]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
ce2212a7caa13111e1d2835fa65be371a4937847
[youtubedl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_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 bool_or_none,
31 clean_html,
32 error_to_compat_str,
33 extract_attributes,
34 ExtractorError,
35 float_or_none,
36 get_element_by_attribute,
37 get_element_by_id,
38 int_or_none,
39 mimetype2ext,
40 orderedSet,
41 parse_codecs,
42 parse_duration,
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|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}'
72
73 _YOUTUBE_CLIENT_HEADERS = {
74 'x-youtube-client-name': '1',
75 'x-youtube-client-version': '1.20200609.04.02',
76 }
77
78 def _set_language(self):
79 self._set_cookie(
80 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
81 # YouTube sets the expire time to about two months
82 expire_time=time.time() + 2 * 30 * 24 * 3600)
83
84 def _ids_to_results(self, ids):
85 return [
86 self.url_result(vid_id, 'Youtube', video_id=vid_id)
87 for vid_id in ids]
88
89 def _login(self):
90 """
91 Attempt to log in to YouTube.
92 True is returned if successful or skipped.
93 False is returned if login failed.
94
95 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
96 """
97 username, password = self._get_login_info()
98 # No authentication to be performed
99 if username is None:
100 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
101 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
102 return True
103
104 login_page = self._download_webpage(
105 self._LOGIN_URL, None,
106 note='Downloading login page',
107 errnote='unable to fetch login page', fatal=False)
108 if login_page is False:
109 return
110
111 login_form = self._hidden_inputs(login_page)
112
113 def req(url, f_req, note, errnote):
114 data = login_form.copy()
115 data.update({
116 'pstMsg': 1,
117 'checkConnection': 'youtube',
118 'checkedDomains': 'youtube',
119 'hl': 'en',
120 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
121 'f.req': json.dumps(f_req),
122 'flowName': 'GlifWebSignIn',
123 'flowEntry': 'ServiceLogin',
124 # TODO: reverse actual botguard identifier generation algo
125 'bgRequest': '["identifier",""]',
126 })
127 return self._download_json(
128 url, None, note=note, errnote=errnote,
129 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
130 fatal=False,
131 data=urlencode_postdata(data), headers={
132 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
133 'Google-Accounts-XSRF': 1,
134 })
135
136 def warn(message):
137 self._downloader.report_warning(message)
138
139 lookup_req = [
140 username,
141 None, [], None, 'US', None, None, 2, False, True,
142 [
143 None, None,
144 [2, 1, None, 1,
145 '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',
146 None, [], 4],
147 1, [None, None, []], None, None, None, True
148 ],
149 username,
150 ]
151
152 lookup_results = req(
153 self._LOOKUP_URL, lookup_req,
154 'Looking up account info', 'Unable to look up account info')
155
156 if lookup_results is False:
157 return False
158
159 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
160 if not user_hash:
161 warn('Unable to extract user hash')
162 return False
163
164 challenge_req = [
165 user_hash,
166 None, 1, None, [1, None, None, None, [password, None, True]],
167 [
168 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],
169 1, [None, None, []], None, None, None, True
170 ]]
171
172 challenge_results = req(
173 self._CHALLENGE_URL, challenge_req,
174 'Logging in', 'Unable to log in')
175
176 if challenge_results is False:
177 return
178
179 login_res = try_get(challenge_results, lambda x: x[0][5], list)
180 if login_res:
181 login_msg = try_get(login_res, lambda x: x[5], compat_str)
182 warn(
183 'Unable to login: %s' % 'Invalid password'
184 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
185 return False
186
187 res = try_get(challenge_results, lambda x: x[0][-1], list)
188 if not res:
189 warn('Unable to extract result entry')
190 return False
191
192 login_challenge = try_get(res, lambda x: x[0][0], list)
193 if login_challenge:
194 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
195 if challenge_str == 'TWO_STEP_VERIFICATION':
196 # SEND_SUCCESS - TFA code has been successfully sent to phone
197 # QUOTA_EXCEEDED - reached the limit of TFA codes
198 status = try_get(login_challenge, lambda x: x[5], compat_str)
199 if status == 'QUOTA_EXCEEDED':
200 warn('Exceeded the limit of TFA codes, try later')
201 return False
202
203 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
204 if not tl:
205 warn('Unable to extract TL')
206 return False
207
208 tfa_code = self._get_tfa_info('2-step verification code')
209
210 if not tfa_code:
211 warn(
212 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
213 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
214 return False
215
216 tfa_code = remove_start(tfa_code, 'G-')
217
218 tfa_req = [
219 user_hash, None, 2, None,
220 [
221 9, None, None, None, None, None, None, None,
222 [None, tfa_code, True, 2]
223 ]]
224
225 tfa_results = req(
226 self._TFA_URL.format(tl), tfa_req,
227 'Submitting TFA code', 'Unable to submit TFA code')
228
229 if tfa_results is False:
230 return False
231
232 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
233 if tfa_res:
234 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
235 warn(
236 'Unable to finish TFA: %s' % 'Invalid TFA code'
237 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
238 return False
239
240 check_cookie_url = try_get(
241 tfa_results, lambda x: x[0][-1][2], compat_str)
242 else:
243 CHALLENGES = {
244 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
245 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
246 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
247 }
248 challenge = CHALLENGES.get(
249 challenge_str,
250 '%s returned error %s.' % (self.IE_NAME, challenge_str))
251 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
252 return False
253 else:
254 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
255
256 if not check_cookie_url:
257 warn('Unable to extract CheckCookie URL')
258 return False
259
260 check_cookie_results = self._download_webpage(
261 check_cookie_url, None, 'Checking cookie', fatal=False)
262
263 if check_cookie_results is False:
264 return False
265
266 if 'https://myaccount.google.com/' not in check_cookie_results:
267 warn('Unable to log in')
268 return False
269
270 return True
271
272 def _download_webpage_handle(self, *args, **kwargs):
273 query = kwargs.get('query', {}).copy()
274 query['disable_polymer'] = 'true'
275 kwargs['query'] = query
276 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
277 *args, **compat_kwargs(kwargs))
278
279 def _real_initialize(self):
280 if self._downloader is None:
281 return
282 self._set_language()
283 if not self._login():
284 return
285
286
287 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
288 # Extract entries from page with "Load more" button
289 def _entries(self, page, playlist_id):
290 more_widget_html = content_html = page
291 for page_num in itertools.count(1):
292 for entry in self._process_page(content_html):
293 yield entry
294
295 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
296 if not mobj:
297 break
298
299 count = 0
300 retries = 3
301 while count <= retries:
302 try:
303 # Downloading page may result in intermittent 5xx HTTP error
304 # that is usually worked around with a retry
305 more = self._download_json(
306 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
307 'Downloading page #%s%s'
308 % (page_num, ' (retry #%d)' % count if count else ''),
309 transform_source=uppercase_escape,
310 headers=self._YOUTUBE_CLIENT_HEADERS)
311 break
312 except ExtractorError as e:
313 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
314 count += 1
315 if count <= retries:
316 continue
317 raise
318
319 content_html = more['content_html']
320 if not content_html.strip():
321 # Some webpages show a "Load more" button but they don't
322 # have more videos
323 break
324 more_widget_html = more['load_more_widget_html']
325
326
327 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
328 def _process_page(self, content):
329 for video_id, video_title in self.extract_videos_from_page(content):
330 yield self.url_result(video_id, 'Youtube', video_id, video_title)
331
332 def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page):
333 for mobj in re.finditer(video_re, page):
334 # The link with index 0 is not the first video of the playlist (not sure if still actual)
335 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
336 continue
337 video_id = mobj.group('id')
338 video_title = unescapeHTML(
339 mobj.group('title')) if 'title' in mobj.groupdict() else None
340 if video_title:
341 video_title = video_title.strip()
342 if video_title == 'ā–ŗ Play all':
343 video_title = None
344 try:
345 idx = ids_in_page.index(video_id)
346 if video_title and not titles_in_page[idx]:
347 titles_in_page[idx] = video_title
348 except ValueError:
349 ids_in_page.append(video_id)
350 titles_in_page.append(video_title)
351
352 def extract_videos_from_page(self, page):
353 ids_in_page = []
354 titles_in_page = []
355 self.extract_videos_from_page_impl(
356 self._VIDEO_RE, page, ids_in_page, titles_in_page)
357 return zip(ids_in_page, titles_in_page)
358
359
360 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
361 def _process_page(self, content):
362 for playlist_id in orderedSet(re.findall(
363 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
364 content)):
365 yield self.url_result(
366 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
367
368 def _real_extract(self, url):
369 playlist_id = self._match_id(url)
370 webpage = self._download_webpage(url, playlist_id)
371 title = self._og_search_title(webpage, fatal=False)
372 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
373
374
375 class YoutubeIE(YoutubeBaseInfoExtractor):
376 IE_DESC = 'YouTube.com'
377 _VALID_URL = r"""(?x)^
378 (
379 (?:https?://|//) # http(s):// or protocol-independent URL
380 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/|
381 (?:www\.)?deturl\.com/www\.youtube\.com/|
382 (?:www\.)?pwnyoutube\.com/|
383 (?:www\.)?hooktube\.com/|
384 (?:www\.)?yourepeat\.com/|
385 tube\.majestyc\.net/|
386 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
387 (?:(?:www|dev)\.)?invidio\.us/|
388 (?:(?:www|no)\.)?invidiou\.sh/|
389 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
390 (?:www\.)?invidious\.kabi\.tk/|
391 (?:www\.)?invidious\.13ad\.de/|
392 (?:www\.)?invidious\.mastodon\.host/|
393 (?:www\.)?invidious\.nixnet\.xyz/|
394 (?:www\.)?invidious\.drycat\.fr/|
395 (?:www\.)?tube\.poal\.co/|
396 (?:www\.)?vid\.wxzm\.sx/|
397 (?:www\.)?yewtu\.be/|
398 (?:www\.)?yt\.elukerio\.org/|
399 (?:www\.)?yt\.lelux\.fi/|
400 (?:www\.)?invidious\.ggc-project\.de/|
401 (?:www\.)?yt\.maisputain\.ovh/|
402 (?:www\.)?invidious\.13ad\.de/|
403 (?:www\.)?invidious\.toot\.koeln/|
404 (?:www\.)?invidious\.fdn\.fr/|
405 (?:www\.)?watch\.nettohikari\.com/|
406 (?:www\.)?kgg2m7yk5aybusll\.onion/|
407 (?:www\.)?qklhadlycap4cnod\.onion/|
408 (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/|
409 (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/|
410 (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/|
411 (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/|
412 (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/|
413 (?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion/|
414 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
415 (?:.*?\#/)? # handle anchor (#/) redirect urls
416 (?: # the various things that can precede the ID:
417 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
418 |(?: # or the v= param in all its forms
419 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
420 (?:\?|\#!?) # the params delimiter ? or # or #!
421 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
422 v=
423 )
424 ))
425 |(?:
426 youtu\.be| # just youtu.be/xxxx
427 vid\.plus| # or vid.plus/xxxx
428 zwearz\.com/watch| # or zwearz.com/watch/xxxx
429 )/
430 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
431 )
432 )? # all until now is optional -> you can pass the naked ID
433 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
434 (?!.*?\blist=
435 (?:
436 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
437 WL # WL are handled by the watch later IE
438 )
439 )
440 (?(1).+)? # if we found the ID, everything can follow
441 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
442 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
443 _PLAYER_INFO_RE = (
444 r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.(?P<ext>[a-z]+)$',
445 r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.(?P<ext>[a-z]+)$',
446 )
447 _formats = {
448 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
449 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
450 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
451 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
452 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
453 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
454 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
455 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
456 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
457 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
458 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
459 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
460 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
461 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
462 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
463 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
464 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
465 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
466
467
468 # 3D videos
469 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
470 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
471 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
472 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
473 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
474 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
475 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
476
477 # Apple HTTP Live Streaming
478 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
479 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
480 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
481 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
482 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
483 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
484 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
485 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
486
487 # DASH mp4 video
488 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
489 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
490 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
491 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
492 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
493 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
494 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
495 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
496 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
497 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
498 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
499 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
500
501 # Dash mp4 audio
502 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
503 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
504 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
505 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
506 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
507 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
508 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
509
510 # Dash webm
511 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
512 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
513 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
514 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
515 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
516 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
517 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
518 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
519 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
520 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
521 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
522 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
523 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
524 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
525 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
526 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
527 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
528 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
529 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
530 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
531 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
532 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
533
534 # Dash webm audio
535 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
536 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
537
538 # Dash webm audio with opus inside
539 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
540 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
541 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
542
543 # RTMP (unnamed)
544 '_rtmp': {'protocol': 'rtmp'},
545
546 # av01 video only formats sometimes served with "unknown" codecs
547 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
548 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
549 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
550 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
551 }
552 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
553
554 _GEO_BYPASS = False
555
556 IE_NAME = 'youtube'
557 _TESTS = [
558 {
559 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
560 'info_dict': {
561 'id': 'BaW_jenozKc',
562 'ext': 'mp4',
563 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
564 'uploader': 'Philipp Hagemeister',
565 'uploader_id': 'phihag',
566 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
567 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
568 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
569 'upload_date': '20121002',
570 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
571 'categories': ['Science & Technology'],
572 'tags': ['youtube-dl'],
573 'duration': 10,
574 'view_count': int,
575 'like_count': int,
576 'dislike_count': int,
577 'start_time': 1,
578 'end_time': 9,
579 }
580 },
581 {
582 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
583 'note': 'Test generic use_cipher_signature video (#897)',
584 'info_dict': {
585 'id': 'UxxajLWwzqY',
586 'ext': 'mp4',
587 'upload_date': '20120506',
588 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
589 'alt_title': 'I Love It (feat. Charli XCX)',
590 'description': 'md5:19a2f98d9032b9311e686ed039564f63',
591 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
592 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
593 'iconic ep', 'iconic', 'love', 'it'],
594 'duration': 180,
595 'uploader': 'Icona Pop',
596 'uploader_id': 'IconaPop',
597 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
598 'creator': 'Icona Pop',
599 'track': 'I Love It (feat. Charli XCX)',
600 'artist': 'Icona Pop',
601 }
602 },
603 {
604 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
605 'note': 'Test VEVO video with age protection (#956)',
606 'info_dict': {
607 'id': '07FYdnEawAQ',
608 'ext': 'mp4',
609 'upload_date': '20130703',
610 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
611 'alt_title': 'Tunnel Vision',
612 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
613 'duration': 419,
614 'uploader': 'justintimberlakeVEVO',
615 'uploader_id': 'justintimberlakeVEVO',
616 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
617 'creator': 'Justin Timberlake',
618 'track': 'Tunnel Vision',
619 'artist': 'Justin Timberlake',
620 'age_limit': 18,
621 }
622 },
623 {
624 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
625 'note': 'Embed-only video (#1746)',
626 'info_dict': {
627 'id': 'yZIXLfi8CZQ',
628 'ext': 'mp4',
629 'upload_date': '20120608',
630 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
631 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
632 'uploader': 'SET India',
633 'uploader_id': 'setindia',
634 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
635 'age_limit': 18,
636 }
637 },
638 {
639 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
640 'note': 'Use the first video ID in the URL',
641 'info_dict': {
642 'id': 'BaW_jenozKc',
643 'ext': 'mp4',
644 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
645 'uploader': 'Philipp Hagemeister',
646 'uploader_id': 'phihag',
647 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
648 'upload_date': '20121002',
649 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
650 'categories': ['Science & Technology'],
651 'tags': ['youtube-dl'],
652 'duration': 10,
653 'view_count': int,
654 'like_count': int,
655 'dislike_count': int,
656 },
657 'params': {
658 'skip_download': True,
659 },
660 },
661 {
662 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
663 'note': '256k DASH audio (format 141) via DASH manifest',
664 'info_dict': {
665 'id': 'a9LDPn-MO4I',
666 'ext': 'm4a',
667 'upload_date': '20121002',
668 'uploader_id': '8KVIDEO',
669 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
670 'description': '',
671 'uploader': '8KVIDEO',
672 'title': 'UHDTV TEST 8K VIDEO.mp4'
673 },
674 'params': {
675 'youtube_include_dash_manifest': True,
676 'format': '141',
677 },
678 'skip': 'format 141 not served anymore',
679 },
680 # DASH manifest with encrypted signature
681 {
682 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
683 'info_dict': {
684 'id': 'IB3lcPjvWLA',
685 'ext': 'm4a',
686 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
687 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
688 'duration': 244,
689 'uploader': 'AfrojackVEVO',
690 'uploader_id': 'AfrojackVEVO',
691 'upload_date': '20131011',
692 },
693 'params': {
694 'youtube_include_dash_manifest': True,
695 'format': '141/bestaudio[ext=m4a]',
696 },
697 },
698 # JS player signature function name containing $
699 {
700 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
701 'info_dict': {
702 'id': 'nfWlot6h_JM',
703 'ext': 'm4a',
704 'title': 'Taylor Swift - Shake It Off',
705 'description': 'md5:307195cd21ff7fa352270fe884570ef0',
706 'duration': 242,
707 'uploader': 'TaylorSwiftVEVO',
708 'uploader_id': 'TaylorSwiftVEVO',
709 'upload_date': '20140818',
710 },
711 'params': {
712 'youtube_include_dash_manifest': True,
713 'format': '141/bestaudio[ext=m4a]',
714 },
715 },
716 # Controversy video
717 {
718 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
719 'info_dict': {
720 'id': 'T4XJQO3qol8',
721 'ext': 'mp4',
722 'duration': 219,
723 'upload_date': '20100909',
724 'uploader': 'Amazing Atheist',
725 'uploader_id': 'TheAmazingAtheist',
726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
727 'title': 'Burning Everyone\'s Koran',
728 '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',
729 }
730 },
731 # Normal age-gate video (No vevo, embed allowed)
732 {
733 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
734 'info_dict': {
735 'id': 'HtVdAasjOgU',
736 'ext': 'mp4',
737 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
738 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
739 'duration': 142,
740 'uploader': 'The Witcher',
741 'uploader_id': 'WitcherGame',
742 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
743 'upload_date': '20140605',
744 'age_limit': 18,
745 },
746 },
747 # Age-gate video with encrypted signature
748 {
749 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
750 'info_dict': {
751 'id': '6kLq3WMV1nU',
752 'ext': 'mp4',
753 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
754 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
755 'duration': 246,
756 'uploader': 'LloydVEVO',
757 'uploader_id': 'LloydVEVO',
758 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
759 'upload_date': '20110629',
760 'age_limit': 18,
761 },
762 },
763 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
764 # YouTube Red ad is not captured for creator
765 {
766 'url': '__2ABJjxzNo',
767 'info_dict': {
768 'id': '__2ABJjxzNo',
769 'ext': 'mp4',
770 'duration': 266,
771 'upload_date': '20100430',
772 'uploader_id': 'deadmau5',
773 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
774 'creator': 'Dada Life, deadmau5',
775 'description': 'md5:12c56784b8032162bb936a5f76d55360',
776 'uploader': 'deadmau5',
777 'title': 'Deadmau5 - Some Chords (HD)',
778 'alt_title': 'This Machine Kills Some Chords',
779 },
780 'expected_warnings': [
781 'DASH manifest missing',
782 ]
783 },
784 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
785 {
786 'url': 'lqQg6PlCWgI',
787 'info_dict': {
788 'id': 'lqQg6PlCWgI',
789 'ext': 'mp4',
790 'duration': 6085,
791 'upload_date': '20150827',
792 'uploader_id': 'olympic',
793 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
794 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
795 'uploader': 'Olympic',
796 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
797 },
798 'params': {
799 'skip_download': 'requires avconv',
800 }
801 },
802 # Non-square pixels
803 {
804 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
805 'info_dict': {
806 'id': '_b-2C3KPAM0',
807 'ext': 'mp4',
808 'stretched_ratio': 16 / 9.,
809 'duration': 85,
810 'upload_date': '20110310',
811 'uploader_id': 'AllenMeow',
812 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
813 'description': 'made by Wacom from Korea | 字幕&åŠ ę²¹ę·»é†‹ by TY\'s Allen | ę„Ÿč¬heylisa00cavey1001同å­øē†±ęƒ…ęä¾›ę¢—åŠēæ»č­Æ',
814 'uploader': 'å­«į„‹į„…',
815 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
816 },
817 },
818 # url_encoded_fmt_stream_map is empty string
819 {
820 'url': 'qEJwOuvDf7I',
821 'info_dict': {
822 'id': 'qEJwOuvDf7I',
823 'ext': 'webm',
824 'title': 'ŠžŠ±ŃŃƒŠ¶Š“ŠµŠ½ŠøŠµ суŠ“ŠµŠ±Š½Š¾Š¹ ŠæрŠ°ŠŗтŠøŠŗŠø ŠæŠ¾ Š²Ń‹Š±Š¾Ń€Š°Š¼ 14 сŠµŠ½Ń‚яŠ±Ń€Ń 2014 Š³Š¾Š“Š° Š² Š”Š°Š½Šŗт-ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Šµ',
825 'description': '',
826 'upload_date': '20150404',
827 'uploader_id': 'spbelect',
828 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
829 },
830 'params': {
831 'skip_download': 'requires avconv',
832 },
833 'skip': 'This live event has ended.',
834 },
835 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
836 {
837 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
838 'info_dict': {
839 'id': 'FIl7x6_3R5Y',
840 'ext': 'webm',
841 'title': 'md5:7b81415841e02ecd4313668cde88737a',
842 'description': 'md5:116377fd2963b81ec4ce64b542173306',
843 'duration': 220,
844 'upload_date': '20150625',
845 'uploader_id': 'dorappi2000',
846 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
847 'uploader': 'dorappi2000',
848 'formats': 'mincount:31',
849 },
850 'skip': 'not actual anymore',
851 },
852 # DASH manifest with segment_list
853 {
854 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
855 'md5': '8ce563a1d667b599d21064e982ab9e31',
856 'info_dict': {
857 'id': 'CsmdDsKjzN8',
858 'ext': 'mp4',
859 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
860 'uploader': 'Airtek',
861 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
862 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
863 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
864 },
865 'params': {
866 'youtube_include_dash_manifest': True,
867 'format': '135', # bestvideo
868 },
869 'skip': 'This live event has ended.',
870 },
871 {
872 # Multifeed videos (multiple cameras), URL is for Main Camera
873 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
874 'info_dict': {
875 'id': 'jqWvoWXjCVs',
876 'title': 'teamPGP: Rocket League Noob Stream',
877 'description': 'md5:dc7872fb300e143831327f1bae3af010',
878 },
879 'playlist': [{
880 'info_dict': {
881 'id': 'jqWvoWXjCVs',
882 'ext': 'mp4',
883 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
884 'description': 'md5:dc7872fb300e143831327f1bae3af010',
885 'duration': 7335,
886 'upload_date': '20150721',
887 'uploader': 'Beer Games Beer',
888 'uploader_id': 'beergamesbeer',
889 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
890 'license': 'Standard YouTube License',
891 },
892 }, {
893 'info_dict': {
894 'id': '6h8e8xoXJzg',
895 'ext': 'mp4',
896 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
897 'description': 'md5:dc7872fb300e143831327f1bae3af010',
898 'duration': 7337,
899 'upload_date': '20150721',
900 'uploader': 'Beer Games Beer',
901 'uploader_id': 'beergamesbeer',
902 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
903 'license': 'Standard YouTube License',
904 },
905 }, {
906 'info_dict': {
907 'id': 'PUOgX5z9xZw',
908 'ext': 'mp4',
909 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
910 'description': 'md5:dc7872fb300e143831327f1bae3af010',
911 'duration': 7337,
912 'upload_date': '20150721',
913 'uploader': 'Beer Games Beer',
914 'uploader_id': 'beergamesbeer',
915 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
916 'license': 'Standard YouTube License',
917 },
918 }, {
919 'info_dict': {
920 'id': 'teuwxikvS5k',
921 'ext': 'mp4',
922 'title': 'teamPGP: Rocket League Noob Stream (zim)',
923 'description': 'md5:dc7872fb300e143831327f1bae3af010',
924 'duration': 7334,
925 'upload_date': '20150721',
926 'uploader': 'Beer Games Beer',
927 'uploader_id': 'beergamesbeer',
928 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
929 'license': 'Standard YouTube License',
930 },
931 }],
932 'params': {
933 'skip_download': True,
934 },
935 'skip': 'This video is not available.',
936 },
937 {
938 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
939 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
940 'info_dict': {
941 'id': 'gVfLd0zydlo',
942 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
943 },
944 'playlist_count': 2,
945 'skip': 'Not multifeed anymore',
946 },
947 {
948 'url': 'https://vid.plus/FlRa-iH7PGw',
949 'only_matching': True,
950 },
951 {
952 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
953 'only_matching': True,
954 },
955 {
956 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
957 # Also tests cut-off URL expansion in video description (see
958 # https://github.com/ytdl-org/youtube-dl/issues/1892,
959 # https://github.com/ytdl-org/youtube-dl/issues/8164)
960 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
961 'info_dict': {
962 'id': 'lsguqyKfVQg',
963 'ext': 'mp4',
964 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
965 'alt_title': 'Dark Walk - Position Music',
966 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
967 'duration': 133,
968 'upload_date': '20151119',
969 'uploader_id': 'IronSoulElf',
970 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
971 'uploader': 'IronSoulElf',
972 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
973 'track': 'Dark Walk - Position Music',
974 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
975 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
976 },
977 'params': {
978 'skip_download': True,
979 },
980 },
981 {
982 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
983 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
984 'only_matching': True,
985 },
986 {
987 # Video with yt:stretch=17:0
988 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
989 'info_dict': {
990 'id': 'Q39EVAstoRM',
991 'ext': 'mp4',
992 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
993 'description': 'md5:ee18a25c350637c8faff806845bddee9',
994 'upload_date': '20151107',
995 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
996 'uploader': 'CH GAMER DROID',
997 },
998 'params': {
999 'skip_download': True,
1000 },
1001 'skip': 'This video does not exist.',
1002 },
1003 {
1004 # Video licensed under Creative Commons
1005 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
1006 'info_dict': {
1007 'id': 'M4gD1WSo5mA',
1008 'ext': 'mp4',
1009 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
1010 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
1011 'duration': 721,
1012 'upload_date': '20150127',
1013 'uploader_id': 'BerkmanCenter',
1014 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
1015 'uploader': 'The Berkman Klein Center for Internet & Society',
1016 'license': 'Creative Commons Attribution license (reuse allowed)',
1017 },
1018 'params': {
1019 'skip_download': True,
1020 },
1021 },
1022 {
1023 # Channel-like uploader_url
1024 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1025 'info_dict': {
1026 'id': 'eQcmzGIKrzg',
1027 'ext': 'mp4',
1028 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1029 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
1030 'duration': 4060,
1031 'upload_date': '20151119',
1032 'uploader': 'Bernie Sanders',
1033 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1034 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1035 'license': 'Creative Commons Attribution license (reuse allowed)',
1036 },
1037 'params': {
1038 'skip_download': True,
1039 },
1040 },
1041 {
1042 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1043 'only_matching': True,
1044 },
1045 {
1046 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1047 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1048 'only_matching': True,
1049 },
1050 {
1051 # Rental video preview
1052 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1053 'info_dict': {
1054 'id': 'uGpuVWrhIzE',
1055 'ext': 'mp4',
1056 'title': 'Piku - Trailer',
1057 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1058 'upload_date': '20150811',
1059 'uploader': 'FlixMatrix',
1060 'uploader_id': 'FlixMatrixKaravan',
1061 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1062 'license': 'Standard YouTube License',
1063 },
1064 'params': {
1065 'skip_download': True,
1066 },
1067 'skip': 'This video is not available.',
1068 },
1069 {
1070 # YouTube Red video with episode data
1071 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1072 'info_dict': {
1073 'id': 'iqKdEhx-dD4',
1074 'ext': 'mp4',
1075 'title': 'Isolation - Mind Field (Ep 1)',
1076 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1077 'duration': 2085,
1078 'upload_date': '20170118',
1079 'uploader': 'Vsauce',
1080 'uploader_id': 'Vsauce',
1081 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1082 'series': 'Mind Field',
1083 'season_number': 1,
1084 'episode_number': 1,
1085 },
1086 'params': {
1087 'skip_download': True,
1088 },
1089 'expected_warnings': [
1090 'Skipping DASH manifest',
1091 ],
1092 },
1093 {
1094 # The following content has been identified by the YouTube community
1095 # as inappropriate or offensive to some audiences.
1096 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1097 'info_dict': {
1098 'id': '6SJNVb0GnPI',
1099 'ext': 'mp4',
1100 'title': 'Race Differences in Intelligence',
1101 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1102 'duration': 965,
1103 'upload_date': '20140124',
1104 'uploader': 'New Century Foundation',
1105 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1106 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1107 },
1108 'params': {
1109 'skip_download': True,
1110 },
1111 },
1112 {
1113 # itag 212
1114 'url': '1t24XAntNCY',
1115 'only_matching': True,
1116 },
1117 {
1118 # geo restricted to JP
1119 'url': 'sJL6WA-aGkQ',
1120 'only_matching': True,
1121 },
1122 {
1123 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1124 'only_matching': True,
1125 },
1126 {
1127 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1128 'only_matching': True,
1129 },
1130 {
1131 # DRM protected
1132 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1133 'only_matching': True,
1134 },
1135 {
1136 # Video with unsupported adaptive stream type formats
1137 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1138 'info_dict': {
1139 'id': 'Z4Vy8R84T1U',
1140 'ext': 'mp4',
1141 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1142 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1143 'duration': 433,
1144 'upload_date': '20130923',
1145 'uploader': 'Amelia Putri Harwita',
1146 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1147 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1148 'formats': 'maxcount:10',
1149 },
1150 'params': {
1151 'skip_download': True,
1152 'youtube_include_dash_manifest': False,
1153 },
1154 'skip': 'not actual anymore',
1155 },
1156 {
1157 # Youtube Music Auto-generated description
1158 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1159 'info_dict': {
1160 'id': 'MgNrAu2pzNs',
1161 'ext': 'mp4',
1162 'title': 'Voyeur Girl',
1163 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1164 'upload_date': '20190312',
1165 'uploader': 'Stephen - Topic',
1166 'uploader_id': 'UC-pWHpBjdGG69N9mM2auIAA',
1167 'artist': 'Stephen',
1168 'track': 'Voyeur Girl',
1169 'album': 'it\'s too much love to know my dear',
1170 'release_date': '20190313',
1171 'release_year': 2019,
1172 },
1173 'params': {
1174 'skip_download': True,
1175 },
1176 },
1177 {
1178 # Youtube Music Auto-generated description
1179 # Retrieve 'artist' field from 'Artist:' in video description
1180 # when it is present on youtube music video
1181 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1182 'info_dict': {
1183 'id': 'k0jLE7tTwjY',
1184 'ext': 'mp4',
1185 'title': 'Latch Feat. Sam Smith',
1186 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1187 'upload_date': '20150110',
1188 'uploader': 'Various Artists - Topic',
1189 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1190 'artist': 'Disclosure',
1191 'track': 'Latch Feat. Sam Smith',
1192 'album': 'Latch Featuring Sam Smith',
1193 'release_date': '20121008',
1194 'release_year': 2012,
1195 },
1196 'params': {
1197 'skip_download': True,
1198 },
1199 },
1200 {
1201 # Youtube Music Auto-generated description
1202 # handle multiple artists on youtube music video
1203 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1204 'info_dict': {
1205 'id': '74qn0eJSjpA',
1206 'ext': 'mp4',
1207 'title': 'Eastside',
1208 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1209 'upload_date': '20180710',
1210 'uploader': 'Benny Blanco - Topic',
1211 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1212 'artist': 'benny blanco, Halsey, Khalid',
1213 'track': 'Eastside',
1214 'album': 'Eastside',
1215 'release_date': '20180713',
1216 'release_year': 2018,
1217 },
1218 'params': {
1219 'skip_download': True,
1220 },
1221 },
1222 {
1223 # Youtube Music Auto-generated description
1224 # handle youtube music video with release_year and no release_date
1225 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1226 'info_dict': {
1227 'id': '-hcAI0g-f5M',
1228 'ext': 'mp4',
1229 'title': 'Put It On Me',
1230 'description': 'md5:f6422397c07c4c907c6638e1fee380a5',
1231 'upload_date': '20180426',
1232 'uploader': 'Matt Maeson - Topic',
1233 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1234 'artist': 'Matt Maeson',
1235 'track': 'Put It On Me',
1236 'album': 'The Hearse',
1237 'release_date': None,
1238 'release_year': 2018,
1239 },
1240 'params': {
1241 'skip_download': True,
1242 },
1243 },
1244 {
1245 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1246 'only_matching': True,
1247 },
1248 {
1249 # invalid -> valid video id redirection
1250 'url': 'DJztXj2GPfl',
1251 'info_dict': {
1252 'id': 'DJztXj2GPfk',
1253 'ext': 'mp4',
1254 'title': 'Panjabi MC - Mundian To Bach Ke (The Dictator Soundtrack)',
1255 'description': 'md5:bf577a41da97918e94fa9798d9228825',
1256 'upload_date': '20090125',
1257 'uploader': 'Prochorowka',
1258 'uploader_id': 'Prochorowka',
1259 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Prochorowka',
1260 'artist': 'Panjabi MC',
1261 'track': 'Beware of the Boys (Mundian to Bach Ke) - Motivo Hi-Lectro Remix',
1262 'album': 'Beware of the Boys (Mundian To Bach Ke)',
1263 },
1264 'params': {
1265 'skip_download': True,
1266 },
1267 }
1268 ]
1269
1270 def __init__(self, *args, **kwargs):
1271 super(YoutubeIE, self).__init__(*args, **kwargs)
1272 self._player_cache = {}
1273
1274 def report_video_info_webpage_download(self, video_id):
1275 """Report attempt to download video info webpage."""
1276 self.to_screen('%s: Downloading video info webpage' % video_id)
1277
1278 def report_information_extraction(self, video_id):
1279 """Report attempt to extract video information."""
1280 self.to_screen('%s: Extracting video information' % video_id)
1281
1282 def report_unavailable_format(self, video_id, format):
1283 """Report extracted video URL."""
1284 self.to_screen('%s: Format %s not available' % (video_id, format))
1285
1286 def report_rtmp_download(self):
1287 """Indicate the download will use the RTMP protocol."""
1288 self.to_screen('RTMP download detected')
1289
1290 def _signature_cache_id(self, example_sig):
1291 """ Return a string representation of a signature """
1292 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1293
1294 @classmethod
1295 def _extract_player_info(cls, player_url):
1296 for player_re in cls._PLAYER_INFO_RE:
1297 id_m = re.search(player_re, player_url)
1298 if id_m:
1299 break
1300 else:
1301 raise ExtractorError('Cannot identify player %r' % player_url)
1302 return id_m.group('ext'), id_m.group('id')
1303
1304 def _extract_signature_function(self, video_id, player_url, example_sig):
1305 player_type, player_id = self._extract_player_info(player_url)
1306
1307 # Read from filesystem cache
1308 func_id = '%s_%s_%s' % (
1309 player_type, player_id, self._signature_cache_id(example_sig))
1310 assert os.path.basename(func_id) == func_id
1311
1312 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1313 if cache_spec is not None:
1314 return lambda s: ''.join(s[i] for i in cache_spec)
1315
1316 download_note = (
1317 'Downloading player %s' % player_url
1318 if self._downloader.params.get('verbose') else
1319 'Downloading %s player %s' % (player_type, player_id)
1320 )
1321 if player_type == 'js':
1322 code = self._download_webpage(
1323 player_url, video_id,
1324 note=download_note,
1325 errnote='Download of %s failed' % player_url)
1326 res = self._parse_sig_js(code)
1327 elif player_type == 'swf':
1328 urlh = self._request_webpage(
1329 player_url, video_id,
1330 note=download_note,
1331 errnote='Download of %s failed' % player_url)
1332 code = urlh.read()
1333 res = self._parse_sig_swf(code)
1334 else:
1335 assert False, 'Invalid player type %r' % player_type
1336
1337 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1338 cache_res = res(test_string)
1339 cache_spec = [ord(c) for c in cache_res]
1340
1341 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1342 return res
1343
1344 def _print_sig_code(self, func, example_sig):
1345 def gen_sig_code(idxs):
1346 def _genslice(start, end, step):
1347 starts = '' if start == 0 else str(start)
1348 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1349 steps = '' if step == 1 else (':%d' % step)
1350 return 's[%s%s%s]' % (starts, ends, steps)
1351
1352 step = None
1353 # Quelch pyflakes warnings - start will be set when step is set
1354 start = '(Never used)'
1355 for i, prev in zip(idxs[1:], idxs[:-1]):
1356 if step is not None:
1357 if i - prev == step:
1358 continue
1359 yield _genslice(start, prev, step)
1360 step = None
1361 continue
1362 if i - prev in [-1, 1]:
1363 step = i - prev
1364 start = prev
1365 continue
1366 else:
1367 yield 's[%d]' % prev
1368 if step is None:
1369 yield 's[%d]' % i
1370 else:
1371 yield _genslice(start, i, step)
1372
1373 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1374 cache_res = func(test_string)
1375 cache_spec = [ord(c) for c in cache_res]
1376 expr_code = ' + '.join(gen_sig_code(cache_spec))
1377 signature_id_tuple = '(%s)' % (
1378 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1379 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1380 ' return %s\n') % (signature_id_tuple, expr_code)
1381 self.to_screen('Extracted signature function:\n' + code)
1382
1383 def _parse_sig_js(self, jscode):
1384 funcname = self._search_regex(
1385 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1386 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1387 r'\b(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1388 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1389 # Obsolete patterns
1390 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1391 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1392 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1393 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1394 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1395 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1396 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1397 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1398 jscode, 'Initial JS player signature function name', group='sig')
1399
1400 jsi = JSInterpreter(jscode)
1401 initial_function = jsi.extract_function(funcname)
1402 return lambda s: initial_function([s])
1403
1404 def _parse_sig_swf(self, file_contents):
1405 swfi = SWFInterpreter(file_contents)
1406 TARGET_CLASSNAME = 'SignatureDecipher'
1407 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1408 initial_function = swfi.extract_function(searched_class, 'decipher')
1409 return lambda s: initial_function([s])
1410
1411 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1412 """Turn the encrypted s field into a working signature"""
1413
1414 if player_url is None:
1415 raise ExtractorError('Cannot decrypt signature without player_url')
1416
1417 if player_url.startswith('//'):
1418 player_url = 'https:' + player_url
1419 elif not re.match(r'https?://', player_url):
1420 player_url = compat_urlparse.urljoin(
1421 'https://www.youtube.com', player_url)
1422 try:
1423 player_id = (player_url, self._signature_cache_id(s))
1424 if player_id not in self._player_cache:
1425 func = self._extract_signature_function(
1426 video_id, player_url, s
1427 )
1428 self._player_cache[player_id] = func
1429 func = self._player_cache[player_id]
1430 if self._downloader.params.get('youtube_print_sig_code'):
1431 self._print_sig_code(func, s)
1432 return func(s)
1433 except Exception as e:
1434 tb = traceback.format_exc()
1435 raise ExtractorError(
1436 'Signature extraction failed: ' + tb, cause=e)
1437
1438 def _get_subtitles(self, video_id, webpage):
1439 try:
1440 subs_doc = self._download_xml(
1441 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1442 video_id, note=False)
1443 except ExtractorError as err:
1444 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1445 return {}
1446
1447 sub_lang_list = {}
1448 for track in subs_doc.findall('track'):
1449 lang = track.attrib['lang_code']
1450 if lang in sub_lang_list:
1451 continue
1452 sub_formats = []
1453 for ext in self._SUBTITLE_FORMATS:
1454 params = compat_urllib_parse_urlencode({
1455 'lang': lang,
1456 'v': video_id,
1457 'fmt': ext,
1458 'name': track.attrib['name'].encode('utf-8'),
1459 })
1460 sub_formats.append({
1461 'url': 'https://www.youtube.com/api/timedtext?' + params,
1462 'ext': ext,
1463 })
1464 sub_lang_list[lang] = sub_formats
1465 if not sub_lang_list:
1466 self._downloader.report_warning('video doesn\'t have subtitles')
1467 return {}
1468 return sub_lang_list
1469
1470 def _get_ytplayer_config(self, video_id, webpage):
1471 patterns = (
1472 # User data may contain arbitrary character sequences that may affect
1473 # JSON extraction with regex, e.g. when '};' is contained the second
1474 # regex won't capture the whole JSON. Yet working around by trying more
1475 # concrete regex first keeping in mind proper quoted string handling
1476 # to be implemented in future that will replace this workaround (see
1477 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1478 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1479 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1480 r';ytplayer\.config\s*=\s*({.+?});',
1481 )
1482 config = self._search_regex(
1483 patterns, webpage, 'ytplayer.config', default=None)
1484 if config:
1485 return self._parse_json(
1486 uppercase_escape(config), video_id, fatal=False)
1487
1488 def _get_automatic_captions(self, video_id, webpage):
1489 """We need the webpage for getting the captions url, pass it as an
1490 argument to speed up the process."""
1491 self.to_screen('%s: Looking for automatic captions' % video_id)
1492 player_config = self._get_ytplayer_config(video_id, webpage)
1493 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1494 if not player_config:
1495 self._downloader.report_warning(err_msg)
1496 return {}
1497 try:
1498 args = player_config['args']
1499 caption_url = args.get('ttsurl')
1500 if caption_url:
1501 timestamp = args['timestamp']
1502 # We get the available subtitles
1503 list_params = compat_urllib_parse_urlencode({
1504 'type': 'list',
1505 'tlangs': 1,
1506 'asrs': 1,
1507 })
1508 list_url = caption_url + '&' + list_params
1509 caption_list = self._download_xml(list_url, video_id)
1510 original_lang_node = caption_list.find('track')
1511 if original_lang_node is None:
1512 self._downloader.report_warning('Video doesn\'t have automatic captions')
1513 return {}
1514 original_lang = original_lang_node.attrib['lang_code']
1515 caption_kind = original_lang_node.attrib.get('kind', '')
1516
1517 sub_lang_list = {}
1518 for lang_node in caption_list.findall('target'):
1519 sub_lang = lang_node.attrib['lang_code']
1520 sub_formats = []
1521 for ext in self._SUBTITLE_FORMATS:
1522 params = compat_urllib_parse_urlencode({
1523 'lang': original_lang,
1524 'tlang': sub_lang,
1525 'fmt': ext,
1526 'ts': timestamp,
1527 'kind': caption_kind,
1528 })
1529 sub_formats.append({
1530 'url': caption_url + '&' + params,
1531 'ext': ext,
1532 })
1533 sub_lang_list[sub_lang] = sub_formats
1534 return sub_lang_list
1535
1536 def make_captions(sub_url, sub_langs):
1537 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1538 caption_qs = compat_parse_qs(parsed_sub_url.query)
1539 captions = {}
1540 for sub_lang in sub_langs:
1541 sub_formats = []
1542 for ext in self._SUBTITLE_FORMATS:
1543 caption_qs.update({
1544 'tlang': [sub_lang],
1545 'fmt': [ext],
1546 })
1547 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1548 query=compat_urllib_parse_urlencode(caption_qs, True)))
1549 sub_formats.append({
1550 'url': sub_url,
1551 'ext': ext,
1552 })
1553 captions[sub_lang] = sub_formats
1554 return captions
1555
1556 # New captions format as of 22.06.2017
1557 player_response = args.get('player_response')
1558 if player_response and isinstance(player_response, compat_str):
1559 player_response = self._parse_json(
1560 player_response, video_id, fatal=False)
1561 if player_response:
1562 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1563 base_url = renderer['captionTracks'][0]['baseUrl']
1564 sub_lang_list = []
1565 for lang in renderer['translationLanguages']:
1566 lang_code = lang.get('languageCode')
1567 if lang_code:
1568 sub_lang_list.append(lang_code)
1569 return make_captions(base_url, sub_lang_list)
1570
1571 # Some videos don't provide ttsurl but rather caption_tracks and
1572 # caption_translation_languages (e.g. 20LmZk1hakA)
1573 # Does not used anymore as of 22.06.2017
1574 caption_tracks = args['caption_tracks']
1575 caption_translation_languages = args['caption_translation_languages']
1576 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1577 sub_lang_list = []
1578 for lang in caption_translation_languages.split(','):
1579 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1580 sub_lang = lang_qs.get('lc', [None])[0]
1581 if sub_lang:
1582 sub_lang_list.append(sub_lang)
1583 return make_captions(caption_url, sub_lang_list)
1584 # An extractor error can be raise by the download process if there are
1585 # no automatic captions but there are subtitles
1586 except (KeyError, IndexError, ExtractorError):
1587 self._downloader.report_warning(err_msg)
1588 return {}
1589
1590 def _mark_watched(self, video_id, video_info, player_response):
1591 playback_url = url_or_none(try_get(
1592 player_response,
1593 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1594 video_info, lambda x: x['videostats_playback_base_url'][0]))
1595 if not playback_url:
1596 return
1597 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1598 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1599
1600 # cpn generation algorithm is reverse engineered from base.js.
1601 # In fact it works even with dummy cpn.
1602 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1603 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1604
1605 qs.update({
1606 'ver': ['2'],
1607 'cpn': [cpn],
1608 })
1609 playback_url = compat_urlparse.urlunparse(
1610 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1611
1612 self._download_webpage(
1613 playback_url, video_id, 'Marking watched',
1614 'Unable to mark watched', fatal=False)
1615
1616 @staticmethod
1617 def _extract_urls(webpage):
1618 # Embedded YouTube player
1619 entries = [
1620 unescapeHTML(mobj.group('url'))
1621 for mobj in re.finditer(r'''(?x)
1622 (?:
1623 <iframe[^>]+?src=|
1624 data-video-url=|
1625 <embed[^>]+?src=|
1626 embedSWF\(?:\s*|
1627 <object[^>]+data=|
1628 new\s+SWFObject\(
1629 )
1630 (["\'])
1631 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1632 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1633 \1''', webpage)]
1634
1635 # lazyYT YouTube embed
1636 entries.extend(list(map(
1637 unescapeHTML,
1638 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1639
1640 # Wordpress "YouTube Video Importer" plugin
1641 matches = re.findall(r'''(?x)<div[^>]+
1642 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1643 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1644 entries.extend(m[-1] for m in matches)
1645
1646 return entries
1647
1648 @staticmethod
1649 def _extract_url(webpage):
1650 urls = YoutubeIE._extract_urls(webpage)
1651 return urls[0] if urls else None
1652
1653 @classmethod
1654 def extract_id(cls, url):
1655 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1656 if mobj is None:
1657 raise ExtractorError('Invalid URL: %s' % url)
1658 video_id = mobj.group(2)
1659 return video_id
1660
1661 def _extract_chapters_from_json(self, webpage, video_id, duration):
1662 if not webpage:
1663 return
1664 player = self._parse_json(
1665 self._search_regex(
1666 r'RELATED_PLAYER_ARGS["\']\s*:\s*({.+})\s*,?\s*\n', webpage,
1667 'player args', default='{}'),
1668 video_id, fatal=False)
1669 if not player or not isinstance(player, dict):
1670 return
1671 watch_next_response = player.get('watch_next_response')
1672 if not isinstance(watch_next_response, compat_str):
1673 return
1674 response = self._parse_json(watch_next_response, video_id, fatal=False)
1675 if not response or not isinstance(response, dict):
1676 return
1677 chapters_list = try_get(
1678 response,
1679 lambda x: x['playerOverlays']
1680 ['playerOverlayRenderer']
1681 ['decoratedPlayerBarRenderer']
1682 ['decoratedPlayerBarRenderer']
1683 ['playerBar']
1684 ['chapteredPlayerBarRenderer']
1685 ['chapters'],
1686 list)
1687 if not chapters_list:
1688 return
1689
1690 def chapter_time(chapter):
1691 return float_or_none(
1692 try_get(
1693 chapter,
1694 lambda x: x['chapterRenderer']['timeRangeStartMillis'],
1695 int),
1696 scale=1000)
1697 chapters = []
1698 for next_num, chapter in enumerate(chapters_list, start=1):
1699 start_time = chapter_time(chapter)
1700 if start_time is None:
1701 continue
1702 end_time = (chapter_time(chapters_list[next_num])
1703 if next_num < len(chapters_list) else duration)
1704 if end_time is None:
1705 continue
1706 title = try_get(
1707 chapter, lambda x: x['chapterRenderer']['title']['simpleText'],
1708 compat_str)
1709 chapters.append({
1710 'start_time': start_time,
1711 'end_time': end_time,
1712 'title': title,
1713 })
1714 return chapters
1715
1716 @staticmethod
1717 def _extract_chapters_from_description(description, duration):
1718 if not description:
1719 return None
1720 chapter_lines = re.findall(
1721 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1722 description)
1723 if not chapter_lines:
1724 return None
1725 chapters = []
1726 for next_num, (chapter_line, time_point) in enumerate(
1727 chapter_lines, start=1):
1728 start_time = parse_duration(time_point)
1729 if start_time is None:
1730 continue
1731 if start_time > duration:
1732 break
1733 end_time = (duration if next_num == len(chapter_lines)
1734 else parse_duration(chapter_lines[next_num][1]))
1735 if end_time is None:
1736 continue
1737 if end_time > duration:
1738 end_time = duration
1739 if start_time > end_time:
1740 break
1741 chapter_title = re.sub(
1742 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1743 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1744 chapters.append({
1745 'start_time': start_time,
1746 'end_time': end_time,
1747 'title': chapter_title,
1748 })
1749 return chapters
1750
1751 def _extract_chapters(self, webpage, description, video_id, duration):
1752 return (self._extract_chapters_from_json(webpage, video_id, duration)
1753 or self._extract_chapters_from_description(description, duration))
1754
1755 def _real_extract(self, url):
1756 url, smuggled_data = unsmuggle_url(url, {})
1757
1758 proto = (
1759 'http' if self._downloader.params.get('prefer_insecure', False)
1760 else 'https')
1761
1762 start_time = None
1763 end_time = None
1764 parsed_url = compat_urllib_parse_urlparse(url)
1765 for component in [parsed_url.fragment, parsed_url.query]:
1766 query = compat_parse_qs(component)
1767 if start_time is None and 't' in query:
1768 start_time = parse_duration(query['t'][0])
1769 if start_time is None and 'start' in query:
1770 start_time = parse_duration(query['start'][0])
1771 if end_time is None and 'end' in query:
1772 end_time = parse_duration(query['end'][0])
1773
1774 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1775 mobj = re.search(self._NEXT_URL_RE, url)
1776 if mobj:
1777 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1778 video_id = self.extract_id(url)
1779
1780 # Get video webpage
1781 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1782 video_webpage, urlh = self._download_webpage_handle(url, video_id)
1783
1784 qs = compat_parse_qs(compat_urllib_parse_urlparse(urlh.geturl()).query)
1785 video_id = qs.get('v', [None])[0] or video_id
1786
1787 # Attempt to extract SWF player URL
1788 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1789 if mobj is not None:
1790 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1791 else:
1792 player_url = None
1793
1794 dash_mpds = []
1795
1796 def add_dash_mpd(video_info):
1797 dash_mpd = video_info.get('dashmpd')
1798 if dash_mpd and dash_mpd[0] not in dash_mpds:
1799 dash_mpds.append(dash_mpd[0])
1800
1801 def add_dash_mpd_pr(pl_response):
1802 dash_mpd = url_or_none(try_get(
1803 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1804 compat_str))
1805 if dash_mpd and dash_mpd not in dash_mpds:
1806 dash_mpds.append(dash_mpd)
1807
1808 is_live = None
1809 view_count = None
1810
1811 def extract_view_count(v_info):
1812 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1813
1814 def extract_player_response(player_response, video_id):
1815 pl_response = str_or_none(player_response)
1816 if not pl_response:
1817 return
1818 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1819 if isinstance(pl_response, dict):
1820 add_dash_mpd_pr(pl_response)
1821 return pl_response
1822
1823 player_response = {}
1824
1825 # Get video info
1826 video_info = {}
1827 embed_webpage = None
1828 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1829 age_gate = True
1830 # We simulate the access to the video from www.youtube.com/v/{video_id}
1831 # this can be viewed without login into Youtube
1832 url = proto + '://www.youtube.com/embed/%s' % video_id
1833 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1834 data = compat_urllib_parse_urlencode({
1835 'video_id': video_id,
1836 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1837 'sts': self._search_regex(
1838 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1839 })
1840 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1841 try:
1842 video_info_webpage = self._download_webpage(
1843 video_info_url, video_id,
1844 note='Refetching age-gated info webpage',
1845 errnote='unable to download video info webpage')
1846 except ExtractorError:
1847 video_info_webpage = None
1848 if video_info_webpage:
1849 video_info = compat_parse_qs(video_info_webpage)
1850 pl_response = video_info.get('player_response', [None])[0]
1851 player_response = extract_player_response(pl_response, video_id)
1852 add_dash_mpd(video_info)
1853 view_count = extract_view_count(video_info)
1854 else:
1855 age_gate = False
1856 # Try looking directly into the video webpage
1857 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1858 if ytplayer_config:
1859 args = ytplayer_config['args']
1860 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1861 # Convert to the same format returned by compat_parse_qs
1862 video_info = dict((k, [v]) for k, v in args.items())
1863 add_dash_mpd(video_info)
1864 # Rental video is not rented but preview is available (e.g.
1865 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1866 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1867 if not video_info and args.get('ypc_vid'):
1868 return self.url_result(
1869 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1870 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1871 is_live = True
1872 if not player_response:
1873 player_response = extract_player_response(args.get('player_response'), video_id)
1874 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1875 add_dash_mpd_pr(player_response)
1876
1877 def extract_unavailable_message():
1878 messages = []
1879 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1880 msg = self._html_search_regex(
1881 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1882 video_webpage, 'unavailable %s' % kind, default=None)
1883 if msg:
1884 messages.append(msg)
1885 if messages:
1886 return '\n'.join(messages)
1887
1888 if not video_info and not player_response:
1889 unavailable_message = extract_unavailable_message()
1890 if not unavailable_message:
1891 unavailable_message = 'Unable to extract video data'
1892 raise ExtractorError(
1893 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1894
1895 if not isinstance(video_info, dict):
1896 video_info = {}
1897
1898 video_details = try_get(
1899 player_response, lambda x: x['videoDetails'], dict) or {}
1900
1901 microformat = try_get(
1902 player_response, lambda x: x['microformat']['playerMicroformatRenderer'], dict) or {}
1903
1904 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1905 if not video_title:
1906 self._downloader.report_warning('Unable to extract video title')
1907 video_title = '_'
1908
1909 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1910 if video_description:
1911
1912 def replace_url(m):
1913 redir_url = compat_urlparse.urljoin(url, m.group(1))
1914 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1915 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1916 qs = compat_parse_qs(parsed_redir_url.query)
1917 q = qs.get('q')
1918 if q and q[0]:
1919 return q[0]
1920 return redir_url
1921
1922 description_original = video_description = re.sub(r'''(?x)
1923 <a\s+
1924 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1925 (?:title|href)="([^"]+)"\s+
1926 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1927 class="[^"]*"[^>]*>
1928 [^<]+\.{3}\s*
1929 </a>
1930 ''', replace_url, video_description)
1931 video_description = clean_html(video_description)
1932 else:
1933 video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
1934
1935 if not smuggled_data.get('force_singlefeed', False):
1936 if not self._downloader.params.get('noplaylist'):
1937 multifeed_metadata_list = try_get(
1938 player_response,
1939 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1940 compat_str) or try_get(
1941 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1942 if multifeed_metadata_list:
1943 entries = []
1944 feed_ids = []
1945 for feed in multifeed_metadata_list.split(','):
1946 # Unquote should take place before split on comma (,) since textual
1947 # fields may contain comma as well (see
1948 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1949 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1950
1951 def feed_entry(name):
1952 return try_get(feed_data, lambda x: x[name][0], compat_str)
1953
1954 feed_id = feed_entry('id')
1955 if not feed_id:
1956 continue
1957 feed_title = feed_entry('title')
1958 title = video_title
1959 if feed_title:
1960 title += ' (%s)' % feed_title
1961 entries.append({
1962 '_type': 'url_transparent',
1963 'ie_key': 'Youtube',
1964 'url': smuggle_url(
1965 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1966 {'force_singlefeed': True}),
1967 'title': title,
1968 })
1969 feed_ids.append(feed_id)
1970 self.to_screen(
1971 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1972 % (', '.join(feed_ids), video_id))
1973 return self.playlist_result(entries, video_id, video_title, video_description)
1974 else:
1975 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1976
1977 if view_count is None:
1978 view_count = extract_view_count(video_info)
1979 if view_count is None and video_details:
1980 view_count = int_or_none(video_details.get('viewCount'))
1981 if view_count is None and microformat:
1982 view_count = int_or_none(microformat.get('viewCount'))
1983
1984 if is_live is None:
1985 is_live = bool_or_none(video_details.get('isLive'))
1986
1987 # Check for "rental" videos
1988 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1989 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1990
1991 def _extract_filesize(media_url):
1992 return int_or_none(self._search_regex(
1993 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1994
1995 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
1996 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
1997
1998 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1999 self.report_rtmp_download()
2000 formats = [{
2001 'format_id': '_rtmp',
2002 'protocol': 'rtmp',
2003 'url': video_info['conn'][0],
2004 'player_url': player_url,
2005 }]
2006 elif not is_live and (streaming_formats or len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
2007 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
2008 if 'rtmpe%3Dyes' in encoded_url_map:
2009 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
2010 formats = []
2011 formats_spec = {}
2012 fmt_list = video_info.get('fmt_list', [''])[0]
2013 if fmt_list:
2014 for fmt in fmt_list.split(','):
2015 spec = fmt.split('/')
2016 if len(spec) > 1:
2017 width_height = spec[1].split('x')
2018 if len(width_height) == 2:
2019 formats_spec[spec[0]] = {
2020 'resolution': spec[1],
2021 'width': int_or_none(width_height[0]),
2022 'height': int_or_none(width_height[1]),
2023 }
2024 for fmt in streaming_formats:
2025 itag = str_or_none(fmt.get('itag'))
2026 if not itag:
2027 continue
2028 quality = fmt.get('quality')
2029 quality_label = fmt.get('qualityLabel') or quality
2030 formats_spec[itag] = {
2031 'asr': int_or_none(fmt.get('audioSampleRate')),
2032 'filesize': int_or_none(fmt.get('contentLength')),
2033 'format_note': quality_label,
2034 'fps': int_or_none(fmt.get('fps')),
2035 'height': int_or_none(fmt.get('height')),
2036 # bitrate for itag 43 is always 2147483647
2037 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
2038 'width': int_or_none(fmt.get('width')),
2039 }
2040
2041 for fmt in streaming_formats:
2042 if fmt.get('drmFamilies') or fmt.get('drm_families'):
2043 continue
2044 url = url_or_none(fmt.get('url'))
2045
2046 if not url:
2047 cipher = fmt.get('cipher') or fmt.get('signatureCipher')
2048 if not cipher:
2049 continue
2050 url_data = compat_parse_qs(cipher)
2051 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
2052 if not url:
2053 continue
2054 else:
2055 cipher = None
2056 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
2057
2058 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
2059 # Unsupported FORMAT_STREAM_TYPE_OTF
2060 if stream_type == 3:
2061 continue
2062
2063 format_id = fmt.get('itag') or url_data['itag'][0]
2064 if not format_id:
2065 continue
2066 format_id = compat_str(format_id)
2067
2068 if cipher:
2069 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
2070 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
2071 jsplayer_url_json = self._search_regex(
2072 ASSETS_RE,
2073 embed_webpage if age_gate else video_webpage,
2074 'JS player URL (1)', default=None)
2075 if not jsplayer_url_json and not age_gate:
2076 # We need the embed website after all
2077 if embed_webpage is None:
2078 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
2079 embed_webpage = self._download_webpage(
2080 embed_url, video_id, 'Downloading embed webpage')
2081 jsplayer_url_json = self._search_regex(
2082 ASSETS_RE, embed_webpage, 'JS player URL')
2083
2084 player_url = json.loads(jsplayer_url_json)
2085 if player_url is None:
2086 player_url_json = self._search_regex(
2087 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2088 video_webpage, 'age gate player URL')
2089 player_url = json.loads(player_url_json)
2090
2091 if 'sig' in url_data:
2092 url += '&signature=' + url_data['sig'][0]
2093 elif 's' in url_data:
2094 encrypted_sig = url_data['s'][0]
2095
2096 if self._downloader.params.get('verbose'):
2097 if player_url is None:
2098 player_desc = 'unknown'
2099 else:
2100 player_type, player_version = self._extract_player_info(player_url)
2101 player_desc = '%s player %s' % ('flash' if player_type == 'swf' else 'html5', player_version)
2102 parts_sizes = self._signature_cache_id(encrypted_sig)
2103 self.to_screen('{%s} signature length %s, %s' %
2104 (format_id, parts_sizes, player_desc))
2105
2106 signature = self._decrypt_signature(
2107 encrypted_sig, video_id, player_url, age_gate)
2108 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2109 url += '&%s=%s' % (sp, signature)
2110 if 'ratebypass' not in url:
2111 url += '&ratebypass=yes'
2112
2113 dct = {
2114 'format_id': format_id,
2115 'url': url,
2116 'player_url': player_url,
2117 }
2118 if format_id in self._formats:
2119 dct.update(self._formats[format_id])
2120 if format_id in formats_spec:
2121 dct.update(formats_spec[format_id])
2122
2123 # Some itags are not included in DASH manifest thus corresponding formats will
2124 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2125 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2126 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2127 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2128
2129 if width is None:
2130 width = int_or_none(fmt.get('width'))
2131 if height is None:
2132 height = int_or_none(fmt.get('height'))
2133
2134 filesize = int_or_none(url_data.get(
2135 'clen', [None])[0]) or _extract_filesize(url)
2136
2137 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2138 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2139
2140 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2141 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2142 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2143
2144 more_fields = {
2145 'filesize': filesize,
2146 'tbr': tbr,
2147 'width': width,
2148 'height': height,
2149 'fps': fps,
2150 'format_note': quality_label or quality,
2151 }
2152 for key, value in more_fields.items():
2153 if value:
2154 dct[key] = value
2155 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2156 if type_:
2157 type_split = type_.split(';')
2158 kind_ext = type_split[0].split('/')
2159 if len(kind_ext) == 2:
2160 kind, _ = kind_ext
2161 dct['ext'] = mimetype2ext(type_split[0])
2162 if kind in ('audio', 'video'):
2163 codecs = None
2164 for mobj in re.finditer(
2165 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2166 if mobj.group('key') == 'codecs':
2167 codecs = mobj.group('val')
2168 break
2169 if codecs:
2170 dct.update(parse_codecs(codecs))
2171 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2172 dct['downloader_options'] = {
2173 # Youtube throttles chunks >~10M
2174 'http_chunk_size': 10485760,
2175 }
2176 formats.append(dct)
2177 else:
2178 manifest_url = (
2179 url_or_none(try_get(
2180 player_response,
2181 lambda x: x['streamingData']['hlsManifestUrl'],
2182 compat_str))
2183 or url_or_none(try_get(
2184 video_info, lambda x: x['hlsvp'][0], compat_str)))
2185 if manifest_url:
2186 formats = []
2187 m3u8_formats = self._extract_m3u8_formats(
2188 manifest_url, video_id, 'mp4', fatal=False)
2189 for a_format in m3u8_formats:
2190 itag = self._search_regex(
2191 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2192 if itag:
2193 a_format['format_id'] = itag
2194 if itag in self._formats:
2195 dct = self._formats[itag].copy()
2196 dct.update(a_format)
2197 a_format = dct
2198 a_format['player_url'] = player_url
2199 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2200 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2201 formats.append(a_format)
2202 else:
2203 error_message = extract_unavailable_message()
2204 if not error_message:
2205 error_message = clean_html(try_get(
2206 player_response, lambda x: x['playabilityStatus']['reason'],
2207 compat_str))
2208 if not error_message:
2209 error_message = clean_html(
2210 try_get(video_info, lambda x: x['reason'][0], compat_str))
2211 if error_message:
2212 raise ExtractorError(error_message, expected=True)
2213 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2214
2215 # uploader
2216 video_uploader = try_get(
2217 video_info, lambda x: x['author'][0],
2218 compat_str) or str_or_none(video_details.get('author'))
2219 if video_uploader:
2220 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2221 else:
2222 self._downloader.report_warning('unable to extract uploader name')
2223
2224 # uploader_id
2225 video_uploader_id = None
2226 video_uploader_url = None
2227 mobj = re.search(
2228 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2229 video_webpage)
2230 if mobj is not None:
2231 video_uploader_id = mobj.group('uploader_id')
2232 video_uploader_url = mobj.group('uploader_url')
2233 else:
2234 owner_profile_url = url_or_none(microformat.get('ownerProfileUrl'))
2235 if owner_profile_url:
2236 video_uploader_id = self._search_regex(
2237 r'(?:user|channel)/([^/]+)', owner_profile_url, 'uploader id',
2238 default=None)
2239 video_uploader_url = owner_profile_url
2240
2241 channel_id = (
2242 str_or_none(video_details.get('channelId'))
2243 or self._html_search_meta(
2244 'channelId', video_webpage, 'channel id', default=None)
2245 or self._search_regex(
2246 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2247 video_webpage, 'channel id', default=None, group='id'))
2248 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2249
2250 thumbnails = []
2251 thumbnails_list = try_get(
2252 video_details, lambda x: x['thumbnail']['thumbnails'], list) or []
2253 for t in thumbnails_list:
2254 if not isinstance(t, dict):
2255 continue
2256 thumbnail_url = url_or_none(t.get('url'))
2257 if not thumbnail_url:
2258 continue
2259 thumbnails.append({
2260 'url': thumbnail_url,
2261 'width': int_or_none(t.get('width')),
2262 'height': int_or_none(t.get('height')),
2263 })
2264
2265 if not thumbnails:
2266 video_thumbnail = None
2267 # We try first to get a high quality image:
2268 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2269 video_webpage, re.DOTALL)
2270 if m_thumb is not None:
2271 video_thumbnail = m_thumb.group(1)
2272 thumbnail_url = try_get(video_info, lambda x: x['thumbnail_url'][0], compat_str)
2273 if thumbnail_url:
2274 video_thumbnail = compat_urllib_parse_unquote_plus(thumbnail_url)
2275 if video_thumbnail:
2276 thumbnails.append({'url': video_thumbnail})
2277
2278 # upload date
2279 upload_date = self._html_search_meta(
2280 'datePublished', video_webpage, 'upload date', default=None)
2281 if not upload_date:
2282 upload_date = self._search_regex(
2283 [r'(?s)id="eow-date.*?>(.*?)</span>',
2284 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2285 video_webpage, 'upload date', default=None)
2286 if not upload_date:
2287 upload_date = microformat.get('publishDate') or microformat.get('uploadDate')
2288 upload_date = unified_strdate(upload_date)
2289
2290 video_license = self._html_search_regex(
2291 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2292 video_webpage, 'license', default=None)
2293
2294 m_music = re.search(
2295 r'''(?x)
2296 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2297 <ul[^>]*>\s*
2298 <li>(?P<title>.+?)
2299 by (?P<creator>.+?)
2300 (?:
2301 \(.+?\)|
2302 <a[^>]*
2303 (?:
2304 \bhref=["\']/red[^>]*>| # drop possible
2305 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2306 )
2307 .*?
2308 )?</li
2309 ''',
2310 video_webpage)
2311 if m_music:
2312 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2313 video_creator = clean_html(m_music.group('creator'))
2314 else:
2315 video_alt_title = video_creator = None
2316
2317 def extract_meta(field):
2318 return self._html_search_regex(
2319 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2320 video_webpage, field, default=None)
2321
2322 track = extract_meta('Song')
2323 artist = extract_meta('Artist')
2324 album = extract_meta('Album')
2325
2326 # Youtube Music Auto-generated description
2327 release_date = release_year = None
2328 if video_description:
2329 mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^Ā·]+)Ā·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?ā„—\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)
2330 if mobj:
2331 if not track:
2332 track = mobj.group('track').strip()
2333 if not artist:
2334 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('Ā·'))
2335 if not album:
2336 album = mobj.group('album'.strip())
2337 release_year = mobj.group('release_year')
2338 release_date = mobj.group('release_date')
2339 if release_date:
2340 release_date = release_date.replace('-', '')
2341 if not release_year:
2342 release_year = int(release_date[:4])
2343 if release_year:
2344 release_year = int(release_year)
2345
2346 m_episode = re.search(
2347 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2348 video_webpage)
2349 if m_episode:
2350 series = unescapeHTML(m_episode.group('series'))
2351 season_number = int(m_episode.group('season'))
2352 episode_number = int(m_episode.group('episode'))
2353 else:
2354 series = season_number = episode_number = None
2355
2356 m_cat_container = self._search_regex(
2357 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2358 video_webpage, 'categories', default=None)
2359 if m_cat_container:
2360 category = self._html_search_regex(
2361 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2362 default=None)
2363 video_categories = None if category is None else [category]
2364 else:
2365 video_categories = None
2366
2367 video_tags = [
2368 unescapeHTML(m.group('content'))
2369 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2370
2371 def _extract_count(count_name):
2372 return str_to_int(self._search_regex(
2373 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2374 % re.escape(count_name),
2375 video_webpage, count_name, default=None))
2376
2377 like_count = _extract_count('like')
2378 dislike_count = _extract_count('dislike')
2379
2380 if view_count is None:
2381 view_count = str_to_int(self._search_regex(
2382 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2383 'view count', default=None))
2384
2385 average_rating = (
2386 float_or_none(video_details.get('averageRating'))
2387 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2388
2389 # subtitles
2390 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2391 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2392
2393 video_duration = try_get(
2394 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2395 if not video_duration:
2396 video_duration = int_or_none(video_details.get('lengthSeconds'))
2397 if not video_duration:
2398 video_duration = parse_duration(self._html_search_meta(
2399 'duration', video_webpage, 'video duration'))
2400
2401 # annotations
2402 video_annotations = None
2403 if self._downloader.params.get('writeannotations', False):
2404 xsrf_token = self._search_regex(
2405 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2406 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2407 invideo_url = try_get(
2408 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2409 if xsrf_token and invideo_url:
2410 xsrf_field_name = self._search_regex(
2411 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2412 video_webpage, 'xsrf field name',
2413 group='xsrf_field_name', default='session_token')
2414 video_annotations = self._download_webpage(
2415 self._proto_relative_url(invideo_url),
2416 video_id, note='Downloading annotations',
2417 errnote='Unable to download video annotations', fatal=False,
2418 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2419
2420 chapters = self._extract_chapters(video_webpage, description_original, video_id, video_duration)
2421
2422 # Look for the DASH manifest
2423 if self._downloader.params.get('youtube_include_dash_manifest', True):
2424 dash_mpd_fatal = True
2425 for mpd_url in dash_mpds:
2426 dash_formats = {}
2427 try:
2428 def decrypt_sig(mobj):
2429 s = mobj.group(1)
2430 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2431 return '/signature/%s' % dec_s
2432
2433 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2434
2435 for df in self._extract_mpd_formats(
2436 mpd_url, video_id, fatal=dash_mpd_fatal,
2437 formats_dict=self._formats):
2438 if not df.get('filesize'):
2439 df['filesize'] = _extract_filesize(df['url'])
2440 # Do not overwrite DASH format found in some previous DASH manifest
2441 if df['format_id'] not in dash_formats:
2442 dash_formats[df['format_id']] = df
2443 # Additional DASH manifests may end up in HTTP Error 403 therefore
2444 # allow them to fail without bug report message if we already have
2445 # some DASH manifest succeeded. This is temporary workaround to reduce
2446 # burst of bug reports until we figure out the reason and whether it
2447 # can be fixed at all.
2448 dash_mpd_fatal = False
2449 except (ExtractorError, KeyError) as e:
2450 self.report_warning(
2451 'Skipping DASH manifest: %r' % e, video_id)
2452 if dash_formats:
2453 # Remove the formats we found through non-DASH, they
2454 # contain less info and it can be wrong, because we use
2455 # fixed values (for example the resolution). See
2456 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2457 # example.
2458 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2459 formats.extend(dash_formats.values())
2460
2461 # Check for malformed aspect ratio
2462 stretched_m = re.search(
2463 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2464 video_webpage)
2465 if stretched_m:
2466 w = float(stretched_m.group('w'))
2467 h = float(stretched_m.group('h'))
2468 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2469 # We will only process correct ratios.
2470 if w > 0 and h > 0:
2471 ratio = w / h
2472 for f in formats:
2473 if f.get('vcodec') != 'none':
2474 f['stretched_ratio'] = ratio
2475
2476 if not formats:
2477 if 'reason' in video_info:
2478 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2479 regions_allowed = self._html_search_meta(
2480 'regionsAllowed', video_webpage, default=None)
2481 countries = regions_allowed.split(',') if regions_allowed else None
2482 self.raise_geo_restricted(
2483 msg=video_info['reason'][0], countries=countries)
2484 reason = video_info['reason'][0]
2485 if 'Invalid parameters' in reason:
2486 unavailable_message = extract_unavailable_message()
2487 if unavailable_message:
2488 reason = unavailable_message
2489 raise ExtractorError(
2490 'YouTube said: %s' % reason,
2491 expected=True, video_id=video_id)
2492 if video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos']):
2493 raise ExtractorError('This video is DRM protected.', expected=True)
2494
2495 self._sort_formats(formats)
2496
2497 self.mark_watched(video_id, video_info, player_response)
2498
2499 return {
2500 'id': video_id,
2501 'uploader': video_uploader,
2502 'uploader_id': video_uploader_id,
2503 'uploader_url': video_uploader_url,
2504 'channel_id': channel_id,
2505 'channel_url': channel_url,
2506 'upload_date': upload_date,
2507 'license': video_license,
2508 'creator': video_creator or artist,
2509 'title': video_title,
2510 'alt_title': video_alt_title or track,
2511 'thumbnails': thumbnails,
2512 'description': video_description,
2513 'categories': video_categories,
2514 'tags': video_tags,
2515 'subtitles': video_subtitles,
2516 'automatic_captions': automatic_captions,
2517 'duration': video_duration,
2518 'age_limit': 18 if age_gate else 0,
2519 'annotations': video_annotations,
2520 'chapters': chapters,
2521 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2522 'view_count': view_count,
2523 'like_count': like_count,
2524 'dislike_count': dislike_count,
2525 'average_rating': average_rating,
2526 'formats': formats,
2527 'is_live': is_live,
2528 'start_time': start_time,
2529 'end_time': end_time,
2530 'series': series,
2531 'season_number': season_number,
2532 'episode_number': episode_number,
2533 'track': track,
2534 'artist': artist,
2535 'album': album,
2536 'release_date': release_date,
2537 'release_year': release_year,
2538 }
2539
2540
2541 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2542 IE_DESC = 'YouTube.com playlists'
2543 _VALID_URL = r"""(?x)(?:
2544 (?:https?://)?
2545 (?:\w+\.)?
2546 (?:
2547 (?:
2548 youtube(?:kids)?\.com|
2549 invidio\.us
2550 )
2551 /
2552 (?:
2553 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2554 \? (?:.*?[&;])*? (?:p|a|list)=
2555 | p/
2556 )|
2557 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2558 )
2559 (
2560 (?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2561 # Top tracks, they can also include dots
2562 |(?:MC)[\w\.]*
2563 )
2564 .*
2565 |
2566 (%(playlist_id)s)
2567 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2568 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2569 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2570 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2571 IE_NAME = 'youtube:playlist'
2572 _TESTS = [{
2573 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2574 'info_dict': {
2575 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2576 'uploader': 'Sergey M.',
2577 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc',
2578 'title': 'youtube-dl public playlist',
2579 },
2580 'playlist_count': 1,
2581 }, {
2582 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2583 'info_dict': {
2584 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA',
2585 'uploader': 'Sergey M.',
2586 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf',
2587 'title': 'youtube-dl empty playlist',
2588 },
2589 'playlist_count': 0,
2590 }, {
2591 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2592 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2593 'info_dict': {
2594 'title': '29C3: Not my department',
2595 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2596 'uploader': 'Christiaan008',
2597 'uploader_id': 'ChRiStIaAn008',
2598 },
2599 'playlist_count': 96,
2600 }, {
2601 'note': 'issue #673',
2602 'url': 'PLBB231211A4F62143',
2603 'info_dict': {
2604 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2605 'id': 'PLBB231211A4F62143',
2606 'uploader': 'Wickydoo',
2607 'uploader_id': 'Wickydoo',
2608 },
2609 'playlist_mincount': 26,
2610 }, {
2611 'note': 'Large playlist',
2612 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2613 'info_dict': {
2614 'title': 'Uploads from Cauchemar',
2615 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2616 'uploader': 'Cauchemar',
2617 'uploader_id': 'Cauchemar89',
2618 },
2619 'playlist_mincount': 799,
2620 }, {
2621 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2622 'info_dict': {
2623 'title': 'YDL_safe_search',
2624 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2625 },
2626 'playlist_count': 2,
2627 'skip': 'This playlist is private',
2628 }, {
2629 'note': 'embedded',
2630 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2631 'playlist_count': 4,
2632 'info_dict': {
2633 'title': 'JODA15',
2634 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2635 'uploader': 'milan',
2636 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2637 }
2638 }, {
2639 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2640 'playlist_mincount': 485,
2641 'info_dict': {
2642 'title': '2018 Chinese New Singles (11/6 updated)',
2643 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2644 'uploader': 'LBK',
2645 'uploader_id': 'sdragonfang',
2646 }
2647 }, {
2648 'note': 'Embedded SWF player',
2649 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2650 'playlist_count': 4,
2651 'info_dict': {
2652 'title': 'JODA7',
2653 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2654 },
2655 'skip': 'This playlist does not exist',
2656 }, {
2657 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2658 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2659 'info_dict': {
2660 'title': 'Uploads from Interstellar Movie',
2661 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2662 'uploader': 'Interstellar Movie',
2663 'uploader_id': 'InterstellarMovie1',
2664 },
2665 'playlist_mincount': 21,
2666 }, {
2667 # Playlist URL that does not actually serve a playlist
2668 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2669 'info_dict': {
2670 'id': 'FqZTN594JQw',
2671 'ext': 'webm',
2672 'title': "Smiley's People 01 detective, Adventure Series, Action",
2673 'uploader': 'STREEM',
2674 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2675 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2676 'upload_date': '20150526',
2677 'license': 'Standard YouTube License',
2678 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2679 'categories': ['People & Blogs'],
2680 'tags': list,
2681 'view_count': int,
2682 'like_count': int,
2683 'dislike_count': int,
2684 },
2685 'params': {
2686 'skip_download': True,
2687 },
2688 'skip': 'This video is not available.',
2689 'add_ie': [YoutubeIE.ie_key()],
2690 }, {
2691 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2692 'info_dict': {
2693 'id': 'yeWKywCrFtk',
2694 'ext': 'mp4',
2695 'title': 'Small Scale Baler and Braiding Rugs',
2696 'uploader': 'Backus-Page House Museum',
2697 'uploader_id': 'backuspagemuseum',
2698 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2699 'upload_date': '20161008',
2700 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2701 'categories': ['Nonprofits & Activism'],
2702 'tags': list,
2703 'like_count': int,
2704 'dislike_count': int,
2705 },
2706 'params': {
2707 'noplaylist': True,
2708 'skip_download': True,
2709 },
2710 }, {
2711 # https://github.com/ytdl-org/youtube-dl/issues/21844
2712 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2713 'info_dict': {
2714 'title': 'Data Analysis with Dr Mike Pound',
2715 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2716 'uploader_id': 'Computerphile',
2717 'uploader': 'Computerphile',
2718 },
2719 'playlist_mincount': 11,
2720 }, {
2721 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2722 'only_matching': True,
2723 }, {
2724 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2725 'only_matching': True,
2726 }, {
2727 # music album playlist
2728 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2729 'only_matching': True,
2730 }, {
2731 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2732 'only_matching': True,
2733 }, {
2734 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2735 'only_matching': True,
2736 }]
2737
2738 def _real_initialize(self):
2739 self._login()
2740
2741 def extract_videos_from_page(self, page):
2742 ids_in_page = []
2743 titles_in_page = []
2744
2745 for item in re.findall(
2746 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2747 attrs = extract_attributes(item)
2748 video_id = attrs['data-video-id']
2749 video_title = unescapeHTML(attrs.get('data-title'))
2750 if video_title:
2751 video_title = video_title.strip()
2752 ids_in_page.append(video_id)
2753 titles_in_page.append(video_title)
2754
2755 # Fallback with old _VIDEO_RE
2756 self.extract_videos_from_page_impl(
2757 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2758
2759 # Relaxed fallbacks
2760 self.extract_videos_from_page_impl(
2761 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2762 ids_in_page, titles_in_page)
2763 self.extract_videos_from_page_impl(
2764 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2765 ids_in_page, titles_in_page)
2766
2767 return zip(ids_in_page, titles_in_page)
2768
2769 def _extract_mix(self, playlist_id):
2770 # The mixes are generated from a single video
2771 # the id of the playlist is just 'RD' + video_id
2772 ids = []
2773 last_id = playlist_id[-11:]
2774 for n in itertools.count(1):
2775 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2776 webpage = self._download_webpage(
2777 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2778 new_ids = orderedSet(re.findall(
2779 r'''(?xs)data-video-username=".*?".*?
2780 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2781 webpage))
2782 # Fetch new pages until all the videos are repeated, it seems that
2783 # there are always 51 unique videos.
2784 new_ids = [_id for _id in new_ids if _id not in ids]
2785 if not new_ids:
2786 break
2787 ids.extend(new_ids)
2788 last_id = ids[-1]
2789
2790 url_results = self._ids_to_results(ids)
2791
2792 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2793 title_span = (
2794 search_title('playlist-title')
2795 or search_title('title long-title')
2796 or search_title('title'))
2797 title = clean_html(title_span)
2798
2799 return self.playlist_result(url_results, playlist_id, title)
2800
2801 def _extract_playlist(self, playlist_id):
2802 url = self._TEMPLATE_URL % playlist_id
2803 page = self._download_webpage(url, playlist_id)
2804
2805 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2806 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2807 match = match.strip()
2808 # Check if the playlist exists or is private
2809 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2810 if mobj:
2811 reason = mobj.group('reason')
2812 message = 'This playlist %s' % reason
2813 if 'private' in reason:
2814 message += ', use --username or --netrc to access it'
2815 message += '.'
2816 raise ExtractorError(message, expected=True)
2817 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2818 raise ExtractorError(
2819 'Invalid parameters. Maybe URL is incorrect.',
2820 expected=True)
2821 elif re.match(r'[^<]*Choose your language[^<]*', match):
2822 continue
2823 else:
2824 self.report_warning('Youtube gives an alert message: ' + match)
2825
2826 playlist_title = self._html_search_regex(
2827 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2828 page, 'title', default=None)
2829
2830 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2831 uploader = self._html_search_regex(
2832 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2833 page, 'uploader', default=None)
2834 mobj = re.search(
2835 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2836 page)
2837 if mobj:
2838 uploader_id = mobj.group('uploader_id')
2839 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2840 else:
2841 uploader_id = uploader_url = None
2842
2843 has_videos = True
2844
2845 if not playlist_title:
2846 try:
2847 # Some playlist URLs don't actually serve a playlist (e.g.
2848 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2849 next(self._entries(page, playlist_id))
2850 except StopIteration:
2851 has_videos = False
2852
2853 playlist = self.playlist_result(
2854 self._entries(page, playlist_id), playlist_id, playlist_title)
2855 playlist.update({
2856 'uploader': uploader,
2857 'uploader_id': uploader_id,
2858 'uploader_url': uploader_url,
2859 })
2860
2861 return has_videos, playlist
2862
2863 def _check_download_just_video(self, url, playlist_id):
2864 # Check if it's a video-specific URL
2865 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2866 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2867 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2868 'video id', default=None)
2869 if video_id:
2870 if self._downloader.params.get('noplaylist'):
2871 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2872 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2873 else:
2874 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2875 return video_id, None
2876 return None, None
2877
2878 def _real_extract(self, url):
2879 # Extract playlist id
2880 mobj = re.match(self._VALID_URL, url)
2881 if mobj is None:
2882 raise ExtractorError('Invalid URL: %s' % url)
2883 playlist_id = mobj.group(1) or mobj.group(2)
2884
2885 video_id, video = self._check_download_just_video(url, playlist_id)
2886 if video:
2887 return video
2888
2889 if playlist_id.startswith(('RD', 'UL', 'PU')):
2890 # Mixes require a custom extraction process
2891 return self._extract_mix(playlist_id)
2892
2893 has_videos, playlist = self._extract_playlist(playlist_id)
2894 if has_videos or not video_id:
2895 return playlist
2896
2897 # Some playlist URLs don't actually serve a playlist (see
2898 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2899 # Fallback to plain video extraction if there is a video id
2900 # along with playlist id.
2901 return self.url_result(video_id, 'Youtube', video_id=video_id)
2902
2903
2904 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2905 IE_DESC = 'YouTube.com channels'
2906 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie|kids)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2907 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2908 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2909 IE_NAME = 'youtube:channel'
2910 _TESTS = [{
2911 'note': 'paginated channel',
2912 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2913 'playlist_mincount': 91,
2914 'info_dict': {
2915 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2916 'title': 'Uploads from lex will',
2917 'uploader': 'lex will',
2918 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2919 }
2920 }, {
2921 'note': 'Age restricted channel',
2922 # from https://www.youtube.com/user/DeusExOfficial
2923 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2924 'playlist_mincount': 64,
2925 'info_dict': {
2926 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2927 'title': 'Uploads from Deus Ex',
2928 'uploader': 'Deus Ex',
2929 'uploader_id': 'DeusExOfficial',
2930 },
2931 }, {
2932 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2933 'only_matching': True,
2934 }, {
2935 'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA',
2936 'only_matching': True,
2937 }]
2938
2939 @classmethod
2940 def suitable(cls, url):
2941 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2942 else super(YoutubeChannelIE, cls).suitable(url))
2943
2944 def _build_template_url(self, url, channel_id):
2945 return self._TEMPLATE_URL % channel_id
2946
2947 def _real_extract(self, url):
2948 channel_id = self._match_id(url)
2949
2950 url = self._build_template_url(url, channel_id)
2951
2952 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2953 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2954 # otherwise fallback on channel by page extraction
2955 channel_page = self._download_webpage(
2956 url + '?view=57', channel_id,
2957 'Downloading channel page', fatal=False)
2958 if channel_page is False:
2959 channel_playlist_id = False
2960 else:
2961 channel_playlist_id = self._html_search_meta(
2962 'channelId', channel_page, 'channel id', default=None)
2963 if not channel_playlist_id:
2964 channel_url = self._html_search_meta(
2965 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2966 channel_page, 'channel url', default=None)
2967 if channel_url:
2968 channel_playlist_id = self._search_regex(
2969 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2970 channel_url, 'channel id', default=None)
2971 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2972 playlist_id = 'UU' + channel_playlist_id[2:]
2973 return self.url_result(
2974 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2975
2976 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2977 autogenerated = re.search(r'''(?x)
2978 class="[^"]*?(?:
2979 channel-header-autogenerated-label|
2980 yt-channel-title-autogenerated
2981 )[^"]*"''', channel_page) is not None
2982
2983 if autogenerated:
2984 # The videos are contained in a single page
2985 # the ajax pages can't be used, they are empty
2986 entries = [
2987 self.url_result(
2988 video_id, 'Youtube', video_id=video_id,
2989 video_title=video_title)
2990 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2991 return self.playlist_result(entries, channel_id)
2992
2993 try:
2994 next(self._entries(channel_page, channel_id))
2995 except StopIteration:
2996 alert_message = self._html_search_regex(
2997 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2998 channel_page, 'alert', default=None, group='alert')
2999 if alert_message:
3000 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
3001
3002 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
3003
3004
3005 class YoutubeUserIE(YoutubeChannelIE):
3006 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
3007 _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_-]+)'
3008 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
3009 IE_NAME = 'youtube:user'
3010
3011 _TESTS = [{
3012 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
3013 'playlist_mincount': 320,
3014 'info_dict': {
3015 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
3016 'title': 'Uploads from The Linux Foundation',
3017 'uploader': 'The Linux Foundation',
3018 'uploader_id': 'TheLinuxFoundation',
3019 }
3020 }, {
3021 # Only available via https://www.youtube.com/c/12minuteathlete/videos
3022 # but not https://www.youtube.com/user/12minuteathlete/videos
3023 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
3024 'playlist_mincount': 249,
3025 'info_dict': {
3026 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
3027 'title': 'Uploads from 12 Minute Athlete',
3028 'uploader': '12 Minute Athlete',
3029 'uploader_id': 'the12minuteathlete',
3030 }
3031 }, {
3032 'url': 'ytuser:phihag',
3033 'only_matching': True,
3034 }, {
3035 'url': 'https://www.youtube.com/c/gametrailers',
3036 'only_matching': True,
3037 }, {
3038 'url': 'https://www.youtube.com/gametrailers',
3039 'only_matching': True,
3040 }, {
3041 # This channel is not available, geo restricted to JP
3042 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
3043 'only_matching': True,
3044 }]
3045
3046 @classmethod
3047 def suitable(cls, url):
3048 # Don't return True if the url can be extracted with other youtube
3049 # extractor, the regex would is too permissive and it would match.
3050 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
3051 if any(ie.suitable(url) for ie in other_yt_ies):
3052 return False
3053 else:
3054 return super(YoutubeUserIE, cls).suitable(url)
3055
3056 def _build_template_url(self, url, channel_id):
3057 mobj = re.match(self._VALID_URL, url)
3058 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
3059
3060
3061 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
3062 IE_DESC = 'YouTube.com live streams'
3063 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
3064 IE_NAME = 'youtube:live'
3065
3066 _TESTS = [{
3067 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
3068 'info_dict': {
3069 'id': 'a48o2S1cPoo',
3070 'ext': 'mp4',
3071 'title': 'The Young Turks - Live Main Show',
3072 'uploader': 'The Young Turks',
3073 'uploader_id': 'TheYoungTurks',
3074 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
3075 'upload_date': '20150715',
3076 'license': 'Standard YouTube License',
3077 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
3078 'categories': ['News & Politics'],
3079 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
3080 'like_count': int,
3081 'dislike_count': int,
3082 },
3083 'params': {
3084 'skip_download': True,
3085 },
3086 }, {
3087 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
3088 'only_matching': True,
3089 }, {
3090 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3091 'only_matching': True,
3092 }, {
3093 'url': 'https://www.youtube.com/TheYoungTurks/live',
3094 'only_matching': True,
3095 }]
3096
3097 def _real_extract(self, url):
3098 mobj = re.match(self._VALID_URL, url)
3099 channel_id = mobj.group('id')
3100 base_url = mobj.group('base_url')
3101 webpage = self._download_webpage(url, channel_id, fatal=False)
3102 if webpage:
3103 page_type = self._og_search_property(
3104 'type', webpage, 'page type', default='')
3105 video_id = self._html_search_meta(
3106 'videoId', webpage, 'video id', default=None)
3107 if page_type.startswith('video') and video_id and re.match(
3108 r'^[0-9A-Za-z_-]{11}$', video_id):
3109 return self.url_result(video_id, YoutubeIE.ie_key())
3110 return self.url_result(base_url)
3111
3112
3113 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
3114 IE_DESC = 'YouTube.com user/channel playlists'
3115 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
3116 IE_NAME = 'youtube:playlists'
3117
3118 _TESTS = [{
3119 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3120 'playlist_mincount': 4,
3121 'info_dict': {
3122 'id': 'ThirstForScience',
3123 'title': 'ThirstForScience',
3124 },
3125 }, {
3126 # with "Load more" button
3127 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3128 'playlist_mincount': 70,
3129 'info_dict': {
3130 'id': 'igorkle1',
3131 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
3132 },
3133 }, {
3134 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3135 'playlist_mincount': 17,
3136 'info_dict': {
3137 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3138 'title': 'Chem Player',
3139 },
3140 'skip': 'Blocked',
3141 }]
3142
3143
3144 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3145 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3146
3147
3148 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3149 IE_DESC = 'YouTube.com searches'
3150 # there doesn't appear to be a real limit, for example if you search for
3151 # 'python' you get more than 8.000.000 results
3152 _MAX_RESULTS = float('inf')
3153 IE_NAME = 'youtube:search'
3154 _SEARCH_KEY = 'ytsearch'
3155 _EXTRA_QUERY_ARGS = {}
3156 _TESTS = []
3157
3158 def _get_n_results(self, query, n):
3159 """Get a specified number of results for a query"""
3160
3161 videos = []
3162 limit = n
3163
3164 url_query = {
3165 'search_query': query.encode('utf-8'),
3166 }
3167 url_query.update(self._EXTRA_QUERY_ARGS)
3168 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3169
3170 for pagenum in itertools.count(1):
3171 data = self._download_json(
3172 result_url, video_id='query "%s"' % query,
3173 note='Downloading page %s' % pagenum,
3174 errnote='Unable to download API page',
3175 query={'spf': 'navigate'})
3176 html_content = data[1]['body']['content']
3177
3178 if 'class="search-message' in html_content:
3179 raise ExtractorError(
3180 '[youtube] No video results', expected=True)
3181
3182 new_videos = list(self._process_page(html_content))
3183 videos += new_videos
3184 if not new_videos or len(videos) > limit:
3185 break
3186 next_link = self._html_search_regex(
3187 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3188 html_content, 'next link', default=None)
3189 if next_link is None:
3190 break
3191 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3192
3193 if len(videos) > n:
3194 videos = videos[:n]
3195 return self.playlist_result(videos, query)
3196
3197
3198 class YoutubeSearchDateIE(YoutubeSearchIE):
3199 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3200 _SEARCH_KEY = 'ytsearchdate'
3201 IE_DESC = 'YouTube.com searches, newest videos first'
3202 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3203
3204
3205 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3206 IE_DESC = 'YouTube.com search URLs'
3207 IE_NAME = 'youtube:search_url'
3208 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3209 _TESTS = [{
3210 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3211 'playlist_mincount': 5,
3212 'info_dict': {
3213 'title': 'youtube-dl test video',
3214 }
3215 }, {
3216 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3217 'only_matching': True,
3218 }]
3219
3220 def _real_extract(self, url):
3221 mobj = re.match(self._VALID_URL, url)
3222 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3223 webpage = self._download_webpage(url, query)
3224 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3225
3226
3227 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3228 IE_DESC = 'YouTube.com (multi-season) shows'
3229 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3230 IE_NAME = 'youtube:show'
3231 _TESTS = [{
3232 'url': 'https://www.youtube.com/show/airdisasters',
3233 'playlist_mincount': 5,
3234 'info_dict': {
3235 'id': 'airdisasters',
3236 'title': 'Air Disasters',
3237 }
3238 }]
3239
3240 def _real_extract(self, url):
3241 playlist_id = self._match_id(url)
3242 return super(YoutubeShowIE, self)._real_extract(
3243 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3244
3245
3246 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3247 """
3248 Base class for feed extractors
3249 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3250 """
3251 _LOGIN_REQUIRED = True
3252
3253 @property
3254 def IE_NAME(self):
3255 return 'youtube:%s' % self._FEED_NAME
3256
3257 def _real_initialize(self):
3258 self._login()
3259
3260 def _entries(self, page):
3261 # The extraction process is the same as for playlists, but the regex
3262 # for the video ids doesn't contain an index
3263 ids = []
3264 more_widget_html = content_html = page
3265 for page_num in itertools.count(1):
3266 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3267
3268 # 'recommended' feed has infinite 'load more' and each new portion spins
3269 # the same videos in (sometimes) slightly different order, so we'll check
3270 # for unicity and break when portion has no new videos
3271 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3272 if not new_ids:
3273 break
3274
3275 ids.extend(new_ids)
3276
3277 for entry in self._ids_to_results(new_ids):
3278 yield entry
3279
3280 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3281 if not mobj:
3282 break
3283
3284 more = self._download_json(
3285 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3286 'Downloading page #%s' % page_num,
3287 transform_source=uppercase_escape,
3288 headers=self._YOUTUBE_CLIENT_HEADERS)
3289 content_html = more['content_html']
3290 more_widget_html = more['load_more_widget_html']
3291
3292 def _real_extract(self, url):
3293 page = self._download_webpage(
3294 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3295 self._PLAYLIST_TITLE)
3296 return self.playlist_result(
3297 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3298
3299
3300 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3301 IE_NAME = 'youtube:watchlater'
3302 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3303 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3304
3305 _TESTS = [{
3306 'url': 'https://www.youtube.com/playlist?list=WL',
3307 'only_matching': True,
3308 }, {
3309 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3310 'only_matching': True,
3311 }]
3312
3313 def _real_extract(self, url):
3314 _, video = self._check_download_just_video(url, 'WL')
3315 if video:
3316 return video
3317 _, playlist = self._extract_playlist('WL')
3318 return playlist
3319
3320
3321 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3322 IE_NAME = 'youtube:favorites'
3323 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3324 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3325 _LOGIN_REQUIRED = True
3326
3327 def _real_extract(self, url):
3328 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3329 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3330 return self.url_result(playlist_id, 'YoutubePlaylist')
3331
3332
3333 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3334 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3335 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3336 _FEED_NAME = 'recommended'
3337 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3338
3339
3340 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3341 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3342 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3343 _FEED_NAME = 'subscriptions'
3344 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3345
3346
3347 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3348 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3349 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3350 _FEED_NAME = 'history'
3351 _PLAYLIST_TITLE = 'Youtube History'
3352
3353
3354 class YoutubeTruncatedURLIE(InfoExtractor):
3355 IE_NAME = 'youtube:truncated_url'
3356 IE_DESC = False # Do not list
3357 _VALID_URL = r'''(?x)
3358 (?:https?://)?
3359 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3360 (?:watch\?(?:
3361 feature=[a-z_]+|
3362 annotation_id=annotation_[^&]+|
3363 x-yt-cl=[0-9]+|
3364 hl=[^&]*|
3365 t=[0-9]+
3366 )?
3367 |
3368 attribution_link\?a=[^&]+
3369 )
3370 $
3371 '''
3372
3373 _TESTS = [{
3374 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3375 'only_matching': True,
3376 }, {
3377 'url': 'https://www.youtube.com/watch?',
3378 'only_matching': True,
3379 }, {
3380 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3381 'only_matching': True,
3382 }, {
3383 'url': 'https://www.youtube.com/watch?feature=foo',
3384 'only_matching': True,
3385 }, {
3386 'url': 'https://www.youtube.com/watch?hl=en-GB',
3387 'only_matching': True,
3388 }, {
3389 'url': 'https://www.youtube.com/watch?t=2372',
3390 'only_matching': True,
3391 }]
3392
3393 def _real_extract(self, url):
3394 raise ExtractorError(
3395 'Did you forget to quote the URL? Remember that & is a meta '
3396 'character in most shells, so you want to put the URL in quotes, '
3397 'like youtube-dl '
3398 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3399 ' or simply youtube-dl BaW_jenozKc .',
3400 expected=True)
3401
3402
3403 class YoutubeTruncatedIDIE(InfoExtractor):
3404 IE_NAME = 'youtube:truncated_id'
3405 IE_DESC = False # Do not list
3406 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3407
3408 _TESTS = [{
3409 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3410 'only_matching': True,
3411 }]
3412
3413 def _real_extract(self, url):
3414 video_id = self._match_id(url)
3415 raise ExtractorError(
3416 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3417 expected=True)