
3149 'url': 'https://www.youtube.com/watch?t=2372',
3150 'only_matching': True,
3151 }]
3152
3153 def _real_extract(self, url):
3154 raise ExtractorError(
3155 'Did you forget to quote the URL? Remember that & is a meta '
3156 'character in most shells, so you want to put the URL in quotes, '
3157 'like youtube-dl '
3158 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3159 ' or simply youtube-dl BaW_jenozKc .',
3160 expected=True)
3161
3162
3163 class YoutubeTruncatedIDIE(InfoExtractor):
3164 IE_NAME = 'youtube:truncated_id'
3165 IE_DESC = False # Do not list
3166 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3167
3168 _TESTS = [{
3169 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3170 'only_matching': True,
3171 }]
3172
3173 def _real_extract(self, url):
3174 video_id = self._match_id(url)
3175 raise ExtractorError(
3176 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3177 expected=True)