]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
9a182fcf67138bf347c677fe69ca73bcc4fc8290
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_HTTPError,
20 compat_kwargs,
21 compat_parse_qs,
22 compat_urllib_parse_unquote,
23 compat_urllib_parse_unquote_plus,
24 compat_urllib_parse_urlencode,
25 compat_urllib_parse_urlparse,
26 compat_urlparse,
27 compat_str,
28 )
29 from ..utils import (
30 bool_or_none,
31 clean_html,
32 dict_get,
33 error_to_compat_str,
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 qualities,
44 remove_quotes,
45 remove_start,
46 smuggle_url,
47 str_or_none,
48 str_to_int,
49 try_get,
50 unescapeHTML,
51 unified_strdate,
52 unsmuggle_url,
53 uppercase_escape,
54 url_or_none,
55 urlencode_postdata,
56 )
57
58
59 class YoutubeBaseInfoExtractor(InfoExtractor):
60 """Provide base functions for Youtube extractors"""
61 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
62 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
63
64 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
65 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
66 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
67
68 _NETRC_MACHINE = 'youtube'
69 # If True it will raise an error if no login info is provided
70 _LOGIN_REQUIRED = False
71
72 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
73
74 def _set_language(self):
75 self._set_cookie(
76 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
77 # YouTube sets the expire time to about two months
78 expire_time=time.time() + 2 * 30 * 24 * 3600)
79
80 def _ids_to_results(self, ids):
81 return [
82 self.url_result(vid_id, 'Youtube', video_id=vid_id)
83 for vid_id in ids]
84
85 def _login(self):
86 """
87 Attempt to log in to YouTube.
88 True is returned if successful or skipped.
89 False is returned if login failed.
90
91 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
92 """
93 username, password = self._get_login_info()
94 # No authentication to be performed
95 if username is None:
96 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
97 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
98 return True
99
100 login_page = self._download_webpage(
101 self._LOGIN_URL, None,
102 note='Downloading login page',
103 errnote='unable to fetch login page', fatal=False)
104 if login_page is False:
105 return
106
107 login_form = self._hidden_inputs(login_page)
108
109 def req(url, f_req, note, errnote):
110 data = login_form.copy()
111 data.update({
112 'pstMsg': 1,
113 'checkConnection': 'youtube',
114 'checkedDomains': 'youtube',
115 'hl': 'en',
116 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
117 'f.req': json.dumps(f_req),
118 'flowName': 'GlifWebSignIn',
119 'flowEntry': 'ServiceLogin',
120 # TODO: reverse actual botguard identifier generation algo
121 'bgRequest': '["identifier",""]',
122 })
123 return self._download_json(
124 url, None, note=note, errnote=errnote,
125 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
126 fatal=False,
127 data=urlencode_postdata(data), headers={
128 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
129 'Google-Accounts-XSRF': 1,
130 })
131
132 def warn(message):
133 self._downloader.report_warning(message)
134
135 lookup_req = [
136 username,
137 None, [], None, 'US', None, None, 2, False, True,
138 [
139 None, None,
140 [2, 1, None, 1,
141 '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',
142 None, [], 4],
143 1, [None, None, []], None, None, None, True
144 ],
145 username,
146 ]
147
148 lookup_results = req(
149 self._LOOKUP_URL, lookup_req,
150 'Looking up account info', 'Unable to look up account info')
151
152 if lookup_results is False:
153 return False
154
155 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
156 if not user_hash:
157 warn('Unable to extract user hash')
158 return False
159
160 challenge_req = [
161 user_hash,
162 None, 1, None, [1, None, None, None, [password, None, True]],
163 [
164 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],
165 1, [None, None, []], None, None, None, True
166 ]]
167
168 challenge_results = req(
169 self._CHALLENGE_URL, challenge_req,
170 'Logging in', 'Unable to log in')
171
172 if challenge_results is False:
173 return
174
175 login_res = try_get(challenge_results, lambda x: x[0][5], list)
176 if login_res:
177 login_msg = try_get(login_res, lambda x: x[5], compat_str)
178 warn(
179 'Unable to login: %s' % 'Invalid password'
180 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
181 return False
182
183 res = try_get(challenge_results, lambda x: x[0][-1], list)
184 if not res:
185 warn('Unable to extract result entry')
186 return False
187
188 login_challenge = try_get(res, lambda x: x[0][0], list)
189 if login_challenge:
190 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
191 if challenge_str == 'TWO_STEP_VERIFICATION':
192 # SEND_SUCCESS - TFA code has been successfully sent to phone
193 # QUOTA_EXCEEDED - reached the limit of TFA codes
194 status = try_get(login_challenge, lambda x: x[5], compat_str)
195 if status == 'QUOTA_EXCEEDED':
196 warn('Exceeded the limit of TFA codes, try later')
197 return False
198
199 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
200 if not tl:
201 warn('Unable to extract TL')
202 return False
203
204 tfa_code = self._get_tfa_info('2-step verification code')
205
206 if not tfa_code:
207 warn(
208 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
209 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
210 return False
211
212 tfa_code = remove_start(tfa_code, 'G-')
213
214 tfa_req = [
215 user_hash, None, 2, None,
216 [
217 9, None, None, None, None, None, None, None,
218 [None, tfa_code, True, 2]
219 ]]
220
221 tfa_results = req(
222 self._TFA_URL.format(tl), tfa_req,
223 'Submitting TFA code', 'Unable to submit TFA code')
224
225 if tfa_results is False:
226 return False
227
228 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
229 if tfa_res:
230 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
231 warn(
232 'Unable to finish TFA: %s' % 'Invalid TFA code'
233 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
234 return False
235
236 check_cookie_url = try_get(
237 tfa_results, lambda x: x[0][-1][2], compat_str)
238 else:
239 CHALLENGES = {
240 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
241 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
242 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
243 }
244 challenge = CHALLENGES.get(
245 challenge_str,
246 '%s returned error %s.' % (self.IE_NAME, challenge_str))
247 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
248 return False
249 else:
250 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
251
252 if not check_cookie_url:
253 warn('Unable to extract CheckCookie URL')
254 return False
255
256 check_cookie_results = self._download_webpage(
257 check_cookie_url, None, 'Checking cookie', fatal=False)
258
259 if check_cookie_results is False:
260 return False
261
262 if 'https://myaccount.google.com/' not in check_cookie_results:
263 warn('Unable to log in')
264 return False
265
266 return True
267
268 def _download_webpage_handle(self, *args, **kwargs):
269 query = kwargs.get('query', {}).copy()
270 query['disable_polymer'] = 'true'
271 kwargs['query'] = query
272 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
273 *args, **compat_kwargs(kwargs))
274
275 def _real_initialize(self):
276 if self._downloader is None:
277 return
278 self._set_language()
279 if not self._login():
280 return
281
282
283 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
284 # Extract entries from page with "Load more" button
285 def _entries(self, page, playlist_id):
286 more_widget_html = content_html = page
287 for page_num in itertools.count(1):
288 for entry in self._process_page(content_html):
289 yield entry
290
291 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
292 if not mobj:
293 break
294
295 count = 0
296 retries = 3
297 while count <= retries:
298 try:
299 # Downloading page may result in intermittent 5xx HTTP error
300 # that is usually worked around with a retry
301 more = self._download_json(
302 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
303 'Downloading page #%s%s'
304 % (page_num, ' (retry #%d)' % count if count else ''),
305 transform_source=uppercase_escape)
306 break
307 except ExtractorError as e:
308 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
309 count += 1
310 if count <= retries:
311 continue
312 raise
313
314 content_html = more['content_html']
315 if not content_html.strip():
316 # Some webpages show a "Load more" button but they don't
317 # have more videos
318 break
319 more_widget_html = more['load_more_widget_html']
320
321
322 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
323 def _process_page(self, content):
324 for video_id, video_title in self.extract_videos_from_page(content):
325 yield self.url_result(video_id, 'Youtube', video_id, video_title)
326
327 def extract_videos_from_page(self, page):
328 ids_in_page = []
329 titles_in_page = []
330 for mobj in re.finditer(self._VIDEO_RE, page):
331 # The link with index 0 is not the first video of the playlist (not sure if still actual)
332 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
333 continue
334 video_id = mobj.group('id')
335 video_title = unescapeHTML(mobj.group('title'))
336 if video_title:
337 video_title = video_title.strip()
338 try:
339 idx = ids_in_page.index(video_id)
340 if video_title and not titles_in_page[idx]:
341 titles_in_page[idx] = video_title
342 except ValueError:
343 ids_in_page.append(video_id)
344 titles_in_page.append(video_title)
345 return zip(ids_in_page, titles_in_page)
346
347
348 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
349 def _process_page(self, content):
350 for playlist_id in orderedSet(re.findall(
351 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
352 content)):
353 yield self.url_result(
354 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
355
356 def _real_extract(self, url):
357 playlist_id = self._match_id(url)
358 webpage = self._download_webpage(url, playlist_id)
359 title = self._og_search_title(webpage, fatal=False)
360 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
361
362
363 class YoutubeIE(YoutubeBaseInfoExtractor):
364 IE_DESC = 'YouTube.com'
365 _VALID_URL = r"""(?x)^
366 (
367 (?:https?://|//) # http(s):// or protocol-independent URL
368 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
369 (?:www\.)?deturl\.com/www\.youtube\.com/|
370 (?:www\.)?pwnyoutube\.com/|
371 (?:www\.)?hooktube\.com/|
372 (?:www\.)?yourepeat\.com/|
373 tube\.majestyc\.net/|
374 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
375 (?:(?:www|dev)\.)?invidio\.us/|
376 (?:(?:www|no)\.)?invidiou\.sh/|
377 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
378 (?:www\.)?invidious\.kabi\.tk/|
379 (?:www\.)?invidious\.enkirton\.net/|
380 (?:www\.)?invidious\.13ad\.de/|
381 (?:www\.)?invidious\.mastodon\.host/|
382 (?:www\.)?tube\.poal\.co/|
383 (?:www\.)?vid\.wxzm\.sx/|
384 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
385 (?:.*?\#/)? # handle anchor (#/) redirect urls
386 (?: # the various things that can precede the ID:
387 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
388 |(?: # or the v= param in all its forms
389 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
390 (?:\?|\#!?) # the params delimiter ? or # or #!
391 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
392 v=
393 )
394 ))
395 |(?:
396 youtu\.be| # just youtu.be/xxxx
397 vid\.plus| # or vid.plus/xxxx
398 zwearz\.com/watch| # or zwearz.com/watch/xxxx
399 )/
400 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
401 )
402 )? # all until now is optional -> you can pass the naked ID
403 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
404 (?!.*?\blist=
405 (?:
406 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
407 WL # WL are handled by the watch later IE
408 )
409 )
410 (?(1).+)? # if we found the ID, everything can follow
411 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
412 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
413 _formats = {
414 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
415 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
416 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
417 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
418 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
419 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
420 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
421 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
422 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
423 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
424 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
425 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
426 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
427 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
428 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
429 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
430 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
431 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
432
433
434 # 3D videos
435 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
436 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
437 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
438 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
439 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
440 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
441 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
442
443 # Apple HTTP Live Streaming
444 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
445 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
446 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
447 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
448 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
449 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
450 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
451 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
452
453 # DASH mp4 video
454 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
455 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
456 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
457 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
458 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
459 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
460 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
461 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
462 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
463 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
464 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
465 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
466
467 # Dash mp4 audio
468 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
469 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
470 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
471 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
472 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
473 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
474 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
475
476 # Dash webm
477 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
478 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
479 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
480 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
481 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
482 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
483 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
484 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
485 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
486 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
487 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
488 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
489 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
490 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
491 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
492 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
493 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
494 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
495 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
496 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
497 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
498 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
499
500 # Dash webm audio
501 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
502 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
503
504 # Dash webm audio with opus inside
505 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
506 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
507 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
508
509 # RTMP (unnamed)
510 '_rtmp': {'protocol': 'rtmp'},
511
512 # av01 video only formats sometimes served with "unknown" codecs
513 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
514 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
515 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
516 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
517 }
518 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
519
520 _GEO_BYPASS = False
521
522 IE_NAME = 'youtube'
523 _TESTS = [
524 {
525 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
526 'info_dict': {
527 'id': 'BaW_jenozKc',
528 'ext': 'mp4',
529 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
530 'uploader': 'Philipp Hagemeister',
531 'uploader_id': 'phihag',
532 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
533 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
534 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
535 'upload_date': '20121002',
536 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
537 'categories': ['Science & Technology'],
538 'tags': ['youtube-dl'],
539 'duration': 10,
540 'view_count': int,
541 'like_count': int,
542 'dislike_count': int,
543 'start_time': 1,
544 'end_time': 9,
545 }
546 },
547 {
548 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
549 'note': 'Test generic use_cipher_signature video (#897)',
550 'info_dict': {
551 'id': 'UxxajLWwzqY',
552 'ext': 'mp4',
553 'upload_date': '20120506',
554 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
555 'alt_title': 'I Love It (feat. Charli XCX)',
556 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
557 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
558 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
559 'iconic ep', 'iconic', 'love', 'it'],
560 'duration': 180,
561 'uploader': 'Icona Pop',
562 'uploader_id': 'IconaPop',
563 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
564 'creator': 'Icona Pop',
565 'track': 'I Love It (feat. Charli XCX)',
566 'artist': 'Icona Pop',
567 }
568 },
569 {
570 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
571 'note': 'Test VEVO video with age protection (#956)',
572 'info_dict': {
573 'id': '07FYdnEawAQ',
574 'ext': 'mp4',
575 'upload_date': '20130703',
576 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
577 'alt_title': 'Tunnel Vision',
578 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
579 'duration': 419,
580 'uploader': 'justintimberlakeVEVO',
581 'uploader_id': 'justintimberlakeVEVO',
582 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
583 'creator': 'Justin Timberlake',
584 'track': 'Tunnel Vision',
585 'artist': 'Justin Timberlake',
586 'age_limit': 18,
587 }
588 },
589 {
590 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
591 'note': 'Embed-only video (#1746)',
592 'info_dict': {
593 'id': 'yZIXLfi8CZQ',
594 'ext': 'mp4',
595 'upload_date': '20120608',
596 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
597 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
598 'uploader': 'SET India',
599 'uploader_id': 'setindia',
600 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
601 'age_limit': 18,
602 }
603 },
604 {
605 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
606 'note': 'Use the first video ID in the URL',
607 'info_dict': {
608 'id': 'BaW_jenozKc',
609 'ext': 'mp4',
610 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
611 'uploader': 'Philipp Hagemeister',
612 'uploader_id': 'phihag',
613 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
614 'upload_date': '20121002',
615 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
616 'categories': ['Science & Technology'],
617 'tags': ['youtube-dl'],
618 'duration': 10,
619 'view_count': int,
620 'like_count': int,
621 'dislike_count': int,
622 },
623 'params': {
624 'skip_download': True,
625 },
626 },
627 {
628 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
629 'note': '256k DASH audio (format 141) via DASH manifest',
630 'info_dict': {
631 'id': 'a9LDPn-MO4I',
632 'ext': 'm4a',
633 'upload_date': '20121002',
634 'uploader_id': '8KVIDEO',
635 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
636 'description': '',
637 'uploader': '8KVIDEO',
638 'title': 'UHDTV TEST 8K VIDEO.mp4'
639 },
640 'params': {
641 'youtube_include_dash_manifest': True,
642 'format': '141',
643 },
644 'skip': 'format 141 not served anymore',
645 },
646 # DASH manifest with encrypted signature
647 {
648 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
649 'info_dict': {
650 'id': 'IB3lcPjvWLA',
651 'ext': 'm4a',
652 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
653 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
654 'duration': 244,
655 'uploader': 'AfrojackVEVO',
656 'uploader_id': 'AfrojackVEVO',
657 'upload_date': '20131011',
658 },
659 'params': {
660 'youtube_include_dash_manifest': True,
661 'format': '141/bestaudio[ext=m4a]',
662 },
663 },
664 # JS player signature function name containing $
665 {
666 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
667 'info_dict': {
668 'id': 'nfWlot6h_JM',
669 'ext': 'm4a',
670 'title': 'Taylor Swift - Shake It Off',
671 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
672 'duration': 242,
673 'uploader': 'TaylorSwiftVEVO',
674 'uploader_id': 'TaylorSwiftVEVO',
675 'upload_date': '20140818',
676 'creator': 'Taylor Swift',
677 },
678 'params': {
679 'youtube_include_dash_manifest': True,
680 'format': '141/bestaudio[ext=m4a]',
681 },
682 },
683 # Controversy video
684 {
685 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
686 'info_dict': {
687 'id': 'T4XJQO3qol8',
688 'ext': 'mp4',
689 'duration': 219,
690 'upload_date': '20100909',
691 'uploader': 'Amazing Atheist',
692 'uploader_id': 'TheAmazingAtheist',
693 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
694 'title': 'Burning Everyone\'s Koran',
695 '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',
696 }
697 },
698 # Normal age-gate video (No vevo, embed allowed)
699 {
700 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
701 'info_dict': {
702 'id': 'HtVdAasjOgU',
703 'ext': 'mp4',
704 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
705 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
706 'duration': 142,
707 'uploader': 'The Witcher',
708 'uploader_id': 'WitcherGame',
709 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
710 'upload_date': '20140605',
711 'age_limit': 18,
712 },
713 },
714 # Age-gate video with encrypted signature
715 {
716 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
717 'info_dict': {
718 'id': '6kLq3WMV1nU',
719 'ext': 'mp4',
720 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
721 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
722 'duration': 246,
723 'uploader': 'LloydVEVO',
724 'uploader_id': 'LloydVEVO',
725 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
726 'upload_date': '20110629',
727 'age_limit': 18,
728 },
729 },
730 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
731 # YouTube Red ad is not captured for creator
732 {
733 'url': '__2ABJjxzNo',
734 'info_dict': {
735 'id': '__2ABJjxzNo',
736 'ext': 'mp4',
737 'duration': 266,
738 'upload_date': '20100430',
739 'uploader_id': 'deadmau5',
740 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
741 'creator': 'deadmau5',
742 'description': 'md5:12c56784b8032162bb936a5f76d55360',
743 'uploader': 'deadmau5',
744 'title': 'Deadmau5 - Some Chords (HD)',
745 'alt_title': 'Some Chords',
746 },
747 'expected_warnings': [
748 'DASH manifest missing',
749 ]
750 },
751 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
752 {
753 'url': 'lqQg6PlCWgI',
754 'info_dict': {
755 'id': 'lqQg6PlCWgI',
756 'ext': 'mp4',
757 'duration': 6085,
758 'upload_date': '20150827',
759 'uploader_id': 'olympic',
760 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
761 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
762 'uploader': 'Olympic',
763 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
764 },
765 'params': {
766 'skip_download': 'requires avconv',
767 }
768 },
769 # Non-square pixels
770 {
771 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
772 'info_dict': {
773 'id': '_b-2C3KPAM0',
774 'ext': 'mp4',
775 'stretched_ratio': 16 / 9.,
776 'duration': 85,
777 'upload_date': '20110310',
778 'uploader_id': 'AllenMeow',
779 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
780 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
781 'uploader': '孫ᄋᄅ',
782 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
783 },
784 },
785 # url_encoded_fmt_stream_map is empty string
786 {
787 'url': 'qEJwOuvDf7I',
788 'info_dict': {
789 'id': 'qEJwOuvDf7I',
790 'ext': 'webm',
791 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
792 'description': '',
793 'upload_date': '20150404',
794 'uploader_id': 'spbelect',
795 'uploader': 'Наблюдатели Петербурга',
796 },
797 'params': {
798 'skip_download': 'requires avconv',
799 },
800 'skip': 'This live event has ended.',
801 },
802 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
803 {
804 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
805 'info_dict': {
806 'id': 'FIl7x6_3R5Y',
807 'ext': 'webm',
808 'title': 'md5:7b81415841e02ecd4313668cde88737a',
809 'description': 'md5:116377fd2963b81ec4ce64b542173306',
810 'duration': 220,
811 'upload_date': '20150625',
812 'uploader_id': 'dorappi2000',
813 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
814 'uploader': 'dorappi2000',
815 'formats': 'mincount:31',
816 },
817 'skip': 'not actual anymore',
818 },
819 # DASH manifest with segment_list
820 {
821 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
822 'md5': '8ce563a1d667b599d21064e982ab9e31',
823 'info_dict': {
824 'id': 'CsmdDsKjzN8',
825 'ext': 'mp4',
826 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
827 'uploader': 'Airtek',
828 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
829 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
830 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
831 },
832 'params': {
833 'youtube_include_dash_manifest': True,
834 'format': '135', # bestvideo
835 },
836 'skip': 'This live event has ended.',
837 },
838 {
839 # Multifeed videos (multiple cameras), URL is for Main Camera
840 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
841 'info_dict': {
842 'id': 'jqWvoWXjCVs',
843 'title': 'teamPGP: Rocket League Noob Stream',
844 'description': 'md5:dc7872fb300e143831327f1bae3af010',
845 },
846 'playlist': [{
847 'info_dict': {
848 'id': 'jqWvoWXjCVs',
849 'ext': 'mp4',
850 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
851 'description': 'md5:dc7872fb300e143831327f1bae3af010',
852 'duration': 7335,
853 'upload_date': '20150721',
854 'uploader': 'Beer Games Beer',
855 'uploader_id': 'beergamesbeer',
856 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
857 'license': 'Standard YouTube License',
858 },
859 }, {
860 'info_dict': {
861 'id': '6h8e8xoXJzg',
862 'ext': 'mp4',
863 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
864 'description': 'md5:dc7872fb300e143831327f1bae3af010',
865 'duration': 7337,
866 'upload_date': '20150721',
867 'uploader': 'Beer Games Beer',
868 'uploader_id': 'beergamesbeer',
869 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
870 'license': 'Standard YouTube License',
871 },
872 }, {
873 'info_dict': {
874 'id': 'PUOgX5z9xZw',
875 'ext': 'mp4',
876 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
877 'description': 'md5:dc7872fb300e143831327f1bae3af010',
878 'duration': 7337,
879 'upload_date': '20150721',
880 'uploader': 'Beer Games Beer',
881 'uploader_id': 'beergamesbeer',
882 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
883 'license': 'Standard YouTube License',
884 },
885 }, {
886 'info_dict': {
887 'id': 'teuwxikvS5k',
888 'ext': 'mp4',
889 'title': 'teamPGP: Rocket League Noob Stream (zim)',
890 'description': 'md5:dc7872fb300e143831327f1bae3af010',
891 'duration': 7334,
892 'upload_date': '20150721',
893 'uploader': 'Beer Games Beer',
894 'uploader_id': 'beergamesbeer',
895 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
896 'license': 'Standard YouTube License',
897 },
898 }],
899 'params': {
900 'skip_download': True,
901 },
902 'skip': 'This video is not available.',
903 },
904 {
905 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
906 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
907 'info_dict': {
908 'id': 'gVfLd0zydlo',
909 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
910 },
911 'playlist_count': 2,
912 'skip': 'Not multifeed anymore',
913 },
914 {
915 'url': 'https://vid.plus/FlRa-iH7PGw',
916 'only_matching': True,
917 },
918 {
919 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
920 'only_matching': True,
921 },
922 {
923 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
924 # Also tests cut-off URL expansion in video description (see
925 # https://github.com/ytdl-org/youtube-dl/issues/1892,
926 # https://github.com/ytdl-org/youtube-dl/issues/8164)
927 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
928 'info_dict': {
929 'id': 'lsguqyKfVQg',
930 'ext': 'mp4',
931 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
932 'alt_title': 'Dark Walk - Position Music',
933 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
934 'duration': 133,
935 'upload_date': '20151119',
936 'uploader_id': 'IronSoulElf',
937 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
938 'uploader': 'IronSoulElf',
939 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
940 'track': 'Dark Walk - Position Music',
941 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
942 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
943 },
944 'params': {
945 'skip_download': True,
946 },
947 },
948 {
949 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
950 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
951 'only_matching': True,
952 },
953 {
954 # Video with yt:stretch=17:0
955 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
956 'info_dict': {
957 'id': 'Q39EVAstoRM',
958 'ext': 'mp4',
959 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
960 'description': 'md5:ee18a25c350637c8faff806845bddee9',
961 'upload_date': '20151107',
962 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
963 'uploader': 'CH GAMER DROID',
964 },
965 'params': {
966 'skip_download': True,
967 },
968 'skip': 'This video does not exist.',
969 },
970 {
971 # Video licensed under Creative Commons
972 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
973 'info_dict': {
974 'id': 'M4gD1WSo5mA',
975 'ext': 'mp4',
976 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
977 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
978 'duration': 721,
979 'upload_date': '20150127',
980 'uploader_id': 'BerkmanCenter',
981 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
982 'uploader': 'The Berkman Klein Center for Internet & Society',
983 'license': 'Creative Commons Attribution license (reuse allowed)',
984 },
985 'params': {
986 'skip_download': True,
987 },
988 },
989 {
990 # Channel-like uploader_url
991 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
992 'info_dict': {
993 'id': 'eQcmzGIKrzg',
994 'ext': 'mp4',
995 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
996 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
997 'duration': 4060,
998 'upload_date': '20151119',
999 'uploader': 'Bernie Sanders',
1000 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1001 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1002 'license': 'Creative Commons Attribution license (reuse allowed)',
1003 },
1004 'params': {
1005 'skip_download': True,
1006 },
1007 },
1008 {
1009 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1010 'only_matching': True,
1011 },
1012 {
1013 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1014 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1015 'only_matching': True,
1016 },
1017 {
1018 # Rental video preview
1019 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1020 'info_dict': {
1021 'id': 'uGpuVWrhIzE',
1022 'ext': 'mp4',
1023 'title': 'Piku - Trailer',
1024 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1025 'upload_date': '20150811',
1026 'uploader': 'FlixMatrix',
1027 'uploader_id': 'FlixMatrixKaravan',
1028 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1029 'license': 'Standard YouTube License',
1030 },
1031 'params': {
1032 'skip_download': True,
1033 },
1034 'skip': 'This video is not available.',
1035 },
1036 {
1037 # YouTube Red video with episode data
1038 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1039 'info_dict': {
1040 'id': 'iqKdEhx-dD4',
1041 'ext': 'mp4',
1042 'title': 'Isolation - Mind Field (Ep 1)',
1043 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1044 'duration': 2085,
1045 'upload_date': '20170118',
1046 'uploader': 'Vsauce',
1047 'uploader_id': 'Vsauce',
1048 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1049 'series': 'Mind Field',
1050 'season_number': 1,
1051 'episode_number': 1,
1052 },
1053 'params': {
1054 'skip_download': True,
1055 },
1056 'expected_warnings': [
1057 'Skipping DASH manifest',
1058 ],
1059 },
1060 {
1061 # The following content has been identified by the YouTube community
1062 # as inappropriate or offensive to some audiences.
1063 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1064 'info_dict': {
1065 'id': '6SJNVb0GnPI',
1066 'ext': 'mp4',
1067 'title': 'Race Differences in Intelligence',
1068 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1069 'duration': 965,
1070 'upload_date': '20140124',
1071 'uploader': 'New Century Foundation',
1072 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1073 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1074 },
1075 'params': {
1076 'skip_download': True,
1077 },
1078 },
1079 {
1080 # itag 212
1081 'url': '1t24XAntNCY',
1082 'only_matching': True,
1083 },
1084 {
1085 # geo restricted to JP
1086 'url': 'sJL6WA-aGkQ',
1087 'only_matching': True,
1088 },
1089 {
1090 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1091 'only_matching': True,
1092 },
1093 {
1094 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1095 'only_matching': True,
1096 },
1097 {
1098 # DRM protected
1099 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1100 'only_matching': True,
1101 },
1102 {
1103 # Video with unsupported adaptive stream type formats
1104 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1105 'info_dict': {
1106 'id': 'Z4Vy8R84T1U',
1107 'ext': 'mp4',
1108 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1109 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1110 'duration': 433,
1111 'upload_date': '20130923',
1112 'uploader': 'Amelia Putri Harwita',
1113 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1114 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1115 'formats': 'maxcount:10',
1116 },
1117 'params': {
1118 'skip_download': True,
1119 'youtube_include_dash_manifest': False,
1120 },
1121 },
1122 {
1123 # Youtube Music Auto-generated description
1124 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1125 'info_dict': {
1126 'id': 'MgNrAu2pzNs',
1127 'ext': 'mp4',
1128 'title': 'Voyeur Girl',
1129 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1130 'upload_date': '20190312',
1131 'uploader': 'Various Artists - Topic',
1132 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1133 'artist': 'Stephen',
1134 'track': 'Voyeur Girl',
1135 'album': 'it\'s too much love to know my dear',
1136 'release_date': '20190313',
1137 'release_year': 2019,
1138 },
1139 'params': {
1140 'skip_download': True,
1141 },
1142 },
1143 {
1144 # Youtube Music Auto-generated description
1145 # Retrieve 'artist' field from 'Artist:' in video description
1146 # when it is present on youtube music video
1147 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1148 'info_dict': {
1149 'id': 'k0jLE7tTwjY',
1150 'ext': 'mp4',
1151 'title': 'Latch Feat. Sam Smith',
1152 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1153 'upload_date': '20150110',
1154 'uploader': 'Various Artists - Topic',
1155 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1156 'artist': 'Disclosure',
1157 'track': 'Latch Feat. Sam Smith',
1158 'album': 'Latch Featuring Sam Smith',
1159 'release_date': '20121008',
1160 'release_year': 2012,
1161 },
1162 'params': {
1163 'skip_download': True,
1164 },
1165 },
1166 {
1167 # Youtube Music Auto-generated description
1168 # handle multiple artists on youtube music video
1169 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1170 'info_dict': {
1171 'id': '74qn0eJSjpA',
1172 'ext': 'mp4',
1173 'title': 'Eastside',
1174 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1175 'upload_date': '20180710',
1176 'uploader': 'Benny Blanco - Topic',
1177 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1178 'artist': 'benny blanco, Halsey, Khalid',
1179 'track': 'Eastside',
1180 'album': 'Eastside',
1181 'release_date': '20180713',
1182 'release_year': 2018,
1183 },
1184 'params': {
1185 'skip_download': True,
1186 },
1187 },
1188 {
1189 # Youtube Music Auto-generated description
1190 # handle youtube music video with release_year and no release_date
1191 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1192 'info_dict': {
1193 'id': '-hcAI0g-f5M',
1194 'ext': 'mp4',
1195 'title': 'Put It On Me',
1196 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1197 'upload_date': '20180426',
1198 'uploader': 'Matt Maeson - Topic',
1199 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1200 'artist': 'Matt Maeson',
1201 'track': 'Put It On Me',
1202 'album': 'The Hearse',
1203 'release_date': None,
1204 'release_year': 2018,
1205 },
1206 'params': {
1207 'skip_download': True,
1208 },
1209 },
1210 ]
1211
1212 def __init__(self, *args, **kwargs):
1213 super(YoutubeIE, self).__init__(*args, **kwargs)
1214 self._player_cache = {}
1215
1216 def report_video_info_webpage_download(self, video_id):
1217 """Report attempt to download video info webpage."""
1218 self.to_screen('%s: Downloading video info webpage' % video_id)
1219
1220 def report_information_extraction(self, video_id):
1221 """Report attempt to extract video information."""
1222 self.to_screen('%s: Extracting video information' % video_id)
1223
1224 def report_unavailable_format(self, video_id, format):
1225 """Report extracted video URL."""
1226 self.to_screen('%s: Format %s not available' % (video_id, format))
1227
1228 def report_rtmp_download(self):
1229 """Indicate the download will use the RTMP protocol."""
1230 self.to_screen('RTMP download detected')
1231
1232 def _signature_cache_id(self, example_sig):
1233 """ Return a string representation of a signature """
1234 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1235
1236 def _extract_signature_function(self, video_id, player_url, example_sig):
1237 id_m = re.match(
1238 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1239 player_url)
1240 if not id_m:
1241 raise ExtractorError('Cannot identify player %r' % player_url)
1242 player_type = id_m.group('ext')
1243 player_id = id_m.group('id')
1244
1245 # Read from filesystem cache
1246 func_id = '%s_%s_%s' % (
1247 player_type, player_id, self._signature_cache_id(example_sig))
1248 assert os.path.basename(func_id) == func_id
1249
1250 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1251 if cache_spec is not None:
1252 return lambda s: ''.join(s[i] for i in cache_spec)
1253
1254 download_note = (
1255 'Downloading player %s' % player_url
1256 if self._downloader.params.get('verbose') else
1257 'Downloading %s player %s' % (player_type, player_id)
1258 )
1259 if player_type == 'js':
1260 code = self._download_webpage(
1261 player_url, video_id,
1262 note=download_note,
1263 errnote='Download of %s failed' % player_url)
1264 res = self._parse_sig_js(code)
1265 elif player_type == 'swf':
1266 urlh = self._request_webpage(
1267 player_url, video_id,
1268 note=download_note,
1269 errnote='Download of %s failed' % player_url)
1270 code = urlh.read()
1271 res = self._parse_sig_swf(code)
1272 else:
1273 assert False, 'Invalid player type %r' % player_type
1274
1275 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1276 cache_res = res(test_string)
1277 cache_spec = [ord(c) for c in cache_res]
1278
1279 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1280 return res
1281
1282 def _print_sig_code(self, func, example_sig):
1283 def gen_sig_code(idxs):
1284 def _genslice(start, end, step):
1285 starts = '' if start == 0 else str(start)
1286 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1287 steps = '' if step == 1 else (':%d' % step)
1288 return 's[%s%s%s]' % (starts, ends, steps)
1289
1290 step = None
1291 # Quelch pyflakes warnings - start will be set when step is set
1292 start = '(Never used)'
1293 for i, prev in zip(idxs[1:], idxs[:-1]):
1294 if step is not None:
1295 if i - prev == step:
1296 continue
1297 yield _genslice(start, prev, step)
1298 step = None
1299 continue
1300 if i - prev in [-1, 1]:
1301 step = i - prev
1302 start = prev
1303 continue
1304 else:
1305 yield 's[%d]' % prev
1306 if step is None:
1307 yield 's[%d]' % i
1308 else:
1309 yield _genslice(start, i, step)
1310
1311 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1312 cache_res = func(test_string)
1313 cache_spec = [ord(c) for c in cache_res]
1314 expr_code = ' + '.join(gen_sig_code(cache_spec))
1315 signature_id_tuple = '(%s)' % (
1316 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1317 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1318 ' return %s\n') % (signature_id_tuple, expr_code)
1319 self.to_screen('Extracted signature function:\n' + code)
1320
1321 def _parse_sig_js(self, jscode):
1322 funcname = self._search_regex(
1323 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1324 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1325 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1326 # Obsolete patterns
1327 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1328 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1329 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1330 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1331 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1332 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1333 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1334 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1335 jscode, 'Initial JS player signature function name', group='sig')
1336
1337 jsi = JSInterpreter(jscode)
1338 initial_function = jsi.extract_function(funcname)
1339 return lambda s: initial_function([s])
1340
1341 def _parse_sig_swf(self, file_contents):
1342 swfi = SWFInterpreter(file_contents)
1343 TARGET_CLASSNAME = 'SignatureDecipher'
1344 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1345 initial_function = swfi.extract_function(searched_class, 'decipher')
1346 return lambda s: initial_function([s])
1347
1348 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1349 """Turn the encrypted s field into a working signature"""
1350
1351 if player_url is None:
1352 raise ExtractorError('Cannot decrypt signature without player_url')
1353
1354 if player_url.startswith('//'):
1355 player_url = 'https:' + player_url
1356 elif not re.match(r'https?://', player_url):
1357 player_url = compat_urlparse.urljoin(
1358 'https://www.youtube.com', player_url)
1359 try:
1360 player_id = (player_url, self._signature_cache_id(s))
1361 if player_id not in self._player_cache:
1362 func = self._extract_signature_function(
1363 video_id, player_url, s
1364 )
1365 self._player_cache[player_id] = func
1366 func = self._player_cache[player_id]
1367 if self._downloader.params.get('youtube_print_sig_code'):
1368 self._print_sig_code(func, s)
1369 return func(s)
1370 except Exception as e:
1371 tb = traceback.format_exc()
1372 raise ExtractorError(
1373 'Signature extraction failed: ' + tb, cause=e)
1374
1375 def _get_subtitles(self, video_id, webpage):
1376 try:
1377 subs_doc = self._download_xml(
1378 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1379 video_id, note=False)
1380 except ExtractorError as err:
1381 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1382 return {}
1383
1384 sub_lang_list = {}
1385 for track in subs_doc.findall('track'):
1386 lang = track.attrib['lang_code']
1387 if lang in sub_lang_list:
1388 continue
1389 sub_formats = []
1390 for ext in self._SUBTITLE_FORMATS:
1391 params = compat_urllib_parse_urlencode({
1392 'lang': lang,
1393 'v': video_id,
1394 'fmt': ext,
1395 'name': track.attrib['name'].encode('utf-8'),
1396 })
1397 sub_formats.append({
1398 'url': 'https://www.youtube.com/api/timedtext?' + params,
1399 'ext': ext,
1400 })
1401 sub_lang_list[lang] = sub_formats
1402 if not sub_lang_list:
1403 self._downloader.report_warning('video doesn\'t have subtitles')
1404 return {}
1405 return sub_lang_list
1406
1407 def _get_ytplayer_config(self, video_id, webpage):
1408 patterns = (
1409 # User data may contain arbitrary character sequences that may affect
1410 # JSON extraction with regex, e.g. when '};' is contained the second
1411 # regex won't capture the whole JSON. Yet working around by trying more
1412 # concrete regex first keeping in mind proper quoted string handling
1413 # to be implemented in future that will replace this workaround (see
1414 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1415 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1416 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1417 r';ytplayer\.config\s*=\s*({.+?});',
1418 )
1419 config = self._search_regex(
1420 patterns, webpage, 'ytplayer.config', default=None)
1421 if config:
1422 return self._parse_json(
1423 uppercase_escape(config), video_id, fatal=False)
1424
1425 def _get_automatic_captions(self, video_id, webpage):
1426 """We need the webpage for getting the captions url, pass it as an
1427 argument to speed up the process."""
1428 self.to_screen('%s: Looking for automatic captions' % video_id)
1429 player_config = self._get_ytplayer_config(video_id, webpage)
1430 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1431 if not player_config:
1432 self._downloader.report_warning(err_msg)
1433 return {}
1434 try:
1435 args = player_config['args']
1436 caption_url = args.get('ttsurl')
1437 if caption_url:
1438 timestamp = args['timestamp']
1439 # We get the available subtitles
1440 list_params = compat_urllib_parse_urlencode({
1441 'type': 'list',
1442 'tlangs': 1,
1443 'asrs': 1,
1444 })
1445 list_url = caption_url + '&' + list_params
1446 caption_list = self._download_xml(list_url, video_id)
1447 original_lang_node = caption_list.find('track')
1448 if original_lang_node is None:
1449 self._downloader.report_warning('Video doesn\'t have automatic captions')
1450 return {}
1451 original_lang = original_lang_node.attrib['lang_code']
1452 caption_kind = original_lang_node.attrib.get('kind', '')
1453
1454 sub_lang_list = {}
1455 for lang_node in caption_list.findall('target'):
1456 sub_lang = lang_node.attrib['lang_code']
1457 sub_formats = []
1458 for ext in self._SUBTITLE_FORMATS:
1459 params = compat_urllib_parse_urlencode({
1460 'lang': original_lang,
1461 'tlang': sub_lang,
1462 'fmt': ext,
1463 'ts': timestamp,
1464 'kind': caption_kind,
1465 })
1466 sub_formats.append({
1467 'url': caption_url + '&' + params,
1468 'ext': ext,
1469 })
1470 sub_lang_list[sub_lang] = sub_formats
1471 return sub_lang_list
1472
1473 def make_captions(sub_url, sub_langs):
1474 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1475 caption_qs = compat_parse_qs(parsed_sub_url.query)
1476 captions = {}
1477 for sub_lang in sub_langs:
1478 sub_formats = []
1479 for ext in self._SUBTITLE_FORMATS:
1480 caption_qs.update({
1481 'tlang': [sub_lang],
1482 'fmt': [ext],
1483 })
1484 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1485 query=compat_urllib_parse_urlencode(caption_qs, True)))
1486 sub_formats.append({
1487 'url': sub_url,
1488 'ext': ext,
1489 })
1490 captions[sub_lang] = sub_formats
1491 return captions
1492
1493 # New captions format as of 22.06.2017
1494 player_response = args.get('player_response')
1495 if player_response and isinstance(player_response, compat_str):
1496 player_response = self._parse_json(
1497 player_response, video_id, fatal=False)
1498 if player_response:
1499 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1500 base_url = renderer['captionTracks'][0]['baseUrl']
1501 sub_lang_list = []
1502 for lang in renderer['translationLanguages']:
1503 lang_code = lang.get('languageCode')
1504 if lang_code:
1505 sub_lang_list.append(lang_code)
1506 return make_captions(base_url, sub_lang_list)
1507
1508 # Some videos don't provide ttsurl but rather caption_tracks and
1509 # caption_translation_languages (e.g. 20LmZk1hakA)
1510 # Does not used anymore as of 22.06.2017
1511 caption_tracks = args['caption_tracks']
1512 caption_translation_languages = args['caption_translation_languages']
1513 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1514 sub_lang_list = []
1515 for lang in caption_translation_languages.split(','):
1516 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1517 sub_lang = lang_qs.get('lc', [None])[0]
1518 if sub_lang:
1519 sub_lang_list.append(sub_lang)
1520 return make_captions(caption_url, sub_lang_list)
1521 # An extractor error can be raise by the download process if there are
1522 # no automatic captions but there are subtitles
1523 except (KeyError, IndexError, ExtractorError):
1524 self._downloader.report_warning(err_msg)
1525 return {}
1526
1527 def _mark_watched(self, video_id, video_info, player_response):
1528 playback_url = url_or_none(try_get(
1529 player_response,
1530 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1531 video_info, lambda x: x['videostats_playback_base_url'][0]))
1532 if not playback_url:
1533 return
1534 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1535 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1536
1537 # cpn generation algorithm is reverse engineered from base.js.
1538 # In fact it works even with dummy cpn.
1539 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1540 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1541
1542 qs.update({
1543 'ver': ['2'],
1544 'cpn': [cpn],
1545 })
1546 playback_url = compat_urlparse.urlunparse(
1547 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1548
1549 self._download_webpage(
1550 playback_url, video_id, 'Marking watched',
1551 'Unable to mark watched', fatal=False)
1552
1553 @staticmethod
1554 def _extract_urls(webpage):
1555 # Embedded YouTube player
1556 entries = [
1557 unescapeHTML(mobj.group('url'))
1558 for mobj in re.finditer(r'''(?x)
1559 (?:
1560 <iframe[^>]+?src=|
1561 data-video-url=|
1562 <embed[^>]+?src=|
1563 embedSWF\(?:\s*|
1564 <object[^>]+data=|
1565 new\s+SWFObject\(
1566 )
1567 (["\'])
1568 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1569 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1570 \1''', webpage)]
1571
1572 # lazyYT YouTube embed
1573 entries.extend(list(map(
1574 unescapeHTML,
1575 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1576
1577 # Wordpress "YouTube Video Importer" plugin
1578 matches = re.findall(r'''(?x)<div[^>]+
1579 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1580 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1581 entries.extend(m[-1] for m in matches)
1582
1583 return entries
1584
1585 @staticmethod
1586 def _extract_url(webpage):
1587 urls = YoutubeIE._extract_urls(webpage)
1588 return urls[0] if urls else None
1589
1590 @classmethod
1591 def extract_id(cls, url):
1592 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1593 if mobj is None:
1594 raise ExtractorError('Invalid URL: %s' % url)
1595 video_id = mobj.group(2)
1596 return video_id
1597
1598 def _extract_annotations(self, video_id):
1599 return self._download_webpage(
1600 'https://www.youtube.com/annotations_invideo', video_id,
1601 note='Downloading annotations',
1602 errnote='Unable to download video annotations', fatal=False,
1603 query={
1604 'features': 1,
1605 'legacy': 1,
1606 'video_id': video_id,
1607 })
1608
1609 @staticmethod
1610 def _extract_chapters(description, duration):
1611 if not description:
1612 return None
1613 chapter_lines = re.findall(
1614 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1615 description)
1616 if not chapter_lines:
1617 return None
1618 chapters = []
1619 for next_num, (chapter_line, time_point) in enumerate(
1620 chapter_lines, start=1):
1621 start_time = parse_duration(time_point)
1622 if start_time is None:
1623 continue
1624 if start_time > duration:
1625 break
1626 end_time = (duration if next_num == len(chapter_lines)
1627 else parse_duration(chapter_lines[next_num][1]))
1628 if end_time is None:
1629 continue
1630 if end_time > duration:
1631 end_time = duration
1632 if start_time > end_time:
1633 break
1634 chapter_title = re.sub(
1635 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1636 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1637 chapters.append({
1638 'start_time': start_time,
1639 'end_time': end_time,
1640 'title': chapter_title,
1641 })
1642 return chapters
1643
1644 def _real_extract(self, url):
1645 url, smuggled_data = unsmuggle_url(url, {})
1646
1647 proto = (
1648 'http' if self._downloader.params.get('prefer_insecure', False)
1649 else 'https')
1650
1651 start_time = None
1652 end_time = None
1653 parsed_url = compat_urllib_parse_urlparse(url)
1654 for component in [parsed_url.fragment, parsed_url.query]:
1655 query = compat_parse_qs(component)
1656 if start_time is None and 't' in query:
1657 start_time = parse_duration(query['t'][0])
1658 if start_time is None and 'start' in query:
1659 start_time = parse_duration(query['start'][0])
1660 if end_time is None and 'end' in query:
1661 end_time = parse_duration(query['end'][0])
1662
1663 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1664 mobj = re.search(self._NEXT_URL_RE, url)
1665 if mobj:
1666 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1667 video_id = self.extract_id(url)
1668
1669 # Get video webpage
1670 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1671 video_webpage = self._download_webpage(url, video_id)
1672
1673 # Attempt to extract SWF player URL
1674 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1675 if mobj is not None:
1676 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1677 else:
1678 player_url = None
1679
1680 dash_mpds = []
1681
1682 def add_dash_mpd(video_info):
1683 dash_mpd = video_info.get('dashmpd')
1684 if dash_mpd and dash_mpd[0] not in dash_mpds:
1685 dash_mpds.append(dash_mpd[0])
1686
1687 def add_dash_mpd_pr(pl_response):
1688 dash_mpd = url_or_none(try_get(
1689 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1690 compat_str))
1691 if dash_mpd and dash_mpd not in dash_mpds:
1692 dash_mpds.append(dash_mpd)
1693
1694 is_live = None
1695 view_count = None
1696
1697 def extract_view_count(v_info):
1698 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1699
1700 def extract_token(v_info):
1701 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1702
1703 player_response = {}
1704
1705 # Get video info
1706 embed_webpage = None
1707 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1708 age_gate = True
1709 # We simulate the access to the video from www.youtube.com/v/{video_id}
1710 # this can be viewed without login into Youtube
1711 url = proto + '://www.youtube.com/embed/%s' % video_id
1712 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1713 data = compat_urllib_parse_urlencode({
1714 'video_id': video_id,
1715 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1716 'sts': self._search_regex(
1717 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1718 })
1719 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1720 video_info_webpage = self._download_webpage(
1721 video_info_url, video_id,
1722 note='Refetching age-gated info webpage',
1723 errnote='unable to download video info webpage')
1724 video_info = compat_parse_qs(video_info_webpage)
1725 add_dash_mpd(video_info)
1726 else:
1727 age_gate = False
1728 video_info = None
1729 sts = None
1730 # Try looking directly into the video webpage
1731 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1732 if ytplayer_config:
1733 args = ytplayer_config['args']
1734 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1735 # Convert to the same format returned by compat_parse_qs
1736 video_info = dict((k, [v]) for k, v in args.items())
1737 add_dash_mpd(video_info)
1738 # Rental video is not rented but preview is available (e.g.
1739 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1740 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1741 if not video_info and args.get('ypc_vid'):
1742 return self.url_result(
1743 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1744 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1745 is_live = True
1746 sts = ytplayer_config.get('sts')
1747 if not player_response:
1748 pl_response = str_or_none(args.get('player_response'))
1749 if pl_response:
1750 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1751 if isinstance(pl_response, dict):
1752 player_response = pl_response
1753 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1754 add_dash_mpd_pr(player_response)
1755 # We also try looking in get_video_info since it may contain different dashmpd
1756 # URL that points to a DASH manifest with possibly different itag set (some itags
1757 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1758 # manifest pointed by get_video_info's dashmpd).
1759 # The general idea is to take a union of itags of both DASH manifests (for example
1760 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1761 self.report_video_info_webpage_download(video_id)
1762 for el in ('embedded', 'detailpage', 'vevo', ''):
1763 query = {
1764 'video_id': video_id,
1765 'ps': 'default',
1766 'eurl': '',
1767 'gl': 'US',
1768 'hl': 'en',
1769 }
1770 if el:
1771 query['el'] = el
1772 if sts:
1773 query['sts'] = sts
1774 video_info_webpage = self._download_webpage(
1775 '%s://www.youtube.com/get_video_info' % proto,
1776 video_id, note=False,
1777 errnote='unable to download video info webpage',
1778 fatal=False, query=query)
1779 if not video_info_webpage:
1780 continue
1781 get_video_info = compat_parse_qs(video_info_webpage)
1782 if not player_response:
1783 pl_response = get_video_info.get('player_response', [None])[0]
1784 if isinstance(pl_response, dict):
1785 player_response = pl_response
1786 add_dash_mpd_pr(player_response)
1787 add_dash_mpd(get_video_info)
1788 if view_count is None:
1789 view_count = extract_view_count(get_video_info)
1790 if not video_info:
1791 video_info = get_video_info
1792 get_token = extract_token(get_video_info)
1793 if get_token:
1794 # Different get_video_info requests may report different results, e.g.
1795 # some may report video unavailability, but some may serve it without
1796 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1797 # the original webpage as well as el=info and el=embedded get_video_info
1798 # requests report video unavailability due to geo restriction while
1799 # el=detailpage succeeds and returns valid data). This is probably
1800 # due to YouTube measures against IP ranges of hosting providers.
1801 # Working around by preferring the first succeeded video_info containing
1802 # the token if no such video_info yet was found.
1803 token = extract_token(video_info)
1804 if not token:
1805 video_info = get_video_info
1806 break
1807
1808 def extract_unavailable_message():
1809 return self._html_search_regex(
1810 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1811 video_webpage, 'unavailable message', default=None)
1812
1813 if not video_info:
1814 unavailable_message = extract_unavailable_message()
1815 if not unavailable_message:
1816 unavailable_message = 'Unable to extract video data'
1817 raise ExtractorError(
1818 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1819
1820 video_details = try_get(
1821 player_response, lambda x: x['videoDetails'], dict) or {}
1822
1823 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1824 if not video_title:
1825 self._downloader.report_warning('Unable to extract video title')
1826 video_title = '_'
1827
1828 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1829 if video_description:
1830
1831 def replace_url(m):
1832 redir_url = compat_urlparse.urljoin(url, m.group(1))
1833 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1834 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1835 qs = compat_parse_qs(parsed_redir_url.query)
1836 q = qs.get('q')
1837 if q and q[0]:
1838 return q[0]
1839 return redir_url
1840
1841 description_original = video_description = re.sub(r'''(?x)
1842 <a\s+
1843 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1844 (?:title|href)="([^"]+)"\s+
1845 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1846 class="[^"]*"[^>]*>
1847 [^<]+\.{3}\s*
1848 </a>
1849 ''', replace_url, video_description)
1850 video_description = clean_html(video_description)
1851 else:
1852 video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
1853
1854 if not smuggled_data.get('force_singlefeed', False):
1855 if not self._downloader.params.get('noplaylist'):
1856 multifeed_metadata_list = try_get(
1857 player_response,
1858 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1859 compat_str) or try_get(
1860 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1861 if multifeed_metadata_list:
1862 entries = []
1863 feed_ids = []
1864 for feed in multifeed_metadata_list.split(','):
1865 # Unquote should take place before split on comma (,) since textual
1866 # fields may contain comma as well (see
1867 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1868 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1869 entries.append({
1870 '_type': 'url_transparent',
1871 'ie_key': 'Youtube',
1872 'url': smuggle_url(
1873 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1874 {'force_singlefeed': True}),
1875 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1876 })
1877 feed_ids.append(feed_data['id'][0])
1878 self.to_screen(
1879 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1880 % (', '.join(feed_ids), video_id))
1881 return self.playlist_result(entries, video_id, video_title, video_description)
1882 else:
1883 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1884
1885 if view_count is None:
1886 view_count = extract_view_count(video_info)
1887 if view_count is None and video_details:
1888 view_count = int_or_none(video_details.get('viewCount'))
1889
1890 if is_live is None:
1891 is_live = bool_or_none(video_details.get('isLive'))
1892
1893 # Check for "rental" videos
1894 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1895 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1896
1897 def _extract_filesize(media_url):
1898 return int_or_none(self._search_regex(
1899 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1900
1901 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1902 self.report_rtmp_download()
1903 formats = [{
1904 'format_id': '_rtmp',
1905 'protocol': 'rtmp',
1906 'url': video_info['conn'][0],
1907 'player_url': player_url,
1908 }]
1909 elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1910 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1911 if 'rtmpe%3Dyes' in encoded_url_map:
1912 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1913 formats_spec = {}
1914 fmt_list = video_info.get('fmt_list', [''])[0]
1915 if fmt_list:
1916 for fmt in fmt_list.split(','):
1917 spec = fmt.split('/')
1918 if len(spec) > 1:
1919 width_height = spec[1].split('x')
1920 if len(width_height) == 2:
1921 formats_spec[spec[0]] = {
1922 'resolution': spec[1],
1923 'width': int_or_none(width_height[0]),
1924 'height': int_or_none(width_height[1]),
1925 }
1926 q = qualities(['small', 'medium', 'hd720'])
1927 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1928 if streaming_formats:
1929 for fmt in streaming_formats:
1930 itag = str_or_none(fmt.get('itag'))
1931 if not itag:
1932 continue
1933 quality = fmt.get('quality')
1934 quality_label = fmt.get('qualityLabel') or quality
1935 formats_spec[itag] = {
1936 'asr': int_or_none(fmt.get('audioSampleRate')),
1937 'filesize': int_or_none(fmt.get('contentLength')),
1938 'format_note': quality_label,
1939 'fps': int_or_none(fmt.get('fps')),
1940 'height': int_or_none(fmt.get('height')),
1941 'quality': q(quality),
1942 # bitrate for itag 43 is always 2147483647
1943 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1944 'width': int_or_none(fmt.get('width')),
1945 }
1946 formats = []
1947 for url_data_str in encoded_url_map.split(','):
1948 url_data = compat_parse_qs(url_data_str)
1949 if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
1950 continue
1951 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1952 # Unsupported FORMAT_STREAM_TYPE_OTF
1953 if stream_type == 3:
1954 continue
1955 format_id = url_data['itag'][0]
1956 url = url_data['url'][0]
1957
1958 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1959 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1960 jsplayer_url_json = self._search_regex(
1961 ASSETS_RE,
1962 embed_webpage if age_gate else video_webpage,
1963 'JS player URL (1)', default=None)
1964 if not jsplayer_url_json and not age_gate:
1965 # We need the embed website after all
1966 if embed_webpage is None:
1967 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1968 embed_webpage = self._download_webpage(
1969 embed_url, video_id, 'Downloading embed webpage')
1970 jsplayer_url_json = self._search_regex(
1971 ASSETS_RE, embed_webpage, 'JS player URL')
1972
1973 player_url = json.loads(jsplayer_url_json)
1974 if player_url is None:
1975 player_url_json = self._search_regex(
1976 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1977 video_webpage, 'age gate player URL')
1978 player_url = json.loads(player_url_json)
1979
1980 if 'sig' in url_data:
1981 url += '&signature=' + url_data['sig'][0]
1982 elif 's' in url_data:
1983 encrypted_sig = url_data['s'][0]
1984
1985 if self._downloader.params.get('verbose'):
1986 if player_url is None:
1987 player_version = 'unknown'
1988 player_desc = 'unknown'
1989 else:
1990 if player_url.endswith('swf'):
1991 player_version = self._search_regex(
1992 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1993 'flash player', fatal=False)
1994 player_desc = 'flash player %s' % player_version
1995 else:
1996 player_version = self._search_regex(
1997 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1998 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1999 player_url,
2000 'html5 player', fatal=False)
2001 player_desc = 'html5 player %s' % player_version
2002
2003 parts_sizes = self._signature_cache_id(encrypted_sig)
2004 self.to_screen('{%s} signature length %s, %s' %
2005 (format_id, parts_sizes, player_desc))
2006
2007 signature = self._decrypt_signature(
2008 encrypted_sig, video_id, player_url, age_gate)
2009 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2010 url += '&%s=%s' % (sp, signature)
2011 if 'ratebypass' not in url:
2012 url += '&ratebypass=yes'
2013
2014 dct = {
2015 'format_id': format_id,
2016 'url': url,
2017 'player_url': player_url,
2018 }
2019 if format_id in self._formats:
2020 dct.update(self._formats[format_id])
2021 if format_id in formats_spec:
2022 dct.update(formats_spec[format_id])
2023
2024 # Some itags are not included in DASH manifest thus corresponding formats will
2025 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2026 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2027 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2028 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2029
2030 filesize = int_or_none(url_data.get(
2031 'clen', [None])[0]) or _extract_filesize(url)
2032
2033 quality = url_data.get('quality', [None])[0]
2034
2035 more_fields = {
2036 'filesize': filesize,
2037 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2038 'width': width,
2039 'height': height,
2040 'fps': int_or_none(url_data.get('fps', [None])[0]),
2041 'format_note': url_data.get('quality_label', [None])[0] or quality,
2042 'quality': q(quality),
2043 }
2044 for key, value in more_fields.items():
2045 if value:
2046 dct[key] = value
2047 type_ = url_data.get('type', [None])[0]
2048 if type_:
2049 type_split = type_.split(';')
2050 kind_ext = type_split[0].split('/')
2051 if len(kind_ext) == 2:
2052 kind, _ = kind_ext
2053 dct['ext'] = mimetype2ext(type_split[0])
2054 if kind in ('audio', 'video'):
2055 codecs = None
2056 for mobj in re.finditer(
2057 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2058 if mobj.group('key') == 'codecs':
2059 codecs = mobj.group('val')
2060 break
2061 if codecs:
2062 dct.update(parse_codecs(codecs))
2063 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2064 dct['downloader_options'] = {
2065 # Youtube throttles chunks >~10M
2066 'http_chunk_size': 10485760,
2067 }
2068 formats.append(dct)
2069 else:
2070 manifest_url = (
2071 url_or_none(try_get(
2072 player_response,
2073 lambda x: x['streamingData']['hlsManifestUrl'],
2074 compat_str))
2075 or url_or_none(try_get(
2076 video_info, lambda x: x['hlsvp'][0], compat_str)))
2077 if manifest_url:
2078 formats = []
2079 m3u8_formats = self._extract_m3u8_formats(
2080 manifest_url, video_id, 'mp4', fatal=False)
2081 for a_format in m3u8_formats:
2082 itag = self._search_regex(
2083 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2084 if itag:
2085 a_format['format_id'] = itag
2086 if itag in self._formats:
2087 dct = self._formats[itag].copy()
2088 dct.update(a_format)
2089 a_format = dct
2090 a_format['player_url'] = player_url
2091 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2092 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2093 formats.append(a_format)
2094 else:
2095 error_message = clean_html(video_info.get('reason', [None])[0])
2096 if not error_message:
2097 error_message = extract_unavailable_message()
2098 if error_message:
2099 raise ExtractorError(error_message, expected=True)
2100 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2101
2102 # uploader
2103 video_uploader = try_get(
2104 video_info, lambda x: x['author'][0],
2105 compat_str) or str_or_none(video_details.get('author'))
2106 if video_uploader:
2107 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2108 else:
2109 self._downloader.report_warning('unable to extract uploader name')
2110
2111 # uploader_id
2112 video_uploader_id = None
2113 video_uploader_url = None
2114 mobj = re.search(
2115 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2116 video_webpage)
2117 if mobj is not None:
2118 video_uploader_id = mobj.group('uploader_id')
2119 video_uploader_url = mobj.group('uploader_url')
2120 else:
2121 self._downloader.report_warning('unable to extract uploader nickname')
2122
2123 channel_id = (
2124 str_or_none(video_details.get('channelId'))
2125 or self._html_search_meta(
2126 'channelId', video_webpage, 'channel id', default=None)
2127 or self._search_regex(
2128 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2129 video_webpage, 'channel id', default=None, group='id'))
2130 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2131
2132 # thumbnail image
2133 # We try first to get a high quality image:
2134 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2135 video_webpage, re.DOTALL)
2136 if m_thumb is not None:
2137 video_thumbnail = m_thumb.group(1)
2138 elif 'thumbnail_url' not in video_info:
2139 self._downloader.report_warning('unable to extract video thumbnail')
2140 video_thumbnail = None
2141 else: # don't panic if we can't find it
2142 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2143
2144 # upload date
2145 upload_date = self._html_search_meta(
2146 'datePublished', video_webpage, 'upload date', default=None)
2147 if not upload_date:
2148 upload_date = self._search_regex(
2149 [r'(?s)id="eow-date.*?>(.*?)</span>',
2150 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2151 video_webpage, 'upload date', default=None)
2152 upload_date = unified_strdate(upload_date)
2153
2154 video_license = self._html_search_regex(
2155 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2156 video_webpage, 'license', default=None)
2157
2158 m_music = re.search(
2159 r'''(?x)
2160 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2161 <ul[^>]*>\s*
2162 <li>(?P<title>.+?)
2163 by (?P<creator>.+?)
2164 (?:
2165 \(.+?\)|
2166 <a[^>]*
2167 (?:
2168 \bhref=["\']/red[^>]*>| # drop possible
2169 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2170 )
2171 .*?
2172 )?</li
2173 ''',
2174 video_webpage)
2175 if m_music:
2176 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2177 video_creator = clean_html(m_music.group('creator'))
2178 else:
2179 video_alt_title = video_creator = None
2180
2181 def extract_meta(field):
2182 return self._html_search_regex(
2183 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2184 video_webpage, field, default=None)
2185
2186 track = extract_meta('Song')
2187 artist = extract_meta('Artist')
2188 album = extract_meta('Album')
2189
2190 # Youtube Music Auto-generated description
2191 release_date = release_year = None
2192 if video_description:
2193 mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^·]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?', video_description)
2194 if mobj:
2195 if not track:
2196 track = mobj.group('track').strip()
2197 if not artist:
2198 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2199 if not album:
2200 album = mobj.group('album'.strip())
2201 release_year = mobj.group('release_year')
2202 release_date = mobj.group('release_date')
2203 if release_date:
2204 release_date = release_date.replace('-', '')
2205 if not release_year:
2206 release_year = int(release_date[:4])
2207 if release_year:
2208 release_year = int(release_year)
2209
2210 m_episode = re.search(
2211 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2212 video_webpage)
2213 if m_episode:
2214 series = unescapeHTML(m_episode.group('series'))
2215 season_number = int(m_episode.group('season'))
2216 episode_number = int(m_episode.group('episode'))
2217 else:
2218 series = season_number = episode_number = None
2219
2220 m_cat_container = self._search_regex(
2221 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2222 video_webpage, 'categories', default=None)
2223 if m_cat_container:
2224 category = self._html_search_regex(
2225 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2226 default=None)
2227 video_categories = None if category is None else [category]
2228 else:
2229 video_categories = None
2230
2231 video_tags = [
2232 unescapeHTML(m.group('content'))
2233 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2234
2235 def _extract_count(count_name):
2236 return str_to_int(self._search_regex(
2237 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2238 % re.escape(count_name),
2239 video_webpage, count_name, default=None))
2240
2241 like_count = _extract_count('like')
2242 dislike_count = _extract_count('dislike')
2243
2244 if view_count is None:
2245 view_count = str_to_int(self._search_regex(
2246 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2247 'view count', default=None))
2248
2249 average_rating = (
2250 float_or_none(video_details.get('averageRating'))
2251 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2252
2253 # subtitles
2254 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2255 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2256
2257 video_duration = try_get(
2258 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2259 if not video_duration:
2260 video_duration = int_or_none(video_details.get('lengthSeconds'))
2261 if not video_duration:
2262 video_duration = parse_duration(self._html_search_meta(
2263 'duration', video_webpage, 'video duration'))
2264
2265 # annotations
2266 video_annotations = None
2267 if self._downloader.params.get('writeannotations', False):
2268 video_annotations = self._extract_annotations(video_id)
2269
2270 chapters = self._extract_chapters(description_original, video_duration)
2271
2272 # Look for the DASH manifest
2273 if self._downloader.params.get('youtube_include_dash_manifest', True):
2274 dash_mpd_fatal = True
2275 for mpd_url in dash_mpds:
2276 dash_formats = {}
2277 try:
2278 def decrypt_sig(mobj):
2279 s = mobj.group(1)
2280 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2281 return '/signature/%s' % dec_s
2282
2283 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2284
2285 for df in self._extract_mpd_formats(
2286 mpd_url, video_id, fatal=dash_mpd_fatal,
2287 formats_dict=self._formats):
2288 if not df.get('filesize'):
2289 df['filesize'] = _extract_filesize(df['url'])
2290 # Do not overwrite DASH format found in some previous DASH manifest
2291 if df['format_id'] not in dash_formats:
2292 dash_formats[df['format_id']] = df
2293 # Additional DASH manifests may end up in HTTP Error 403 therefore
2294 # allow them to fail without bug report message if we already have
2295 # some DASH manifest succeeded. This is temporary workaround to reduce
2296 # burst of bug reports until we figure out the reason and whether it
2297 # can be fixed at all.
2298 dash_mpd_fatal = False
2299 except (ExtractorError, KeyError) as e:
2300 self.report_warning(
2301 'Skipping DASH manifest: %r' % e, video_id)
2302 if dash_formats:
2303 # Remove the formats we found through non-DASH, they
2304 # contain less info and it can be wrong, because we use
2305 # fixed values (for example the resolution). See
2306 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2307 # example.
2308 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2309 formats.extend(dash_formats.values())
2310
2311 # Check for malformed aspect ratio
2312 stretched_m = re.search(
2313 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2314 video_webpage)
2315 if stretched_m:
2316 w = float(stretched_m.group('w'))
2317 h = float(stretched_m.group('h'))
2318 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2319 # We will only process correct ratios.
2320 if w > 0 and h > 0:
2321 ratio = w / h
2322 for f in formats:
2323 if f.get('vcodec') != 'none':
2324 f['stretched_ratio'] = ratio
2325
2326 if not formats:
2327 token = extract_token(video_info)
2328 if not token:
2329 if 'reason' in video_info:
2330 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2331 regions_allowed = self._html_search_meta(
2332 'regionsAllowed', video_webpage, default=None)
2333 countries = regions_allowed.split(',') if regions_allowed else None
2334 self.raise_geo_restricted(
2335 msg=video_info['reason'][0], countries=countries)
2336 reason = video_info['reason'][0]
2337 if 'Invalid parameters' in reason:
2338 unavailable_message = extract_unavailable_message()
2339 if unavailable_message:
2340 reason = unavailable_message
2341 raise ExtractorError(
2342 'YouTube said: %s' % reason,
2343 expected=True, video_id=video_id)
2344 else:
2345 raise ExtractorError(
2346 '"token" parameter not in video info for unknown reason',
2347 video_id=video_id)
2348
2349 if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2350 raise ExtractorError('This video is DRM protected.', expected=True)
2351
2352 self._sort_formats(formats)
2353
2354 self.mark_watched(video_id, video_info, player_response)
2355
2356 return {
2357 'id': video_id,
2358 'uploader': video_uploader,
2359 'uploader_id': video_uploader_id,
2360 'uploader_url': video_uploader_url,
2361 'channel_id': channel_id,
2362 'channel_url': channel_url,
2363 'upload_date': upload_date,
2364 'license': video_license,
2365 'creator': video_creator or artist,
2366 'title': video_title,
2367 'alt_title': video_alt_title or track,
2368 'thumbnail': video_thumbnail,
2369 'description': video_description,
2370 'categories': video_categories,
2371 'tags': video_tags,
2372 'subtitles': video_subtitles,
2373 'automatic_captions': automatic_captions,
2374 'duration': video_duration,
2375 'age_limit': 18 if age_gate else 0,
2376 'annotations': video_annotations,
2377 'chapters': chapters,
2378 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2379 'view_count': view_count,
2380 'like_count': like_count,
2381 'dislike_count': dislike_count,
2382 'average_rating': average_rating,
2383 'formats': formats,
2384 'is_live': is_live,
2385 'start_time': start_time,
2386 'end_time': end_time,
2387 'series': series,
2388 'season_number': season_number,
2389 'episode_number': episode_number,
2390 'track': track,
2391 'artist': artist,
2392 'album': album,
2393 'release_date': release_date,
2394 'release_year': release_year,
2395 }
2396
2397
2398 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2399 IE_DESC = 'YouTube.com playlists'
2400 _VALID_URL = r"""(?x)(?:
2401 (?:https?://)?
2402 (?:\w+\.)?
2403 (?:
2404 (?:
2405 youtube\.com|
2406 invidio\.us
2407 )
2408 /
2409 (?:
2410 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2411 \? (?:.*?[&;])*? (?:p|a|list)=
2412 | p/
2413 )|
2414 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2415 )
2416 (
2417 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2418 # Top tracks, they can also include dots
2419 |(?:MC)[\w\.]*
2420 )
2421 .*
2422 |
2423 (%(playlist_id)s)
2424 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2425 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2426 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2427 IE_NAME = 'youtube:playlist'
2428 _TESTS = [{
2429 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2430 'info_dict': {
2431 'title': 'ytdl test PL',
2432 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2433 },
2434 'playlist_count': 3,
2435 }, {
2436 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2437 'info_dict': {
2438 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2439 'title': 'YDL_Empty_List',
2440 },
2441 'playlist_count': 0,
2442 'skip': 'This playlist is private',
2443 }, {
2444 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2445 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2446 'info_dict': {
2447 'title': '29C3: Not my department',
2448 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2449 'uploader': 'Christiaan008',
2450 'uploader_id': 'ChRiStIaAn008',
2451 },
2452 'playlist_count': 95,
2453 }, {
2454 'note': 'issue #673',
2455 'url': 'PLBB231211A4F62143',
2456 'info_dict': {
2457 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2458 'id': 'PLBB231211A4F62143',
2459 'uploader': 'Wickydoo',
2460 'uploader_id': 'Wickydoo',
2461 },
2462 'playlist_mincount': 26,
2463 }, {
2464 'note': 'Large playlist',
2465 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2466 'info_dict': {
2467 'title': 'Uploads from Cauchemar',
2468 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2469 'uploader': 'Cauchemar',
2470 'uploader_id': 'Cauchemar89',
2471 },
2472 'playlist_mincount': 799,
2473 }, {
2474 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2475 'info_dict': {
2476 'title': 'YDL_safe_search',
2477 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2478 },
2479 'playlist_count': 2,
2480 'skip': 'This playlist is private',
2481 }, {
2482 'note': 'embedded',
2483 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2484 'playlist_count': 4,
2485 'info_dict': {
2486 'title': 'JODA15',
2487 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2488 'uploader': 'milan',
2489 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2490 }
2491 }, {
2492 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2493 'playlist_mincount': 485,
2494 'info_dict': {
2495 'title': '2018 Chinese New Singles (11/6 updated)',
2496 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2497 'uploader': 'LBK',
2498 'uploader_id': 'sdragonfang',
2499 }
2500 }, {
2501 'note': 'Embedded SWF player',
2502 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2503 'playlist_count': 4,
2504 'info_dict': {
2505 'title': 'JODA7',
2506 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2507 },
2508 'skip': 'This playlist does not exist',
2509 }, {
2510 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2511 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2512 'info_dict': {
2513 'title': 'Uploads from Interstellar Movie',
2514 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2515 'uploader': 'Interstellar Movie',
2516 'uploader_id': 'InterstellarMovie1',
2517 },
2518 'playlist_mincount': 21,
2519 }, {
2520 # Playlist URL that does not actually serve a playlist
2521 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2522 'info_dict': {
2523 'id': 'FqZTN594JQw',
2524 'ext': 'webm',
2525 'title': "Smiley's People 01 detective, Adventure Series, Action",
2526 'uploader': 'STREEM',
2527 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2528 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2529 'upload_date': '20150526',
2530 'license': 'Standard YouTube License',
2531 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2532 'categories': ['People & Blogs'],
2533 'tags': list,
2534 'view_count': int,
2535 'like_count': int,
2536 'dislike_count': int,
2537 },
2538 'params': {
2539 'skip_download': True,
2540 },
2541 'skip': 'This video is not available.',
2542 'add_ie': [YoutubeIE.ie_key()],
2543 }, {
2544 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2545 'info_dict': {
2546 'id': 'yeWKywCrFtk',
2547 'ext': 'mp4',
2548 'title': 'Small Scale Baler and Braiding Rugs',
2549 'uploader': 'Backus-Page House Museum',
2550 'uploader_id': 'backuspagemuseum',
2551 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2552 'upload_date': '20161008',
2553 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2554 'categories': ['Nonprofits & Activism'],
2555 'tags': list,
2556 'like_count': int,
2557 'dislike_count': int,
2558 },
2559 'params': {
2560 'noplaylist': True,
2561 'skip_download': True,
2562 },
2563 }, {
2564 # https://github.com/ytdl-org/youtube-dl/issues/21844
2565 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2566 'info_dict': {
2567 'title': 'Data Analysis with Dr Mike Pound',
2568 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2569 'uploader_id': 'Computerphile',
2570 'uploader': 'Computerphile',
2571 },
2572 'playlist_mincount': 11,
2573 }, {
2574 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2575 'only_matching': True,
2576 }, {
2577 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2578 'only_matching': True,
2579 }, {
2580 # music album playlist
2581 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2582 'only_matching': True,
2583 }, {
2584 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2585 'only_matching': True,
2586 }]
2587
2588 def _real_initialize(self):
2589 self._login()
2590
2591 def _extract_mix(self, playlist_id):
2592 # The mixes are generated from a single video
2593 # the id of the playlist is just 'RD' + video_id
2594 ids = []
2595 last_id = playlist_id[-11:]
2596 for n in itertools.count(1):
2597 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2598 webpage = self._download_webpage(
2599 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2600 new_ids = orderedSet(re.findall(
2601 r'''(?xs)data-video-username=".*?".*?
2602 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2603 webpage))
2604 # Fetch new pages until all the videos are repeated, it seems that
2605 # there are always 51 unique videos.
2606 new_ids = [_id for _id in new_ids if _id not in ids]
2607 if not new_ids:
2608 break
2609 ids.extend(new_ids)
2610 last_id = ids[-1]
2611
2612 url_results = self._ids_to_results(ids)
2613
2614 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2615 title_span = (
2616 search_title('playlist-title')
2617 or search_title('title long-title')
2618 or search_title('title'))
2619 title = clean_html(title_span)
2620
2621 return self.playlist_result(url_results, playlist_id, title)
2622
2623 def _extract_playlist(self, playlist_id):
2624 url = self._TEMPLATE_URL % playlist_id
2625 page = self._download_webpage(url, playlist_id)
2626
2627 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2628 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2629 match = match.strip()
2630 # Check if the playlist exists or is private
2631 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2632 if mobj:
2633 reason = mobj.group('reason')
2634 message = 'This playlist %s' % reason
2635 if 'private' in reason:
2636 message += ', use --username or --netrc to access it'
2637 message += '.'
2638 raise ExtractorError(message, expected=True)
2639 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2640 raise ExtractorError(
2641 'Invalid parameters. Maybe URL is incorrect.',
2642 expected=True)
2643 elif re.match(r'[^<]*Choose your language[^<]*', match):
2644 continue
2645 else:
2646 self.report_warning('Youtube gives an alert message: ' + match)
2647
2648 playlist_title = self._html_search_regex(
2649 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2650 page, 'title', default=None)
2651
2652 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2653 uploader = self._search_regex(
2654 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2655 page, 'uploader', default=None)
2656 mobj = re.search(
2657 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2658 page)
2659 if mobj:
2660 uploader_id = mobj.group('uploader_id')
2661 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2662 else:
2663 uploader_id = uploader_url = None
2664
2665 has_videos = True
2666
2667 if not playlist_title:
2668 try:
2669 # Some playlist URLs don't actually serve a playlist (e.g.
2670 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2671 next(self._entries(page, playlist_id))
2672 except StopIteration:
2673 has_videos = False
2674
2675 playlist = self.playlist_result(
2676 self._entries(page, playlist_id), playlist_id, playlist_title)
2677 playlist.update({
2678 'uploader': uploader,
2679 'uploader_id': uploader_id,
2680 'uploader_url': uploader_url,
2681 })
2682
2683 return has_videos, playlist
2684
2685 def _check_download_just_video(self, url, playlist_id):
2686 # Check if it's a video-specific URL
2687 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2688 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2689 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2690 'video id', default=None)
2691 if video_id:
2692 if self._downloader.params.get('noplaylist'):
2693 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2694 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2695 else:
2696 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2697 return video_id, None
2698 return None, None
2699
2700 def _real_extract(self, url):
2701 # Extract playlist id
2702 mobj = re.match(self._VALID_URL, url)
2703 if mobj is None:
2704 raise ExtractorError('Invalid URL: %s' % url)
2705 playlist_id = mobj.group(1) or mobj.group(2)
2706
2707 video_id, video = self._check_download_just_video(url, playlist_id)
2708 if video:
2709 return video
2710
2711 if playlist_id.startswith(('RD', 'UL', 'PU')):
2712 # Mixes require a custom extraction process
2713 return self._extract_mix(playlist_id)
2714
2715 has_videos, playlist = self._extract_playlist(playlist_id)
2716 if has_videos or not video_id:
2717 return playlist
2718
2719 # Some playlist URLs don't actually serve a playlist (see
2720 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2721 # Fallback to plain video extraction if there is a video id
2722 # along with playlist id.
2723 return self.url_result(video_id, 'Youtube', video_id=video_id)
2724
2725
2726 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2727 IE_DESC = 'YouTube.com channels'
2728 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2729 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2730 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2731 IE_NAME = 'youtube:channel'
2732 _TESTS = [{
2733 'note': 'paginated channel',
2734 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2735 'playlist_mincount': 91,
2736 'info_dict': {
2737 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2738 'title': 'Uploads from lex will',
2739 'uploader': 'lex will',
2740 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2741 }
2742 }, {
2743 'note': 'Age restricted channel',
2744 # from https://www.youtube.com/user/DeusExOfficial
2745 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2746 'playlist_mincount': 64,
2747 'info_dict': {
2748 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2749 'title': 'Uploads from Deus Ex',
2750 'uploader': 'Deus Ex',
2751 'uploader_id': 'DeusExOfficial',
2752 },
2753 }, {
2754 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2755 'only_matching': True,
2756 }]
2757
2758 @classmethod
2759 def suitable(cls, url):
2760 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2761 else super(YoutubeChannelIE, cls).suitable(url))
2762
2763 def _build_template_url(self, url, channel_id):
2764 return self._TEMPLATE_URL % channel_id
2765
2766 def _real_extract(self, url):
2767 channel_id = self._match_id(url)
2768
2769 url = self._build_template_url(url, channel_id)
2770
2771 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2772 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2773 # otherwise fallback on channel by page extraction
2774 channel_page = self._download_webpage(
2775 url + '?view=57', channel_id,
2776 'Downloading channel page', fatal=False)
2777 if channel_page is False:
2778 channel_playlist_id = False
2779 else:
2780 channel_playlist_id = self._html_search_meta(
2781 'channelId', channel_page, 'channel id', default=None)
2782 if not channel_playlist_id:
2783 channel_url = self._html_search_meta(
2784 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2785 channel_page, 'channel url', default=None)
2786 if channel_url:
2787 channel_playlist_id = self._search_regex(
2788 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2789 channel_url, 'channel id', default=None)
2790 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2791 playlist_id = 'UU' + channel_playlist_id[2:]
2792 return self.url_result(
2793 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2794
2795 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2796 autogenerated = re.search(r'''(?x)
2797 class="[^"]*?(?:
2798 channel-header-autogenerated-label|
2799 yt-channel-title-autogenerated
2800 )[^"]*"''', channel_page) is not None
2801
2802 if autogenerated:
2803 # The videos are contained in a single page
2804 # the ajax pages can't be used, they are empty
2805 entries = [
2806 self.url_result(
2807 video_id, 'Youtube', video_id=video_id,
2808 video_title=video_title)
2809 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2810 return self.playlist_result(entries, channel_id)
2811
2812 try:
2813 next(self._entries(channel_page, channel_id))
2814 except StopIteration:
2815 alert_message = self._html_search_regex(
2816 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2817 channel_page, 'alert', default=None, group='alert')
2818 if alert_message:
2819 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2820
2821 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2822
2823
2824 class YoutubeUserIE(YoutubeChannelIE):
2825 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2826 _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_-]+)'
2827 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2828 IE_NAME = 'youtube:user'
2829
2830 _TESTS = [{
2831 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2832 'playlist_mincount': 320,
2833 'info_dict': {
2834 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2835 'title': 'Uploads from The Linux Foundation',
2836 'uploader': 'The Linux Foundation',
2837 'uploader_id': 'TheLinuxFoundation',
2838 }
2839 }, {
2840 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2841 # but not https://www.youtube.com/user/12minuteathlete/videos
2842 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2843 'playlist_mincount': 249,
2844 'info_dict': {
2845 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2846 'title': 'Uploads from 12 Minute Athlete',
2847 'uploader': '12 Minute Athlete',
2848 'uploader_id': 'the12minuteathlete',
2849 }
2850 }, {
2851 'url': 'ytuser:phihag',
2852 'only_matching': True,
2853 }, {
2854 'url': 'https://www.youtube.com/c/gametrailers',
2855 'only_matching': True,
2856 }, {
2857 'url': 'https://www.youtube.com/gametrailers',
2858 'only_matching': True,
2859 }, {
2860 # This channel is not available, geo restricted to JP
2861 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2862 'only_matching': True,
2863 }]
2864
2865 @classmethod
2866 def suitable(cls, url):
2867 # Don't return True if the url can be extracted with other youtube
2868 # extractor, the regex would is too permissive and it would match.
2869 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2870 if any(ie.suitable(url) for ie in other_yt_ies):
2871 return False
2872 else:
2873 return super(YoutubeUserIE, cls).suitable(url)
2874
2875 def _build_template_url(self, url, channel_id):
2876 mobj = re.match(self._VALID_URL, url)
2877 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2878
2879
2880 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2881 IE_DESC = 'YouTube.com live streams'
2882 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2883 IE_NAME = 'youtube:live'
2884
2885 _TESTS = [{
2886 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2887 'info_dict': {
2888 'id': 'a48o2S1cPoo',
2889 'ext': 'mp4',
2890 'title': 'The Young Turks - Live Main Show',
2891 'uploader': 'The Young Turks',
2892 'uploader_id': 'TheYoungTurks',
2893 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2894 'upload_date': '20150715',
2895 'license': 'Standard YouTube License',
2896 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2897 'categories': ['News & Politics'],
2898 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2899 'like_count': int,
2900 'dislike_count': int,
2901 },
2902 'params': {
2903 'skip_download': True,
2904 },
2905 }, {
2906 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2907 'only_matching': True,
2908 }, {
2909 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2910 'only_matching': True,
2911 }, {
2912 'url': 'https://www.youtube.com/TheYoungTurks/live',
2913 'only_matching': True,
2914 }]
2915
2916 def _real_extract(self, url):
2917 mobj = re.match(self._VALID_URL, url)
2918 channel_id = mobj.group('id')
2919 base_url = mobj.group('base_url')
2920 webpage = self._download_webpage(url, channel_id, fatal=False)
2921 if webpage:
2922 page_type = self._og_search_property(
2923 'type', webpage, 'page type', default='')
2924 video_id = self._html_search_meta(
2925 'videoId', webpage, 'video id', default=None)
2926 if page_type.startswith('video') and video_id and re.match(
2927 r'^[0-9A-Za-z_-]{11}$', video_id):
2928 return self.url_result(video_id, YoutubeIE.ie_key())
2929 return self.url_result(base_url)
2930
2931
2932 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2933 IE_DESC = 'YouTube.com user/channel playlists'
2934 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2935 IE_NAME = 'youtube:playlists'
2936
2937 _TESTS = [{
2938 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2939 'playlist_mincount': 4,
2940 'info_dict': {
2941 'id': 'ThirstForScience',
2942 'title': 'ThirstForScience',
2943 },
2944 }, {
2945 # with "Load more" button
2946 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2947 'playlist_mincount': 70,
2948 'info_dict': {
2949 'id': 'igorkle1',
2950 'title': 'Игорь Клейнер',
2951 },
2952 }, {
2953 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2954 'playlist_mincount': 17,
2955 'info_dict': {
2956 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2957 'title': 'Chem Player',
2958 },
2959 'skip': 'Blocked',
2960 }]
2961
2962
2963 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2964 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2965
2966
2967 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2968 IE_DESC = 'YouTube.com searches'
2969 # there doesn't appear to be a real limit, for example if you search for
2970 # 'python' you get more than 8.000.000 results
2971 _MAX_RESULTS = float('inf')
2972 IE_NAME = 'youtube:search'
2973 _SEARCH_KEY = 'ytsearch'
2974 _EXTRA_QUERY_ARGS = {}
2975 _TESTS = []
2976
2977 def _get_n_results(self, query, n):
2978 """Get a specified number of results for a query"""
2979
2980 videos = []
2981 limit = n
2982
2983 url_query = {
2984 'search_query': query.encode('utf-8'),
2985 }
2986 url_query.update(self._EXTRA_QUERY_ARGS)
2987 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2988
2989 for pagenum in itertools.count(1):
2990 data = self._download_json(
2991 result_url, video_id='query "%s"' % query,
2992 note='Downloading page %s' % pagenum,
2993 errnote='Unable to download API page',
2994 query={'spf': 'navigate'})
2995 html_content = data[1]['body']['content']
2996
2997 if 'class="search-message' in html_content:
2998 raise ExtractorError(
2999 '[youtube] No video results', expected=True)
3000
3001 new_videos = list(self._process_page(html_content))
3002 videos += new_videos
3003 if not new_videos or len(videos) > limit:
3004 break
3005 next_link = self._html_search_regex(
3006 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3007 html_content, 'next link', default=None)
3008 if next_link is None:
3009 break
3010 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3011
3012 if len(videos) > n:
3013 videos = videos[:n]
3014 return self.playlist_result(videos, query)
3015
3016
3017 class YoutubeSearchDateIE(YoutubeSearchIE):
3018 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3019 _SEARCH_KEY = 'ytsearchdate'
3020 IE_DESC = 'YouTube.com searches, newest videos first'
3021 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3022
3023
3024 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3025 IE_DESC = 'YouTube.com search URLs'
3026 IE_NAME = 'youtube:search_url'
3027 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3028 _TESTS = [{
3029 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3030 'playlist_mincount': 5,
3031 'info_dict': {
3032 'title': 'youtube-dl test video',
3033 }
3034 }, {
3035 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3036 'only_matching': True,
3037 }]
3038
3039 def _real_extract(self, url):
3040 mobj = re.match(self._VALID_URL, url)
3041 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3042 webpage = self._download_webpage(url, query)
3043 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3044
3045
3046 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3047 IE_DESC = 'YouTube.com (multi-season) shows'
3048 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3049 IE_NAME = 'youtube:show'
3050 _TESTS = [{
3051 'url': 'https://www.youtube.com/show/airdisasters',
3052 'playlist_mincount': 5,
3053 'info_dict': {
3054 'id': 'airdisasters',
3055 'title': 'Air Disasters',
3056 }
3057 }]
3058
3059 def _real_extract(self, url):
3060 playlist_id = self._match_id(url)
3061 return super(YoutubeShowIE, self)._real_extract(
3062 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3063
3064
3065 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3066 """
3067 Base class for feed extractors
3068 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3069 """
3070 _LOGIN_REQUIRED = True
3071
3072 @property
3073 def IE_NAME(self):
3074 return 'youtube:%s' % self._FEED_NAME
3075
3076 def _real_initialize(self):
3077 self._login()
3078
3079 def _entries(self, page):
3080 # The extraction process is the same as for playlists, but the regex
3081 # for the video ids doesn't contain an index
3082 ids = []
3083 more_widget_html = content_html = page
3084 for page_num in itertools.count(1):
3085 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3086
3087 # 'recommended' feed has infinite 'load more' and each new portion spins
3088 # the same videos in (sometimes) slightly different order, so we'll check
3089 # for unicity and break when portion has no new videos
3090 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3091 if not new_ids:
3092 break
3093
3094 ids.extend(new_ids)
3095
3096 for entry in self._ids_to_results(new_ids):
3097 yield entry
3098
3099 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3100 if not mobj:
3101 break
3102
3103 more = self._download_json(
3104 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3105 'Downloading page #%s' % page_num,
3106 transform_source=uppercase_escape)
3107 content_html = more['content_html']
3108 more_widget_html = more['load_more_widget_html']
3109
3110 def _real_extract(self, url):
3111 page = self._download_webpage(
3112 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3113 self._PLAYLIST_TITLE)
3114 return self.playlist_result(
3115 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3116
3117
3118 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3119 IE_NAME = 'youtube:watchlater'
3120 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3121 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3122
3123 _TESTS = [{
3124 'url': 'https://www.youtube.com/playlist?list=WL',
3125 'only_matching': True,
3126 }, {
3127 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3128 'only_matching': True,
3129 }]
3130
3131 def _real_extract(self, url):
3132 _, video = self._check_download_just_video(url, 'WL')
3133 if video:
3134 return video
3135 _, playlist = self._extract_playlist('WL')
3136 return playlist
3137
3138
3139 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3140 IE_NAME = 'youtube:favorites'
3141 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3142 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3143 _LOGIN_REQUIRED = True
3144
3145 def _real_extract(self, url):
3146 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3147 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3148 return self.url_result(playlist_id, 'YoutubePlaylist')
3149
3150
3151 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3152 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3153 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3154 _FEED_NAME = 'recommended'
3155 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3156
3157
3158 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3159 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3160 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3161 _FEED_NAME = 'subscriptions'
3162 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3163
3164
3165 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3166 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3167 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3168 _FEED_NAME = 'history'
3169 _PLAYLIST_TITLE = 'Youtube History'
3170
3171
3172 class YoutubeTruncatedURLIE(InfoExtractor):
3173 IE_NAME = 'youtube:truncated_url'
3174 IE_DESC = False # Do not list
3175 _VALID_URL = r'''(?x)
3176 (?:https?://)?
3177 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3178 (?:watch\?(?:
3179 feature=[a-z_]+|
3180 annotation_id=annotation_[^&]+|
3181 x-yt-cl=[0-9]+|
3182 hl=[^&]*|
3183 t=[0-9]+
3184 )?
3185 |
3186 attribution_link\?a=[^&]+
3187 )
3188 $
3189 '''
3190
3191 _TESTS = [{
3192 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3193 'only_matching': True,
3194 }, {
3195 'url': 'https://www.youtube.com/watch?',
3196 'only_matching': True,
3197 }, {
3198 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3199 'only_matching': True,
3200 }, {
3201 'url': 'https://www.youtube.com/watch?feature=foo',
3202 'only_matching': True,
3203 }, {
3204 'url': 'https://www.youtube.com/watch?hl=en-GB',
3205 'only_matching': True,
3206 }, {
3207 'url': 'https://www.youtube.com/watch?t=2372',
3208 'only_matching': True,
3209 }]
3210
3211 def _real_extract(self, url):
3212 raise ExtractorError(
3213 'Did you forget to quote the URL? Remember that & is a meta '
3214 'character in most shells, so you want to put the URL in quotes, '
3215 'like youtube-dl '
3216 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3217 ' or simply youtube-dl BaW_jenozKc .',
3218 expected=True)
3219
3220
3221 class YoutubeTruncatedIDIE(InfoExtractor):
3222 IE_NAME = 'youtube:truncated_id'
3223 IE_DESC = False # Do not list
3224 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3225
3226 _TESTS = [{
3227 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3228 'only_matching': True,
3229 }]
3230
3231 def _real_extract(self, url):
3232 video_id = self._match_id(url)
3233 raise ExtractorError(
3234 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3235 expected=True)