]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
b570d5bae9b6128afb544e16fc9aef292e64fa01
[youtubedl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_HTTPError,
20 compat_kwargs,
21 compat_parse_qs,
22 compat_urllib_parse_unquote,
23 compat_urllib_parse_unquote_plus,
24 compat_urllib_parse_urlencode,
25 compat_urllib_parse_urlparse,
26 compat_urlparse,
27 compat_str,
28 )
29 from ..utils import (
30 clean_html,
31 dict_get,
32 error_to_compat_str,
33 ExtractorError,
34 float_or_none,
35 get_element_by_attribute,
36 get_element_by_id,
37 int_or_none,
38 mimetype2ext,
39 orderedSet,
40 parse_codecs,
41 parse_duration,
42 qualities,
43 remove_quotes,
44 remove_start,
45 smuggle_url,
46 str_or_none,
47 str_to_int,
48 try_get,
49 unescapeHTML,
50 unified_strdate,
51 unsmuggle_url,
52 uppercase_escape,
53 url_or_none,
54 urlencode_postdata,
55 )
56
57
58 class YoutubeBaseInfoExtractor(InfoExtractor):
59 """Provide base functions for Youtube extractors"""
60 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
61 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
62
63 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
64 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
65 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
66
67 _NETRC_MACHINE = 'youtube'
68 # If True it will raise an error if no login info is provided
69 _LOGIN_REQUIRED = False
70
71 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
72
73 def _set_language(self):
74 self._set_cookie(
75 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
76 # YouTube sets the expire time to about two months
77 expire_time=time.time() + 2 * 30 * 24 * 3600)
78
79 def _ids_to_results(self, ids):
80 return [
81 self.url_result(vid_id, 'Youtube', video_id=vid_id)
82 for vid_id in ids]
83
84 def _login(self):
85 """
86 Attempt to log in to YouTube.
87 True is returned if successful or skipped.
88 False is returned if login failed.
89
90 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
91 """
92 username, password = self._get_login_info()
93 # No authentication to be performed
94 if username is None:
95 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
96 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
97 return True
98
99 login_page = self._download_webpage(
100 self._LOGIN_URL, None,
101 note='Downloading login page',
102 errnote='unable to fetch login page', fatal=False)
103 if login_page is False:
104 return
105
106 login_form = self._hidden_inputs(login_page)
107
108 def req(url, f_req, note, errnote):
109 data = login_form.copy()
110 data.update({
111 'pstMsg': 1,
112 'checkConnection': 'youtube',
113 'checkedDomains': 'youtube',
114 'hl': 'en',
115 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
116 'f.req': json.dumps(f_req),
117 'flowName': 'GlifWebSignIn',
118 'flowEntry': 'ServiceLogin',
119 })
120 return self._download_json(
121 url, None, note=note, errnote=errnote,
122 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
123 fatal=False,
124 data=urlencode_postdata(data), headers={
125 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
126 'Google-Accounts-XSRF': 1,
127 })
128
129 def warn(message):
130 self._downloader.report_warning(message)
131
132 lookup_req = [
133 username,
134 None, [], None, 'US', None, None, 2, False, True,
135 [
136 None, None,
137 [2, 1, None, 1,
138 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
139 None, [], 4],
140 1, [None, None, []], None, None, None, True
141 ],
142 username,
143 ]
144
145 lookup_results = req(
146 self._LOOKUP_URL, lookup_req,
147 'Looking up account info', 'Unable to look up account info')
148
149 if lookup_results is False:
150 return False
151
152 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
153 if not user_hash:
154 warn('Unable to extract user hash')
155 return False
156
157 challenge_req = [
158 user_hash,
159 None, 1, None, [1, None, None, None, [password, None, True]],
160 [
161 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
162 1, [None, None, []], None, None, None, True
163 ]]
164
165 challenge_results = req(
166 self._CHALLENGE_URL, challenge_req,
167 'Logging in', 'Unable to log in')
168
169 if challenge_results is False:
170 return
171
172 login_res = try_get(challenge_results, lambda x: x[0][5], list)
173 if login_res:
174 login_msg = try_get(login_res, lambda x: x[5], compat_str)
175 warn(
176 'Unable to login: %s' % 'Invalid password'
177 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
178 return False
179
180 res = try_get(challenge_results, lambda x: x[0][-1], list)
181 if not res:
182 warn('Unable to extract result entry')
183 return False
184
185 login_challenge = try_get(res, lambda x: x[0][0], list)
186 if login_challenge:
187 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
188 if challenge_str == 'TWO_STEP_VERIFICATION':
189 # SEND_SUCCESS - TFA code has been successfully sent to phone
190 # QUOTA_EXCEEDED - reached the limit of TFA codes
191 status = try_get(login_challenge, lambda x: x[5], compat_str)
192 if status == 'QUOTA_EXCEEDED':
193 warn('Exceeded the limit of TFA codes, try later')
194 return False
195
196 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
197 if not tl:
198 warn('Unable to extract TL')
199 return False
200
201 tfa_code = self._get_tfa_info('2-step verification code')
202
203 if not tfa_code:
204 warn(
205 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
206 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
207 return False
208
209 tfa_code = remove_start(tfa_code, 'G-')
210
211 tfa_req = [
212 user_hash, None, 2, None,
213 [
214 9, None, None, None, None, None, None, None,
215 [None, tfa_code, True, 2]
216 ]]
217
218 tfa_results = req(
219 self._TFA_URL.format(tl), tfa_req,
220 'Submitting TFA code', 'Unable to submit TFA code')
221
222 if tfa_results is False:
223 return False
224
225 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
226 if tfa_res:
227 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
228 warn(
229 'Unable to finish TFA: %s' % 'Invalid TFA code'
230 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
231 return False
232
233 check_cookie_url = try_get(
234 tfa_results, lambda x: x[0][-1][2], compat_str)
235 else:
236 CHALLENGES = {
237 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
238 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
239 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
240 }
241 challenge = CHALLENGES.get(
242 challenge_str,
243 '%s returned error %s.' % (self.IE_NAME, challenge_str))
244 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
245 return False
246 else:
247 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
248
249 if not check_cookie_url:
250 warn('Unable to extract CheckCookie URL')
251 return False
252
253 check_cookie_results = self._download_webpage(
254 check_cookie_url, None, 'Checking cookie', fatal=False)
255
256 if check_cookie_results is False:
257 return False
258
259 if 'https://myaccount.google.com/' not in check_cookie_results:
260 warn('Unable to log in')
261 return False
262
263 return True
264
265 def _download_webpage_handle(self, *args, **kwargs):
266 query = kwargs.get('query', {}).copy()
267 query['disable_polymer'] = 'true'
268 kwargs['query'] = query
269 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
270 *args, **compat_kwargs(kwargs))
271
272 def _real_initialize(self):
273 if self._downloader is None:
274 return
275 self._set_language()
276 if not self._login():
277 return
278
279
280 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
281 # Extract entries from page with "Load more" button
282 def _entries(self, page, playlist_id):
283 more_widget_html = content_html = page
284 for page_num in itertools.count(1):
285 for entry in self._process_page(content_html):
286 yield entry
287
288 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
289 if not mobj:
290 break
291
292 count = 0
293 retries = 3
294 while count <= retries:
295 try:
296 # Downloading page may result in intermittent 5xx HTTP error
297 # that is usually worked around with a retry
298 more = self._download_json(
299 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
300 'Downloading page #%s%s'
301 % (page_num, ' (retry #%d)' % count if count else ''),
302 transform_source=uppercase_escape)
303 break
304 except ExtractorError as e:
305 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
306 count += 1
307 if count <= retries:
308 continue
309 raise
310
311 content_html = more['content_html']
312 if not content_html.strip():
313 # Some webpages show a "Load more" button but they don't
314 # have more videos
315 break
316 more_widget_html = more['load_more_widget_html']
317
318
319 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
320 def _process_page(self, content):
321 for video_id, video_title in self.extract_videos_from_page(content):
322 yield self.url_result(video_id, 'Youtube', video_id, video_title)
323
324 def extract_videos_from_page(self, page):
325 ids_in_page = []
326 titles_in_page = []
327 for mobj in re.finditer(self._VIDEO_RE, page):
328 # The link with index 0 is not the first video of the playlist (not sure if still actual)
329 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
330 continue
331 video_id = mobj.group('id')
332 video_title = unescapeHTML(mobj.group('title'))
333 if video_title:
334 video_title = video_title.strip()
335 try:
336 idx = ids_in_page.index(video_id)
337 if video_title and not titles_in_page[idx]:
338 titles_in_page[idx] = video_title
339 except ValueError:
340 ids_in_page.append(video_id)
341 titles_in_page.append(video_title)
342 return zip(ids_in_page, titles_in_page)
343
344
345 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
346 def _process_page(self, content):
347 for playlist_id in orderedSet(re.findall(
348 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
349 content)):
350 yield self.url_result(
351 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
352
353 def _real_extract(self, url):
354 playlist_id = self._match_id(url)
355 webpage = self._download_webpage(url, playlist_id)
356 title = self._og_search_title(webpage, fatal=False)
357 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
358
359
360 class YoutubeIE(YoutubeBaseInfoExtractor):
361 IE_DESC = 'YouTube.com'
362 _VALID_URL = r"""(?x)^
363 (
364 (?:https?://|//) # http(s):// or protocol-independent URL
365 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
366 (?:www\.)?deturl\.com/www\.youtube\.com/|
367 (?:www\.)?pwnyoutube\.com/|
368 (?:www\.)?hooktube\.com/|
369 (?:www\.)?yourepeat\.com/|
370 tube\.majestyc\.net/|
371 (?:(?:www|dev)\.)?invidio\.us/|
372 (?:www\.)?invidiou\.sh/|
373 (?:www\.)?invidious\.snopyta\.org/|
374 (?:www\.)?invidious\.kabi\.tk/|
375 (?:www\.)?vid\.wxzm\.sx/|
376 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
377 (?:.*?\#/)? # handle anchor (#/) redirect urls
378 (?: # the various things that can precede the ID:
379 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
380 |(?: # or the v= param in all its forms
381 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
382 (?:\?|\#!?) # the params delimiter ? or # or #!
383 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
384 v=
385 )
386 ))
387 |(?:
388 youtu\.be| # just youtu.be/xxxx
389 vid\.plus| # or vid.plus/xxxx
390 zwearz\.com/watch| # or zwearz.com/watch/xxxx
391 )/
392 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
393 )
394 )? # all until now is optional -> you can pass the naked ID
395 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
396 (?!.*?\blist=
397 (?:
398 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
399 WL # WL are handled by the watch later IE
400 )
401 )
402 (?(1).+)? # if we found the ID, everything can follow
403 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
404 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
405 _formats = {
406 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
407 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
408 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
409 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
410 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
411 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
412 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
413 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
414 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
415 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
416 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
417 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
418 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
419 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
420 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
421 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
422 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
423 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
424
425
426 # 3D videos
427 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
428 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
429 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
430 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
431 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
432 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
433 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
434
435 # Apple HTTP Live Streaming
436 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
437 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
438 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
439 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
440 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
441 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
442 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
443 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
444
445 # DASH mp4 video
446 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
447 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
448 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
449 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
450 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
451 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
452 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
453 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
454 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
455 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
456 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
457 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
458
459 # Dash mp4 audio
460 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
461 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
462 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
463 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
464 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
465 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
466 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
467
468 # Dash webm
469 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
470 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
471 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
472 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
473 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
474 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
475 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
476 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
477 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
478 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
479 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
480 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
481 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
482 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
483 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
484 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
485 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
486 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
487 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
488 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
489 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
490 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
491
492 # Dash webm audio
493 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
494 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
495
496 # Dash webm audio with opus inside
497 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
498 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
499 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
500
501 # RTMP (unnamed)
502 '_rtmp': {'protocol': 'rtmp'},
503
504 # av01 video only formats sometimes served with "unknown" codecs
505 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
506 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
507 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
508 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
509 }
510 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
511
512 _GEO_BYPASS = False
513
514 IE_NAME = 'youtube'
515 _TESTS = [
516 {
517 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
518 'info_dict': {
519 'id': 'BaW_jenozKc',
520 'ext': 'mp4',
521 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
522 'uploader': 'Philipp Hagemeister',
523 'uploader_id': 'phihag',
524 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
525 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
526 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
527 'upload_date': '20121002',
528 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
529 'categories': ['Science & Technology'],
530 'tags': ['youtube-dl'],
531 'duration': 10,
532 'view_count': int,
533 'like_count': int,
534 'dislike_count': int,
535 'start_time': 1,
536 'end_time': 9,
537 }
538 },
539 {
540 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
541 'note': 'Test generic use_cipher_signature video (#897)',
542 'info_dict': {
543 'id': 'UxxajLWwzqY',
544 'ext': 'mp4',
545 'upload_date': '20120506',
546 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
547 'alt_title': 'I Love It (feat. Charli XCX)',
548 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
549 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
550 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
551 'iconic ep', 'iconic', 'love', 'it'],
552 'duration': 180,
553 'uploader': 'Icona Pop',
554 'uploader_id': 'IconaPop',
555 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
556 'creator': 'Icona Pop',
557 'track': 'I Love It (feat. Charli XCX)',
558 'artist': 'Icona Pop',
559 }
560 },
561 {
562 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
563 'note': 'Test VEVO video with age protection (#956)',
564 'info_dict': {
565 'id': '07FYdnEawAQ',
566 'ext': 'mp4',
567 'upload_date': '20130703',
568 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
569 'alt_title': 'Tunnel Vision',
570 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
571 'duration': 419,
572 'uploader': 'justintimberlakeVEVO',
573 'uploader_id': 'justintimberlakeVEVO',
574 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
575 'creator': 'Justin Timberlake',
576 'track': 'Tunnel Vision',
577 'artist': 'Justin Timberlake',
578 'age_limit': 18,
579 }
580 },
581 {
582 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
583 'note': 'Embed-only video (#1746)',
584 'info_dict': {
585 'id': 'yZIXLfi8CZQ',
586 'ext': 'mp4',
587 'upload_date': '20120608',
588 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
589 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
590 'uploader': 'SET India',
591 'uploader_id': 'setindia',
592 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
593 'age_limit': 18,
594 }
595 },
596 {
597 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
598 'note': 'Use the first video ID in the URL',
599 'info_dict': {
600 'id': 'BaW_jenozKc',
601 'ext': 'mp4',
602 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
603 'uploader': 'Philipp Hagemeister',
604 'uploader_id': 'phihag',
605 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
606 'upload_date': '20121002',
607 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
608 'categories': ['Science & Technology'],
609 'tags': ['youtube-dl'],
610 'duration': 10,
611 'view_count': int,
612 'like_count': int,
613 'dislike_count': int,
614 },
615 'params': {
616 'skip_download': True,
617 },
618 },
619 {
620 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
621 'note': '256k DASH audio (format 141) via DASH manifest',
622 'info_dict': {
623 'id': 'a9LDPn-MO4I',
624 'ext': 'm4a',
625 'upload_date': '20121002',
626 'uploader_id': '8KVIDEO',
627 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
628 'description': '',
629 'uploader': '8KVIDEO',
630 'title': 'UHDTV TEST 8K VIDEO.mp4'
631 },
632 'params': {
633 'youtube_include_dash_manifest': True,
634 'format': '141',
635 },
636 'skip': 'format 141 not served anymore',
637 },
638 # DASH manifest with encrypted signature
639 {
640 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
641 'info_dict': {
642 'id': 'IB3lcPjvWLA',
643 'ext': 'm4a',
644 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
645 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
646 'duration': 244,
647 'uploader': 'AfrojackVEVO',
648 'uploader_id': 'AfrojackVEVO',
649 'upload_date': '20131011',
650 },
651 'params': {
652 'youtube_include_dash_manifest': True,
653 'format': '141/bestaudio[ext=m4a]',
654 },
655 },
656 # JS player signature function name containing $
657 {
658 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
659 'info_dict': {
660 'id': 'nfWlot6h_JM',
661 'ext': 'm4a',
662 'title': 'Taylor Swift - Shake It Off',
663 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
664 'duration': 242,
665 'uploader': 'TaylorSwiftVEVO',
666 'uploader_id': 'TaylorSwiftVEVO',
667 'upload_date': '20140818',
668 'creator': 'Taylor Swift',
669 },
670 'params': {
671 'youtube_include_dash_manifest': True,
672 'format': '141/bestaudio[ext=m4a]',
673 },
674 },
675 # Controversy video
676 {
677 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
678 'info_dict': {
679 'id': 'T4XJQO3qol8',
680 'ext': 'mp4',
681 'duration': 219,
682 'upload_date': '20100909',
683 'uploader': 'Amazing Atheist',
684 'uploader_id': 'TheAmazingAtheist',
685 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
686 'title': 'Burning Everyone\'s Koran',
687 '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',
688 }
689 },
690 # Normal age-gate video (No vevo, embed allowed)
691 {
692 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
693 'info_dict': {
694 'id': 'HtVdAasjOgU',
695 'ext': 'mp4',
696 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
697 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
698 'duration': 142,
699 'uploader': 'The Witcher',
700 'uploader_id': 'WitcherGame',
701 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
702 'upload_date': '20140605',
703 'age_limit': 18,
704 },
705 },
706 # Age-gate video with encrypted signature
707 {
708 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
709 'info_dict': {
710 'id': '6kLq3WMV1nU',
711 'ext': 'mp4',
712 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
713 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
714 'duration': 246,
715 'uploader': 'LloydVEVO',
716 'uploader_id': 'LloydVEVO',
717 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
718 'upload_date': '20110629',
719 'age_limit': 18,
720 },
721 },
722 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
723 # YouTube Red ad is not captured for creator
724 {
725 'url': '__2ABJjxzNo',
726 'info_dict': {
727 'id': '__2ABJjxzNo',
728 'ext': 'mp4',
729 'duration': 266,
730 'upload_date': '20100430',
731 'uploader_id': 'deadmau5',
732 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
733 'creator': 'deadmau5',
734 'description': 'md5:12c56784b8032162bb936a5f76d55360',
735 'uploader': 'deadmau5',
736 'title': 'Deadmau5 - Some Chords (HD)',
737 'alt_title': 'Some Chords',
738 },
739 'expected_warnings': [
740 'DASH manifest missing',
741 ]
742 },
743 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
744 {
745 'url': 'lqQg6PlCWgI',
746 'info_dict': {
747 'id': 'lqQg6PlCWgI',
748 'ext': 'mp4',
749 'duration': 6085,
750 'upload_date': '20150827',
751 'uploader_id': 'olympic',
752 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
753 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
754 'uploader': 'Olympic',
755 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
756 },
757 'params': {
758 'skip_download': 'requires avconv',
759 }
760 },
761 # Non-square pixels
762 {
763 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
764 'info_dict': {
765 'id': '_b-2C3KPAM0',
766 'ext': 'mp4',
767 'stretched_ratio': 16 / 9.,
768 'duration': 85,
769 'upload_date': '20110310',
770 'uploader_id': 'AllenMeow',
771 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
772 'description': 'made by Wacom from Korea | 字幕&åŠ ę²¹ę·»é†‹ by TY\'s Allen | ę„Ÿč¬heylisa00cavey1001同å­øē†±ęƒ…ęä¾›ę¢—åŠēæ»č­Æ',
773 'uploader': 'å­«į„‹į„…',
774 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
775 },
776 },
777 # url_encoded_fmt_stream_map is empty string
778 {
779 'url': 'qEJwOuvDf7I',
780 'info_dict': {
781 'id': 'qEJwOuvDf7I',
782 'ext': 'webm',
783 'title': 'ŠžŠ±ŃŃƒŠ¶Š“ŠµŠ½ŠøŠµ суŠ“ŠµŠ±Š½Š¾Š¹ ŠæрŠ°ŠŗтŠøŠŗŠø ŠæŠ¾ Š²Ń‹Š±Š¾Ń€Š°Š¼ 14 сŠµŠ½Ń‚яŠ±Ń€Ń 2014 Š³Š¾Š“Š° Š² Š”Š°Š½Šŗт-ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Šµ',
784 'description': '',
785 'upload_date': '20150404',
786 'uploader_id': 'spbelect',
787 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
788 },
789 'params': {
790 'skip_download': 'requires avconv',
791 },
792 'skip': 'This live event has ended.',
793 },
794 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
795 {
796 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
797 'info_dict': {
798 'id': 'FIl7x6_3R5Y',
799 'ext': 'webm',
800 'title': 'md5:7b81415841e02ecd4313668cde88737a',
801 'description': 'md5:116377fd2963b81ec4ce64b542173306',
802 'duration': 220,
803 'upload_date': '20150625',
804 'uploader_id': 'dorappi2000',
805 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
806 'uploader': 'dorappi2000',
807 'formats': 'mincount:31',
808 },
809 'skip': 'not actual anymore',
810 },
811 # DASH manifest with segment_list
812 {
813 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
814 'md5': '8ce563a1d667b599d21064e982ab9e31',
815 'info_dict': {
816 'id': 'CsmdDsKjzN8',
817 'ext': 'mp4',
818 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
819 'uploader': 'Airtek',
820 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
821 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
822 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
823 },
824 'params': {
825 'youtube_include_dash_manifest': True,
826 'format': '135', # bestvideo
827 },
828 'skip': 'This live event has ended.',
829 },
830 {
831 # Multifeed videos (multiple cameras), URL is for Main Camera
832 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
833 'info_dict': {
834 'id': 'jqWvoWXjCVs',
835 'title': 'teamPGP: Rocket League Noob Stream',
836 'description': 'md5:dc7872fb300e143831327f1bae3af010',
837 },
838 'playlist': [{
839 'info_dict': {
840 'id': 'jqWvoWXjCVs',
841 'ext': 'mp4',
842 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
843 'description': 'md5:dc7872fb300e143831327f1bae3af010',
844 'duration': 7335,
845 'upload_date': '20150721',
846 'uploader': 'Beer Games Beer',
847 'uploader_id': 'beergamesbeer',
848 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
849 'license': 'Standard YouTube License',
850 },
851 }, {
852 'info_dict': {
853 'id': '6h8e8xoXJzg',
854 'ext': 'mp4',
855 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
856 'description': 'md5:dc7872fb300e143831327f1bae3af010',
857 'duration': 7337,
858 'upload_date': '20150721',
859 'uploader': 'Beer Games Beer',
860 'uploader_id': 'beergamesbeer',
861 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
862 'license': 'Standard YouTube License',
863 },
864 }, {
865 'info_dict': {
866 'id': 'PUOgX5z9xZw',
867 'ext': 'mp4',
868 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
869 'description': 'md5:dc7872fb300e143831327f1bae3af010',
870 'duration': 7337,
871 'upload_date': '20150721',
872 'uploader': 'Beer Games Beer',
873 'uploader_id': 'beergamesbeer',
874 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
875 'license': 'Standard YouTube License',
876 },
877 }, {
878 'info_dict': {
879 'id': 'teuwxikvS5k',
880 'ext': 'mp4',
881 'title': 'teamPGP: Rocket League Noob Stream (zim)',
882 'description': 'md5:dc7872fb300e143831327f1bae3af010',
883 'duration': 7334,
884 'upload_date': '20150721',
885 'uploader': 'Beer Games Beer',
886 'uploader_id': 'beergamesbeer',
887 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
888 'license': 'Standard YouTube License',
889 },
890 }],
891 'params': {
892 'skip_download': True,
893 },
894 'skip': 'This video is not available.',
895 },
896 {
897 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
898 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
899 'info_dict': {
900 'id': 'gVfLd0zydlo',
901 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
902 },
903 'playlist_count': 2,
904 'skip': 'Not multifeed anymore',
905 },
906 {
907 'url': 'https://vid.plus/FlRa-iH7PGw',
908 'only_matching': True,
909 },
910 {
911 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
912 'only_matching': True,
913 },
914 {
915 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
916 # Also tests cut-off URL expansion in video description (see
917 # https://github.com/ytdl-org/youtube-dl/issues/1892,
918 # https://github.com/ytdl-org/youtube-dl/issues/8164)
919 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
920 'info_dict': {
921 'id': 'lsguqyKfVQg',
922 'ext': 'mp4',
923 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
924 'alt_title': 'Dark Walk - Position Music',
925 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
926 'duration': 133,
927 'upload_date': '20151119',
928 'uploader_id': 'IronSoulElf',
929 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
930 'uploader': 'IronSoulElf',
931 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
932 'track': 'Dark Walk - Position Music',
933 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
934 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
935 },
936 'params': {
937 'skip_download': True,
938 },
939 },
940 {
941 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
942 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
943 'only_matching': True,
944 },
945 {
946 # Video with yt:stretch=17:0
947 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
948 'info_dict': {
949 'id': 'Q39EVAstoRM',
950 'ext': 'mp4',
951 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
952 'description': 'md5:ee18a25c350637c8faff806845bddee9',
953 'upload_date': '20151107',
954 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
955 'uploader': 'CH GAMER DROID',
956 },
957 'params': {
958 'skip_download': True,
959 },
960 'skip': 'This video does not exist.',
961 },
962 {
963 # Video licensed under Creative Commons
964 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
965 'info_dict': {
966 'id': 'M4gD1WSo5mA',
967 'ext': 'mp4',
968 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
969 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
970 'duration': 721,
971 'upload_date': '20150127',
972 'uploader_id': 'BerkmanCenter',
973 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
974 'uploader': 'The Berkman Klein Center for Internet & Society',
975 'license': 'Creative Commons Attribution license (reuse allowed)',
976 },
977 'params': {
978 'skip_download': True,
979 },
980 },
981 {
982 # Channel-like uploader_url
983 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
984 'info_dict': {
985 'id': 'eQcmzGIKrzg',
986 'ext': 'mp4',
987 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
988 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
989 'duration': 4060,
990 'upload_date': '20151119',
991 'uploader': 'Bernie Sanders',
992 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
993 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
994 'license': 'Creative Commons Attribution license (reuse allowed)',
995 },
996 'params': {
997 'skip_download': True,
998 },
999 },
1000 {
1001 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1002 'only_matching': True,
1003 },
1004 {
1005 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1006 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1007 'only_matching': True,
1008 },
1009 {
1010 # Rental video preview
1011 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1012 'info_dict': {
1013 'id': 'uGpuVWrhIzE',
1014 'ext': 'mp4',
1015 'title': 'Piku - Trailer',
1016 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1017 'upload_date': '20150811',
1018 'uploader': 'FlixMatrix',
1019 'uploader_id': 'FlixMatrixKaravan',
1020 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1021 'license': 'Standard YouTube License',
1022 },
1023 'params': {
1024 'skip_download': True,
1025 },
1026 'skip': 'This video is not available.',
1027 },
1028 {
1029 # YouTube Red video with episode data
1030 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1031 'info_dict': {
1032 'id': 'iqKdEhx-dD4',
1033 'ext': 'mp4',
1034 'title': 'Isolation - Mind Field (Ep 1)',
1035 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1036 'duration': 2085,
1037 'upload_date': '20170118',
1038 'uploader': 'Vsauce',
1039 'uploader_id': 'Vsauce',
1040 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1041 'series': 'Mind Field',
1042 'season_number': 1,
1043 'episode_number': 1,
1044 },
1045 'params': {
1046 'skip_download': True,
1047 },
1048 'expected_warnings': [
1049 'Skipping DASH manifest',
1050 ],
1051 },
1052 {
1053 # The following content has been identified by the YouTube community
1054 # as inappropriate or offensive to some audiences.
1055 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1056 'info_dict': {
1057 'id': '6SJNVb0GnPI',
1058 'ext': 'mp4',
1059 'title': 'Race Differences in Intelligence',
1060 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1061 'duration': 965,
1062 'upload_date': '20140124',
1063 'uploader': 'New Century Foundation',
1064 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1065 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1066 },
1067 'params': {
1068 'skip_download': True,
1069 },
1070 },
1071 {
1072 # itag 212
1073 'url': '1t24XAntNCY',
1074 'only_matching': True,
1075 },
1076 {
1077 # geo restricted to JP
1078 'url': 'sJL6WA-aGkQ',
1079 'only_matching': True,
1080 },
1081 {
1082 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1083 'only_matching': True,
1084 },
1085 {
1086 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1087 'only_matching': True,
1088 },
1089 {
1090 # DRM protected
1091 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1092 'only_matching': True,
1093 },
1094 {
1095 # Video with unsupported adaptive stream type formats
1096 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1097 'info_dict': {
1098 'id': 'Z4Vy8R84T1U',
1099 'ext': 'mp4',
1100 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1101 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1102 'duration': 433,
1103 'upload_date': '20130923',
1104 'uploader': 'Amelia Putri Harwita',
1105 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1106 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1107 'formats': 'maxcount:10',
1108 },
1109 'params': {
1110 'skip_download': True,
1111 'youtube_include_dash_manifest': False,
1112 },
1113 },
1114 {
1115 # Youtube Music Auto-generated description
1116 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1117 'info_dict': {
1118 'id': 'MgNrAu2pzNs',
1119 'ext': 'mp4',
1120 'title': 'Voyeur Girl',
1121 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1122 'upload_date': '20190312',
1123 'uploader': 'Various Artists - Topic',
1124 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1125 'artist': 'Stephen',
1126 'track': 'Voyeur Girl',
1127 'album': 'it\'s too much love to know my dear',
1128 'release_date': '20190313',
1129 'release_year': 2019,
1130 },
1131 'params': {
1132 'skip_download': True,
1133 },
1134 },
1135 {
1136 # Youtube Music Auto-generated description
1137 # Retrieve 'artist' field from 'Artist:' in video description
1138 # when it is present on youtube music video
1139 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1140 'info_dict': {
1141 'id': 'k0jLE7tTwjY',
1142 'ext': 'mp4',
1143 'title': 'Latch Feat. Sam Smith',
1144 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1145 'upload_date': '20150110',
1146 'uploader': 'Various Artists - Topic',
1147 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1148 'artist': 'Disclosure',
1149 'track': 'Latch Feat. Sam Smith',
1150 'album': 'Latch Featuring Sam Smith',
1151 'release_date': '20121008',
1152 'release_year': 2012,
1153 },
1154 'params': {
1155 'skip_download': True,
1156 },
1157 },
1158 {
1159 # Youtube Music Auto-generated description
1160 # handle multiple artists on youtube music video
1161 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1162 'info_dict': {
1163 'id': '74qn0eJSjpA',
1164 'ext': 'mp4',
1165 'title': 'Eastside',
1166 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1167 'upload_date': '20180710',
1168 'uploader': 'Benny Blanco - Topic',
1169 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1170 'artist': 'benny blanco, Halsey, Khalid',
1171 'track': 'Eastside',
1172 'album': 'Eastside',
1173 'release_date': '20180713',
1174 'release_year': 2018,
1175 },
1176 'params': {
1177 'skip_download': True,
1178 },
1179 },
1180 {
1181 # Youtube Music Auto-generated description
1182 # handle youtube music video with release_year and no release_date
1183 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1184 'info_dict': {
1185 'id': '-hcAI0g-f5M',
1186 'ext': 'mp4',
1187 'title': 'Put It On Me',
1188 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1189 'upload_date': '20180426',
1190 'uploader': 'Matt Maeson - Topic',
1191 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1192 'artist': 'Matt Maeson',
1193 'track': 'Put It On Me',
1194 'album': 'The Hearse',
1195 'release_date': None,
1196 'release_year': 2018,
1197 },
1198 'params': {
1199 'skip_download': True,
1200 },
1201 },
1202 ]
1203
1204 def __init__(self, *args, **kwargs):
1205 super(YoutubeIE, self).__init__(*args, **kwargs)
1206 self._player_cache = {}
1207
1208 def report_video_info_webpage_download(self, video_id):
1209 """Report attempt to download video info webpage."""
1210 self.to_screen('%s: Downloading video info webpage' % video_id)
1211
1212 def report_information_extraction(self, video_id):
1213 """Report attempt to extract video information."""
1214 self.to_screen('%s: Extracting video information' % video_id)
1215
1216 def report_unavailable_format(self, video_id, format):
1217 """Report extracted video URL."""
1218 self.to_screen('%s: Format %s not available' % (video_id, format))
1219
1220 def report_rtmp_download(self):
1221 """Indicate the download will use the RTMP protocol."""
1222 self.to_screen('RTMP download detected')
1223
1224 def _signature_cache_id(self, example_sig):
1225 """ Return a string representation of a signature """
1226 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1227
1228 def _extract_signature_function(self, video_id, player_url, example_sig):
1229 id_m = re.match(
1230 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1231 player_url)
1232 if not id_m:
1233 raise ExtractorError('Cannot identify player %r' % player_url)
1234 player_type = id_m.group('ext')
1235 player_id = id_m.group('id')
1236
1237 # Read from filesystem cache
1238 func_id = '%s_%s_%s' % (
1239 player_type, player_id, self._signature_cache_id(example_sig))
1240 assert os.path.basename(func_id) == func_id
1241
1242 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1243 if cache_spec is not None:
1244 return lambda s: ''.join(s[i] for i in cache_spec)
1245
1246 download_note = (
1247 'Downloading player %s' % player_url
1248 if self._downloader.params.get('verbose') else
1249 'Downloading %s player %s' % (player_type, player_id)
1250 )
1251 if player_type == 'js':
1252 code = self._download_webpage(
1253 player_url, video_id,
1254 note=download_note,
1255 errnote='Download of %s failed' % player_url)
1256 res = self._parse_sig_js(code)
1257 elif player_type == 'swf':
1258 urlh = self._request_webpage(
1259 player_url, video_id,
1260 note=download_note,
1261 errnote='Download of %s failed' % player_url)
1262 code = urlh.read()
1263 res = self._parse_sig_swf(code)
1264 else:
1265 assert False, 'Invalid player type %r' % player_type
1266
1267 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1268 cache_res = res(test_string)
1269 cache_spec = [ord(c) for c in cache_res]
1270
1271 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1272 return res
1273
1274 def _print_sig_code(self, func, example_sig):
1275 def gen_sig_code(idxs):
1276 def _genslice(start, end, step):
1277 starts = '' if start == 0 else str(start)
1278 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1279 steps = '' if step == 1 else (':%d' % step)
1280 return 's[%s%s%s]' % (starts, ends, steps)
1281
1282 step = None
1283 # Quelch pyflakes warnings - start will be set when step is set
1284 start = '(Never used)'
1285 for i, prev in zip(idxs[1:], idxs[:-1]):
1286 if step is not None:
1287 if i - prev == step:
1288 continue
1289 yield _genslice(start, prev, step)
1290 step = None
1291 continue
1292 if i - prev in [-1, 1]:
1293 step = i - prev
1294 start = prev
1295 continue
1296 else:
1297 yield 's[%d]' % prev
1298 if step is None:
1299 yield 's[%d]' % i
1300 else:
1301 yield _genslice(start, i, step)
1302
1303 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1304 cache_res = func(test_string)
1305 cache_spec = [ord(c) for c in cache_res]
1306 expr_code = ' + '.join(gen_sig_code(cache_spec))
1307 signature_id_tuple = '(%s)' % (
1308 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1309 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1310 ' return %s\n') % (signature_id_tuple, expr_code)
1311 self.to_screen('Extracted signature function:\n' + code)
1312
1313 def _parse_sig_js(self, jscode):
1314 funcname = self._search_regex(
1315 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1316 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1317 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1318 # Obsolete patterns
1319 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1320 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1321 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1322 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1323 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1324 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1325 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1326 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1327 jscode, 'Initial JS player signature function name', group='sig')
1328
1329 jsi = JSInterpreter(jscode)
1330 initial_function = jsi.extract_function(funcname)
1331 return lambda s: initial_function([s])
1332
1333 def _parse_sig_swf(self, file_contents):
1334 swfi = SWFInterpreter(file_contents)
1335 TARGET_CLASSNAME = 'SignatureDecipher'
1336 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1337 initial_function = swfi.extract_function(searched_class, 'decipher')
1338 return lambda s: initial_function([s])
1339
1340 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1341 """Turn the encrypted s field into a working signature"""
1342
1343 if player_url is None:
1344 raise ExtractorError('Cannot decrypt signature without player_url')
1345
1346 if player_url.startswith('//'):
1347 player_url = 'https:' + player_url
1348 elif not re.match(r'https?://', player_url):
1349 player_url = compat_urlparse.urljoin(
1350 'https://www.youtube.com', player_url)
1351 try:
1352 player_id = (player_url, self._signature_cache_id(s))
1353 if player_id not in self._player_cache:
1354 func = self._extract_signature_function(
1355 video_id, player_url, s
1356 )
1357 self._player_cache[player_id] = func
1358 func = self._player_cache[player_id]
1359 if self._downloader.params.get('youtube_print_sig_code'):
1360 self._print_sig_code(func, s)
1361 return func(s)
1362 except Exception as e:
1363 tb = traceback.format_exc()
1364 raise ExtractorError(
1365 'Signature extraction failed: ' + tb, cause=e)
1366
1367 def _get_subtitles(self, video_id, webpage):
1368 try:
1369 subs_doc = self._download_xml(
1370 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1371 video_id, note=False)
1372 except ExtractorError as err:
1373 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1374 return {}
1375
1376 sub_lang_list = {}
1377 for track in subs_doc.findall('track'):
1378 lang = track.attrib['lang_code']
1379 if lang in sub_lang_list:
1380 continue
1381 sub_formats = []
1382 for ext in self._SUBTITLE_FORMATS:
1383 params = compat_urllib_parse_urlencode({
1384 'lang': lang,
1385 'v': video_id,
1386 'fmt': ext,
1387 'name': track.attrib['name'].encode('utf-8'),
1388 })
1389 sub_formats.append({
1390 'url': 'https://www.youtube.com/api/timedtext?' + params,
1391 'ext': ext,
1392 })
1393 sub_lang_list[lang] = sub_formats
1394 if not sub_lang_list:
1395 self._downloader.report_warning('video doesn\'t have subtitles')
1396 return {}
1397 return sub_lang_list
1398
1399 def _get_ytplayer_config(self, video_id, webpage):
1400 patterns = (
1401 # User data may contain arbitrary character sequences that may affect
1402 # JSON extraction with regex, e.g. when '};' is contained the second
1403 # regex won't capture the whole JSON. Yet working around by trying more
1404 # concrete regex first keeping in mind proper quoted string handling
1405 # to be implemented in future that will replace this workaround (see
1406 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1407 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1408 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1409 r';ytplayer\.config\s*=\s*({.+?});',
1410 )
1411 config = self._search_regex(
1412 patterns, webpage, 'ytplayer.config', default=None)
1413 if config:
1414 return self._parse_json(
1415 uppercase_escape(config), video_id, fatal=False)
1416
1417 def _get_automatic_captions(self, video_id, webpage):
1418 """We need the webpage for getting the captions url, pass it as an
1419 argument to speed up the process."""
1420 self.to_screen('%s: Looking for automatic captions' % video_id)
1421 player_config = self._get_ytplayer_config(video_id, webpage)
1422 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1423 if not player_config:
1424 self._downloader.report_warning(err_msg)
1425 return {}
1426 try:
1427 args = player_config['args']
1428 caption_url = args.get('ttsurl')
1429 if caption_url:
1430 timestamp = args['timestamp']
1431 # We get the available subtitles
1432 list_params = compat_urllib_parse_urlencode({
1433 'type': 'list',
1434 'tlangs': 1,
1435 'asrs': 1,
1436 })
1437 list_url = caption_url + '&' + list_params
1438 caption_list = self._download_xml(list_url, video_id)
1439 original_lang_node = caption_list.find('track')
1440 if original_lang_node is None:
1441 self._downloader.report_warning('Video doesn\'t have automatic captions')
1442 return {}
1443 original_lang = original_lang_node.attrib['lang_code']
1444 caption_kind = original_lang_node.attrib.get('kind', '')
1445
1446 sub_lang_list = {}
1447 for lang_node in caption_list.findall('target'):
1448 sub_lang = lang_node.attrib['lang_code']
1449 sub_formats = []
1450 for ext in self._SUBTITLE_FORMATS:
1451 params = compat_urllib_parse_urlencode({
1452 'lang': original_lang,
1453 'tlang': sub_lang,
1454 'fmt': ext,
1455 'ts': timestamp,
1456 'kind': caption_kind,
1457 })
1458 sub_formats.append({
1459 'url': caption_url + '&' + params,
1460 'ext': ext,
1461 })
1462 sub_lang_list[sub_lang] = sub_formats
1463 return sub_lang_list
1464
1465 def make_captions(sub_url, sub_langs):
1466 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1467 caption_qs = compat_parse_qs(parsed_sub_url.query)
1468 captions = {}
1469 for sub_lang in sub_langs:
1470 sub_formats = []
1471 for ext in self._SUBTITLE_FORMATS:
1472 caption_qs.update({
1473 'tlang': [sub_lang],
1474 'fmt': [ext],
1475 })
1476 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1477 query=compat_urllib_parse_urlencode(caption_qs, True)))
1478 sub_formats.append({
1479 'url': sub_url,
1480 'ext': ext,
1481 })
1482 captions[sub_lang] = sub_formats
1483 return captions
1484
1485 # New captions format as of 22.06.2017
1486 player_response = args.get('player_response')
1487 if player_response and isinstance(player_response, compat_str):
1488 player_response = self._parse_json(
1489 player_response, video_id, fatal=False)
1490 if player_response:
1491 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1492 base_url = renderer['captionTracks'][0]['baseUrl']
1493 sub_lang_list = []
1494 for lang in renderer['translationLanguages']:
1495 lang_code = lang.get('languageCode')
1496 if lang_code:
1497 sub_lang_list.append(lang_code)
1498 return make_captions(base_url, sub_lang_list)
1499
1500 # Some videos don't provide ttsurl but rather caption_tracks and
1501 # caption_translation_languages (e.g. 20LmZk1hakA)
1502 # Does not used anymore as of 22.06.2017
1503 caption_tracks = args['caption_tracks']
1504 caption_translation_languages = args['caption_translation_languages']
1505 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1506 sub_lang_list = []
1507 for lang in caption_translation_languages.split(','):
1508 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1509 sub_lang = lang_qs.get('lc', [None])[0]
1510 if sub_lang:
1511 sub_lang_list.append(sub_lang)
1512 return make_captions(caption_url, sub_lang_list)
1513 # An extractor error can be raise by the download process if there are
1514 # no automatic captions but there are subtitles
1515 except (KeyError, IndexError, ExtractorError):
1516 self._downloader.report_warning(err_msg)
1517 return {}
1518
1519 def _mark_watched(self, video_id, video_info, player_response):
1520 playback_url = url_or_none(try_get(
1521 player_response,
1522 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1523 video_info, lambda x: x['videostats_playback_base_url'][0]))
1524 if not playback_url:
1525 return
1526 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1527 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1528
1529 # cpn generation algorithm is reverse engineered from base.js.
1530 # In fact it works even with dummy cpn.
1531 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1532 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1533
1534 qs.update({
1535 'ver': ['2'],
1536 'cpn': [cpn],
1537 })
1538 playback_url = compat_urlparse.urlunparse(
1539 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1540
1541 self._download_webpage(
1542 playback_url, video_id, 'Marking watched',
1543 'Unable to mark watched', fatal=False)
1544
1545 @staticmethod
1546 def _extract_urls(webpage):
1547 # Embedded YouTube player
1548 entries = [
1549 unescapeHTML(mobj.group('url'))
1550 for mobj in re.finditer(r'''(?x)
1551 (?:
1552 <iframe[^>]+?src=|
1553 data-video-url=|
1554 <embed[^>]+?src=|
1555 embedSWF\(?:\s*|
1556 <object[^>]+data=|
1557 new\s+SWFObject\(
1558 )
1559 (["\'])
1560 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1561 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1562 \1''', webpage)]
1563
1564 # lazyYT YouTube embed
1565 entries.extend(list(map(
1566 unescapeHTML,
1567 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1568
1569 # Wordpress "YouTube Video Importer" plugin
1570 matches = re.findall(r'''(?x)<div[^>]+
1571 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1572 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1573 entries.extend(m[-1] for m in matches)
1574
1575 return entries
1576
1577 @staticmethod
1578 def _extract_url(webpage):
1579 urls = YoutubeIE._extract_urls(webpage)
1580 return urls[0] if urls else None
1581
1582 @classmethod
1583 def extract_id(cls, url):
1584 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1585 if mobj is None:
1586 raise ExtractorError('Invalid URL: %s' % url)
1587 video_id = mobj.group(2)
1588 return video_id
1589
1590 def _extract_annotations(self, video_id):
1591 return self._download_webpage(
1592 'https://www.youtube.com/annotations_invideo', video_id,
1593 note='Downloading annotations',
1594 errnote='Unable to download video annotations', fatal=False,
1595 query={
1596 'features': 1,
1597 'legacy': 1,
1598 'video_id': video_id,
1599 })
1600
1601 @staticmethod
1602 def _extract_chapters(description, duration):
1603 if not description:
1604 return None
1605 chapter_lines = re.findall(
1606 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1607 description)
1608 if not chapter_lines:
1609 return None
1610 chapters = []
1611 for next_num, (chapter_line, time_point) in enumerate(
1612 chapter_lines, start=1):
1613 start_time = parse_duration(time_point)
1614 if start_time is None:
1615 continue
1616 if start_time > duration:
1617 break
1618 end_time = (duration if next_num == len(chapter_lines)
1619 else parse_duration(chapter_lines[next_num][1]))
1620 if end_time is None:
1621 continue
1622 if end_time > duration:
1623 end_time = duration
1624 if start_time > end_time:
1625 break
1626 chapter_title = re.sub(
1627 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1628 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1629 chapters.append({
1630 'start_time': start_time,
1631 'end_time': end_time,
1632 'title': chapter_title,
1633 })
1634 return chapters
1635
1636 def _real_extract(self, url):
1637 url, smuggled_data = unsmuggle_url(url, {})
1638
1639 proto = (
1640 'http' if self._downloader.params.get('prefer_insecure', False)
1641 else 'https')
1642
1643 start_time = None
1644 end_time = None
1645 parsed_url = compat_urllib_parse_urlparse(url)
1646 for component in [parsed_url.fragment, parsed_url.query]:
1647 query = compat_parse_qs(component)
1648 if start_time is None and 't' in query:
1649 start_time = parse_duration(query['t'][0])
1650 if start_time is None and 'start' in query:
1651 start_time = parse_duration(query['start'][0])
1652 if end_time is None and 'end' in query:
1653 end_time = parse_duration(query['end'][0])
1654
1655 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1656 mobj = re.search(self._NEXT_URL_RE, url)
1657 if mobj:
1658 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1659 video_id = self.extract_id(url)
1660
1661 # Get video webpage
1662 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1663 video_webpage = self._download_webpage(url, video_id)
1664
1665 # Attempt to extract SWF player URL
1666 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1667 if mobj is not None:
1668 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1669 else:
1670 player_url = None
1671
1672 dash_mpds = []
1673
1674 def add_dash_mpd(video_info):
1675 dash_mpd = video_info.get('dashmpd')
1676 if dash_mpd and dash_mpd[0] not in dash_mpds:
1677 dash_mpds.append(dash_mpd[0])
1678
1679 def add_dash_mpd_pr(pl_response):
1680 dash_mpd = url_or_none(try_get(
1681 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1682 compat_str))
1683 if dash_mpd and dash_mpd not in dash_mpds:
1684 dash_mpds.append(dash_mpd)
1685
1686 is_live = None
1687 view_count = None
1688
1689 def extract_view_count(v_info):
1690 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1691
1692 def extract_token(v_info):
1693 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1694
1695 player_response = {}
1696
1697 # Get video info
1698 embed_webpage = None
1699 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1700 age_gate = True
1701 # We simulate the access to the video from www.youtube.com/v/{video_id}
1702 # this can be viewed without login into Youtube
1703 url = proto + '://www.youtube.com/embed/%s' % video_id
1704 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1705 data = compat_urllib_parse_urlencode({
1706 'video_id': video_id,
1707 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1708 'sts': self._search_regex(
1709 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1710 })
1711 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1712 video_info_webpage = self._download_webpage(
1713 video_info_url, video_id,
1714 note='Refetching age-gated info webpage',
1715 errnote='unable to download video info webpage')
1716 video_info = compat_parse_qs(video_info_webpage)
1717 add_dash_mpd(video_info)
1718 else:
1719 age_gate = False
1720 video_info = None
1721 sts = None
1722 # Try looking directly into the video webpage
1723 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1724 if ytplayer_config:
1725 args = ytplayer_config['args']
1726 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1727 # Convert to the same format returned by compat_parse_qs
1728 video_info = dict((k, [v]) for k, v in args.items())
1729 add_dash_mpd(video_info)
1730 # Rental video is not rented but preview is available (e.g.
1731 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1732 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1733 if not video_info and args.get('ypc_vid'):
1734 return self.url_result(
1735 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1736 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1737 is_live = True
1738 sts = ytplayer_config.get('sts')
1739 if not player_response:
1740 pl_response = str_or_none(args.get('player_response'))
1741 if pl_response:
1742 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1743 if isinstance(pl_response, dict):
1744 player_response = pl_response
1745 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1746 add_dash_mpd_pr(player_response)
1747 # We also try looking in get_video_info since it may contain different dashmpd
1748 # URL that points to a DASH manifest with possibly different itag set (some itags
1749 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1750 # manifest pointed by get_video_info's dashmpd).
1751 # The general idea is to take a union of itags of both DASH manifests (for example
1752 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1753 self.report_video_info_webpage_download(video_id)
1754 for el in ('embedded', 'detailpage', 'vevo', ''):
1755 query = {
1756 'video_id': video_id,
1757 'ps': 'default',
1758 'eurl': '',
1759 'gl': 'US',
1760 'hl': 'en',
1761 }
1762 if el:
1763 query['el'] = el
1764 if sts:
1765 query['sts'] = sts
1766 video_info_webpage = self._download_webpage(
1767 '%s://www.youtube.com/get_video_info' % proto,
1768 video_id, note=False,
1769 errnote='unable to download video info webpage',
1770 fatal=False, query=query)
1771 if not video_info_webpage:
1772 continue
1773 get_video_info = compat_parse_qs(video_info_webpage)
1774 if not player_response:
1775 pl_response = get_video_info.get('player_response', [None])[0]
1776 if isinstance(pl_response, dict):
1777 player_response = pl_response
1778 add_dash_mpd_pr(player_response)
1779 add_dash_mpd(get_video_info)
1780 if view_count is None:
1781 view_count = extract_view_count(get_video_info)
1782 if not video_info:
1783 video_info = get_video_info
1784 get_token = extract_token(get_video_info)
1785 if get_token:
1786 # Different get_video_info requests may report different results, e.g.
1787 # some may report video unavailability, but some may serve it without
1788 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1789 # the original webpage as well as el=info and el=embedded get_video_info
1790 # requests report video unavailability due to geo restriction while
1791 # el=detailpage succeeds and returns valid data). This is probably
1792 # due to YouTube measures against IP ranges of hosting providers.
1793 # Working around by preferring the first succeeded video_info containing
1794 # the token if no such video_info yet was found.
1795 token = extract_token(video_info)
1796 if not token:
1797 video_info = get_video_info
1798 break
1799
1800 def extract_unavailable_message():
1801 return self._html_search_regex(
1802 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1803 video_webpage, 'unavailable message', default=None)
1804
1805 if not video_info:
1806 unavailable_message = extract_unavailable_message()
1807 if not unavailable_message:
1808 unavailable_message = 'Unable to extract video data'
1809 raise ExtractorError(
1810 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1811
1812 video_details = try_get(
1813 player_response, lambda x: x['videoDetails'], dict) or {}
1814
1815 # title
1816 if 'title' in video_info:
1817 video_title = video_info['title'][0]
1818 elif 'title' in player_response:
1819 video_title = video_details['title']
1820 else:
1821 self._downloader.report_warning('Unable to extract video title')
1822 video_title = '_'
1823
1824 # description
1825 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1826 if video_description:
1827
1828 def replace_url(m):
1829 redir_url = compat_urlparse.urljoin(url, m.group(1))
1830 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1831 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1832 qs = compat_parse_qs(parsed_redir_url.query)
1833 q = qs.get('q')
1834 if q and q[0]:
1835 return q[0]
1836 return redir_url
1837
1838 description_original = video_description = re.sub(r'''(?x)
1839 <a\s+
1840 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1841 (?:title|href)="([^"]+)"\s+
1842 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1843 class="[^"]*"[^>]*>
1844 [^<]+\.{3}\s*
1845 </a>
1846 ''', replace_url, video_description)
1847 video_description = clean_html(video_description)
1848 else:
1849 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1850 if fd_mobj:
1851 video_description = unescapeHTML(fd_mobj.group(1))
1852 else:
1853 video_description = ''
1854
1855 if not smuggled_data.get('force_singlefeed', False):
1856 if not self._downloader.params.get('noplaylist'):
1857 multifeed_metadata_list = try_get(
1858 player_response,
1859 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1860 compat_str) or try_get(
1861 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1862 if multifeed_metadata_list:
1863 entries = []
1864 feed_ids = []
1865 for feed in multifeed_metadata_list.split(','):
1866 # Unquote should take place before split on comma (,) since textual
1867 # fields may contain comma as well (see
1868 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1869 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1870 entries.append({
1871 '_type': 'url_transparent',
1872 'ie_key': 'Youtube',
1873 'url': smuggle_url(
1874 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1875 {'force_singlefeed': True}),
1876 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1877 })
1878 feed_ids.append(feed_data['id'][0])
1879 self.to_screen(
1880 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1881 % (', '.join(feed_ids), video_id))
1882 return self.playlist_result(entries, video_id, video_title, video_description)
1883 else:
1884 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1885
1886 if view_count is None:
1887 view_count = extract_view_count(video_info)
1888 if view_count is None and video_details:
1889 view_count = int_or_none(video_details.get('viewCount'))
1890
1891 # Check for "rental" videos
1892 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1893 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1894
1895 def _extract_filesize(media_url):
1896 return int_or_none(self._search_regex(
1897 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1898
1899 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1900 self.report_rtmp_download()
1901 formats = [{
1902 'format_id': '_rtmp',
1903 'protocol': 'rtmp',
1904 'url': video_info['conn'][0],
1905 'player_url': player_url,
1906 }]
1907 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):
1908 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1909 if 'rtmpe%3Dyes' in encoded_url_map:
1910 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1911 formats_spec = {}
1912 fmt_list = video_info.get('fmt_list', [''])[0]
1913 if fmt_list:
1914 for fmt in fmt_list.split(','):
1915 spec = fmt.split('/')
1916 if len(spec) > 1:
1917 width_height = spec[1].split('x')
1918 if len(width_height) == 2:
1919 formats_spec[spec[0]] = {
1920 'resolution': spec[1],
1921 'width': int_or_none(width_height[0]),
1922 'height': int_or_none(width_height[1]),
1923 }
1924 q = qualities(['small', 'medium', 'hd720'])
1925 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1926 if streaming_formats:
1927 for fmt in streaming_formats:
1928 itag = str_or_none(fmt.get('itag'))
1929 if not itag:
1930 continue
1931 quality = fmt.get('quality')
1932 quality_label = fmt.get('qualityLabel') or quality
1933 formats_spec[itag] = {
1934 'asr': int_or_none(fmt.get('audioSampleRate')),
1935 'filesize': int_or_none(fmt.get('contentLength')),
1936 'format_note': quality_label,
1937 'fps': int_or_none(fmt.get('fps')),
1938 'height': int_or_none(fmt.get('height')),
1939 'quality': q(quality),
1940 # bitrate for itag 43 is always 2147483647
1941 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1942 'width': int_or_none(fmt.get('width')),
1943 }
1944 formats = []
1945 for url_data_str in encoded_url_map.split(','):
1946 url_data = compat_parse_qs(url_data_str)
1947 if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
1948 continue
1949 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1950 # Unsupported FORMAT_STREAM_TYPE_OTF
1951 if stream_type == 3:
1952 continue
1953 format_id = url_data['itag'][0]
1954 url = url_data['url'][0]
1955
1956 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1957 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1958 jsplayer_url_json = self._search_regex(
1959 ASSETS_RE,
1960 embed_webpage if age_gate else video_webpage,
1961 'JS player URL (1)', default=None)
1962 if not jsplayer_url_json and not age_gate:
1963 # We need the embed website after all
1964 if embed_webpage is None:
1965 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1966 embed_webpage = self._download_webpage(
1967 embed_url, video_id, 'Downloading embed webpage')
1968 jsplayer_url_json = self._search_regex(
1969 ASSETS_RE, embed_webpage, 'JS player URL')
1970
1971 player_url = json.loads(jsplayer_url_json)
1972 if player_url is None:
1973 player_url_json = self._search_regex(
1974 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1975 video_webpage, 'age gate player URL')
1976 player_url = json.loads(player_url_json)
1977
1978 if 'sig' in url_data:
1979 url += '&signature=' + url_data['sig'][0]
1980 elif 's' in url_data:
1981 encrypted_sig = url_data['s'][0]
1982
1983 if self._downloader.params.get('verbose'):
1984 if player_url is None:
1985 player_version = 'unknown'
1986 player_desc = 'unknown'
1987 else:
1988 if player_url.endswith('swf'):
1989 player_version = self._search_regex(
1990 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1991 'flash player', fatal=False)
1992 player_desc = 'flash player %s' % player_version
1993 else:
1994 player_version = self._search_regex(
1995 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1996 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1997 player_url,
1998 'html5 player', fatal=False)
1999 player_desc = 'html5 player %s' % player_version
2000
2001 parts_sizes = self._signature_cache_id(encrypted_sig)
2002 self.to_screen('{%s} signature length %s, %s' %
2003 (format_id, parts_sizes, player_desc))
2004
2005 signature = self._decrypt_signature(
2006 encrypted_sig, video_id, player_url, age_gate)
2007 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2008 url += '&%s=%s' % (sp, signature)
2009 if 'ratebypass' not in url:
2010 url += '&ratebypass=yes'
2011
2012 dct = {
2013 'format_id': format_id,
2014 'url': url,
2015 'player_url': player_url,
2016 }
2017 if format_id in self._formats:
2018 dct.update(self._formats[format_id])
2019 if format_id in formats_spec:
2020 dct.update(formats_spec[format_id])
2021
2022 # Some itags are not included in DASH manifest thus corresponding formats will
2023 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2024 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2025 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2026 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2027
2028 filesize = int_or_none(url_data.get(
2029 'clen', [None])[0]) or _extract_filesize(url)
2030
2031 quality = url_data.get('quality', [None])[0]
2032
2033 more_fields = {
2034 'filesize': filesize,
2035 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2036 'width': width,
2037 'height': height,
2038 'fps': int_or_none(url_data.get('fps', [None])[0]),
2039 'format_note': url_data.get('quality_label', [None])[0] or quality,
2040 'quality': q(quality),
2041 }
2042 for key, value in more_fields.items():
2043 if value:
2044 dct[key] = value
2045 type_ = url_data.get('type', [None])[0]
2046 if type_:
2047 type_split = type_.split(';')
2048 kind_ext = type_split[0].split('/')
2049 if len(kind_ext) == 2:
2050 kind, _ = kind_ext
2051 dct['ext'] = mimetype2ext(type_split[0])
2052 if kind in ('audio', 'video'):
2053 codecs = None
2054 for mobj in re.finditer(
2055 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2056 if mobj.group('key') == 'codecs':
2057 codecs = mobj.group('val')
2058 break
2059 if codecs:
2060 dct.update(parse_codecs(codecs))
2061 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2062 dct['downloader_options'] = {
2063 # Youtube throttles chunks >~10M
2064 'http_chunk_size': 10485760,
2065 }
2066 formats.append(dct)
2067 else:
2068 manifest_url = (
2069 url_or_none(try_get(
2070 player_response,
2071 lambda x: x['streamingData']['hlsManifestUrl'],
2072 compat_str))
2073 or url_or_none(try_get(
2074 video_info, lambda x: x['hlsvp'][0], compat_str)))
2075 if manifest_url:
2076 formats = []
2077 m3u8_formats = self._extract_m3u8_formats(
2078 manifest_url, video_id, 'mp4', fatal=False)
2079 for a_format in m3u8_formats:
2080 itag = self._search_regex(
2081 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2082 if itag:
2083 a_format['format_id'] = itag
2084 if itag in self._formats:
2085 dct = self._formats[itag].copy()
2086 dct.update(a_format)
2087 a_format = dct
2088 a_format['player_url'] = player_url
2089 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2090 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2091 formats.append(a_format)
2092 else:
2093 error_message = clean_html(video_info.get('reason', [None])[0])
2094 if not error_message:
2095 error_message = extract_unavailable_message()
2096 if error_message:
2097 raise ExtractorError(error_message, expected=True)
2098 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2099
2100 # uploader
2101 video_uploader = try_get(
2102 video_info, lambda x: x['author'][0],
2103 compat_str) or str_or_none(video_details.get('author'))
2104 if video_uploader:
2105 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2106 else:
2107 self._downloader.report_warning('unable to extract uploader name')
2108
2109 # uploader_id
2110 video_uploader_id = None
2111 video_uploader_url = None
2112 mobj = re.search(
2113 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2114 video_webpage)
2115 if mobj is not None:
2116 video_uploader_id = mobj.group('uploader_id')
2117 video_uploader_url = mobj.group('uploader_url')
2118 else:
2119 self._downloader.report_warning('unable to extract uploader nickname')
2120
2121 channel_id = (
2122 str_or_none(video_details.get('channelId'))
2123 or self._html_search_meta(
2124 'channelId', video_webpage, 'channel id', default=None)
2125 or self._search_regex(
2126 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2127 video_webpage, 'channel id', default=None, group='id'))
2128 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2129
2130 # thumbnail image
2131 # We try first to get a high quality image:
2132 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2133 video_webpage, re.DOTALL)
2134 if m_thumb is not None:
2135 video_thumbnail = m_thumb.group(1)
2136 elif 'thumbnail_url' not in video_info:
2137 self._downloader.report_warning('unable to extract video thumbnail')
2138 video_thumbnail = None
2139 else: # don't panic if we can't find it
2140 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2141
2142 # upload date
2143 upload_date = self._html_search_meta(
2144 'datePublished', video_webpage, 'upload date', default=None)
2145 if not upload_date:
2146 upload_date = self._search_regex(
2147 [r'(?s)id="eow-date.*?>(.*?)</span>',
2148 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2149 video_webpage, 'upload date', default=None)
2150 upload_date = unified_strdate(upload_date)
2151
2152 video_license = self._html_search_regex(
2153 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2154 video_webpage, 'license', default=None)
2155
2156 m_music = re.search(
2157 r'''(?x)
2158 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2159 <ul[^>]*>\s*
2160 <li>(?P<title>.+?)
2161 by (?P<creator>.+?)
2162 (?:
2163 \(.+?\)|
2164 <a[^>]*
2165 (?:
2166 \bhref=["\']/red[^>]*>| # drop possible
2167 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2168 )
2169 .*?
2170 )?</li
2171 ''',
2172 video_webpage)
2173 if m_music:
2174 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2175 video_creator = clean_html(m_music.group('creator'))
2176 else:
2177 video_alt_title = video_creator = None
2178
2179 def extract_meta(field):
2180 return self._html_search_regex(
2181 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2182 video_webpage, field, default=None)
2183
2184 track = extract_meta('Song')
2185 artist = extract_meta('Artist')
2186 album = extract_meta('Album')
2187
2188 # Youtube Music Auto-generated description
2189 release_date = release_year = None
2190 if video_description:
2191 mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^Ā·]+)Ā·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?ā„—\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?', video_description)
2192 if mobj:
2193 if not track:
2194 track = mobj.group('track').strip()
2195 if not artist:
2196 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('Ā·'))
2197 if not album:
2198 album = mobj.group('album'.strip())
2199 release_year = mobj.group('release_year')
2200 release_date = mobj.group('release_date')
2201 if release_date:
2202 release_date = release_date.replace('-', '')
2203 if not release_year:
2204 release_year = int(release_date[:4])
2205 if release_year:
2206 release_year = int(release_year)
2207
2208 m_episode = re.search(
2209 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2210 video_webpage)
2211 if m_episode:
2212 series = unescapeHTML(m_episode.group('series'))
2213 season_number = int(m_episode.group('season'))
2214 episode_number = int(m_episode.group('episode'))
2215 else:
2216 series = season_number = episode_number = None
2217
2218 m_cat_container = self._search_regex(
2219 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2220 video_webpage, 'categories', default=None)
2221 if m_cat_container:
2222 category = self._html_search_regex(
2223 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2224 default=None)
2225 video_categories = None if category is None else [category]
2226 else:
2227 video_categories = None
2228
2229 video_tags = [
2230 unescapeHTML(m.group('content'))
2231 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2232
2233 def _extract_count(count_name):
2234 return str_to_int(self._search_regex(
2235 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2236 % re.escape(count_name),
2237 video_webpage, count_name, default=None))
2238
2239 like_count = _extract_count('like')
2240 dislike_count = _extract_count('dislike')
2241
2242 if view_count is None:
2243 view_count = str_to_int(self._search_regex(
2244 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2245 'view count', default=None))
2246
2247 average_rating = (
2248 float_or_none(video_details.get('averageRating'))
2249 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2250
2251 # subtitles
2252 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2253 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2254
2255 video_duration = try_get(
2256 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2257 if not video_duration:
2258 video_duration = int_or_none(video_details.get('lengthSeconds'))
2259 if not video_duration:
2260 video_duration = parse_duration(self._html_search_meta(
2261 'duration', video_webpage, 'video duration'))
2262
2263 # annotations
2264 video_annotations = None
2265 if self._downloader.params.get('writeannotations', False):
2266 video_annotations = self._extract_annotations(video_id)
2267
2268 chapters = self._extract_chapters(description_original, video_duration)
2269
2270 # Look for the DASH manifest
2271 if self._downloader.params.get('youtube_include_dash_manifest', True):
2272 dash_mpd_fatal = True
2273 for mpd_url in dash_mpds:
2274 dash_formats = {}
2275 try:
2276 def decrypt_sig(mobj):
2277 s = mobj.group(1)
2278 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2279 return '/signature/%s' % dec_s
2280
2281 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2282
2283 for df in self._extract_mpd_formats(
2284 mpd_url, video_id, fatal=dash_mpd_fatal,
2285 formats_dict=self._formats):
2286 if not df.get('filesize'):
2287 df['filesize'] = _extract_filesize(df['url'])
2288 # Do not overwrite DASH format found in some previous DASH manifest
2289 if df['format_id'] not in dash_formats:
2290 dash_formats[df['format_id']] = df
2291 # Additional DASH manifests may end up in HTTP Error 403 therefore
2292 # allow them to fail without bug report message if we already have
2293 # some DASH manifest succeeded. This is temporary workaround to reduce
2294 # burst of bug reports until we figure out the reason and whether it
2295 # can be fixed at all.
2296 dash_mpd_fatal = False
2297 except (ExtractorError, KeyError) as e:
2298 self.report_warning(
2299 'Skipping DASH manifest: %r' % e, video_id)
2300 if dash_formats:
2301 # Remove the formats we found through non-DASH, they
2302 # contain less info and it can be wrong, because we use
2303 # fixed values (for example the resolution). See
2304 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2305 # example.
2306 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2307 formats.extend(dash_formats.values())
2308
2309 # Check for malformed aspect ratio
2310 stretched_m = re.search(
2311 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2312 video_webpage)
2313 if stretched_m:
2314 w = float(stretched_m.group('w'))
2315 h = float(stretched_m.group('h'))
2316 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2317 # We will only process correct ratios.
2318 if w > 0 and h > 0:
2319 ratio = w / h
2320 for f in formats:
2321 if f.get('vcodec') != 'none':
2322 f['stretched_ratio'] = ratio
2323
2324 if not formats:
2325 token = extract_token(video_info)
2326 if not token:
2327 if 'reason' in video_info:
2328 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2329 regions_allowed = self._html_search_meta(
2330 'regionsAllowed', video_webpage, default=None)
2331 countries = regions_allowed.split(',') if regions_allowed else None
2332 self.raise_geo_restricted(
2333 msg=video_info['reason'][0], countries=countries)
2334 reason = video_info['reason'][0]
2335 if 'Invalid parameters' in reason:
2336 unavailable_message = extract_unavailable_message()
2337 if unavailable_message:
2338 reason = unavailable_message
2339 raise ExtractorError(
2340 'YouTube said: %s' % reason,
2341 expected=True, video_id=video_id)
2342 else:
2343 raise ExtractorError(
2344 '"token" parameter not in video info for unknown reason',
2345 video_id=video_id)
2346
2347 if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2348 raise ExtractorError('This video is DRM protected.', expected=True)
2349
2350 self._sort_formats(formats)
2351
2352 self.mark_watched(video_id, video_info, player_response)
2353
2354 return {
2355 'id': video_id,
2356 'uploader': video_uploader,
2357 'uploader_id': video_uploader_id,
2358 'uploader_url': video_uploader_url,
2359 'channel_id': channel_id,
2360 'channel_url': channel_url,
2361 'upload_date': upload_date,
2362 'license': video_license,
2363 'creator': video_creator or artist,
2364 'title': video_title,
2365 'alt_title': video_alt_title or track,
2366 'thumbnail': video_thumbnail,
2367 'description': video_description,
2368 'categories': video_categories,
2369 'tags': video_tags,
2370 'subtitles': video_subtitles,
2371 'automatic_captions': automatic_captions,
2372 'duration': video_duration,
2373 'age_limit': 18 if age_gate else 0,
2374 'annotations': video_annotations,
2375 'chapters': chapters,
2376 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2377 'view_count': view_count,
2378 'like_count': like_count,
2379 'dislike_count': dislike_count,
2380 'average_rating': average_rating,
2381 'formats': formats,
2382 'is_live': is_live,
2383 'start_time': start_time,
2384 'end_time': end_time,
2385 'series': series,
2386 'season_number': season_number,
2387 'episode_number': episode_number,
2388 'track': track,
2389 'artist': artist,
2390 'album': album,
2391 'release_date': release_date,
2392 'release_year': release_year,
2393 }
2394
2395
2396 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2397 IE_DESC = 'YouTube.com playlists'
2398 _VALID_URL = r"""(?x)(?:
2399 (?:https?://)?
2400 (?:\w+\.)?
2401 (?:
2402 (?:
2403 youtube\.com|
2404 invidio\.us
2405 )
2406 /
2407 (?:
2408 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2409 \? (?:.*?[&;])*? (?:p|a|list)=
2410 | p/
2411 )|
2412 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2413 )
2414 (
2415 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2416 # Top tracks, they can also include dots
2417 |(?:MC)[\w\.]*
2418 )
2419 .*
2420 |
2421 (%(playlist_id)s)
2422 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2423 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2424 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2425 IE_NAME = 'youtube:playlist'
2426 _TESTS = [{
2427 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2428 'info_dict': {
2429 'title': 'ytdl test PL',
2430 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2431 },
2432 'playlist_count': 3,
2433 }, {
2434 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2435 'info_dict': {
2436 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2437 'title': 'YDL_Empty_List',
2438 },
2439 'playlist_count': 0,
2440 'skip': 'This playlist is private',
2441 }, {
2442 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2443 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2444 'info_dict': {
2445 'title': '29C3: Not my department',
2446 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2447 },
2448 'playlist_count': 95,
2449 }, {
2450 'note': 'issue #673',
2451 'url': 'PLBB231211A4F62143',
2452 'info_dict': {
2453 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2454 'id': 'PLBB231211A4F62143',
2455 },
2456 'playlist_mincount': 26,
2457 }, {
2458 'note': 'Large playlist',
2459 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2460 'info_dict': {
2461 'title': 'Uploads from Cauchemar',
2462 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2463 },
2464 'playlist_mincount': 799,
2465 }, {
2466 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2467 'info_dict': {
2468 'title': 'YDL_safe_search',
2469 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2470 },
2471 'playlist_count': 2,
2472 'skip': 'This playlist is private',
2473 }, {
2474 'note': 'embedded',
2475 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2476 'playlist_count': 4,
2477 'info_dict': {
2478 'title': 'JODA15',
2479 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2480 }
2481 }, {
2482 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2483 'playlist_mincount': 485,
2484 'info_dict': {
2485 'title': '2017 čÆčŖžęœ€ę–°å–®ę›² (2/24ꛓꖰ)',
2486 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2487 }
2488 }, {
2489 'note': 'Embedded SWF player',
2490 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2491 'playlist_count': 4,
2492 'info_dict': {
2493 'title': 'JODA7',
2494 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2495 }
2496 }, {
2497 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2498 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2499 'info_dict': {
2500 'title': 'Uploads from Interstellar Movie',
2501 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2502 },
2503 'playlist_mincount': 21,
2504 }, {
2505 # Playlist URL that does not actually serve a playlist
2506 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2507 'info_dict': {
2508 'id': 'FqZTN594JQw',
2509 'ext': 'webm',
2510 'title': "Smiley's People 01 detective, Adventure Series, Action",
2511 'uploader': 'STREEM',
2512 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2513 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2514 'upload_date': '20150526',
2515 'license': 'Standard YouTube License',
2516 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2517 'categories': ['People & Blogs'],
2518 'tags': list,
2519 'view_count': int,
2520 'like_count': int,
2521 'dislike_count': int,
2522 },
2523 'params': {
2524 'skip_download': True,
2525 },
2526 'add_ie': [YoutubeIE.ie_key()],
2527 }, {
2528 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2529 'info_dict': {
2530 'id': 'yeWKywCrFtk',
2531 'ext': 'mp4',
2532 'title': 'Small Scale Baler and Braiding Rugs',
2533 'uploader': 'Backus-Page House Museum',
2534 'uploader_id': 'backuspagemuseum',
2535 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2536 'upload_date': '20161008',
2537 'license': 'Standard YouTube License',
2538 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2539 'categories': ['Nonprofits & Activism'],
2540 'tags': list,
2541 'like_count': int,
2542 'dislike_count': int,
2543 },
2544 'params': {
2545 'noplaylist': True,
2546 'skip_download': True,
2547 },
2548 }, {
2549 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2550 'only_matching': True,
2551 }, {
2552 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2553 'only_matching': True,
2554 }, {
2555 # music album playlist
2556 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2557 'only_matching': True,
2558 }, {
2559 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2560 'only_matching': True,
2561 }]
2562
2563 def _real_initialize(self):
2564 self._login()
2565
2566 def _extract_mix(self, playlist_id):
2567 # The mixes are generated from a single video
2568 # the id of the playlist is just 'RD' + video_id
2569 ids = []
2570 last_id = playlist_id[-11:]
2571 for n in itertools.count(1):
2572 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2573 webpage = self._download_webpage(
2574 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2575 new_ids = orderedSet(re.findall(
2576 r'''(?xs)data-video-username=".*?".*?
2577 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2578 webpage))
2579 # Fetch new pages until all the videos are repeated, it seems that
2580 # there are always 51 unique videos.
2581 new_ids = [_id for _id in new_ids if _id not in ids]
2582 if not new_ids:
2583 break
2584 ids.extend(new_ids)
2585 last_id = ids[-1]
2586
2587 url_results = self._ids_to_results(ids)
2588
2589 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2590 title_span = (
2591 search_title('playlist-title')
2592 or search_title('title long-title')
2593 or search_title('title'))
2594 title = clean_html(title_span)
2595
2596 return self.playlist_result(url_results, playlist_id, title)
2597
2598 def _extract_playlist(self, playlist_id):
2599 url = self._TEMPLATE_URL % playlist_id
2600 page = self._download_webpage(url, playlist_id)
2601
2602 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2603 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2604 match = match.strip()
2605 # Check if the playlist exists or is private
2606 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2607 if mobj:
2608 reason = mobj.group('reason')
2609 message = 'This playlist %s' % reason
2610 if 'private' in reason:
2611 message += ', use --username or --netrc to access it'
2612 message += '.'
2613 raise ExtractorError(message, expected=True)
2614 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2615 raise ExtractorError(
2616 'Invalid parameters. Maybe URL is incorrect.',
2617 expected=True)
2618 elif re.match(r'[^<]*Choose your language[^<]*', match):
2619 continue
2620 else:
2621 self.report_warning('Youtube gives an alert message: ' + match)
2622
2623 playlist_title = self._html_search_regex(
2624 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2625 page, 'title', default=None)
2626
2627 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2628 uploader = self._search_regex(
2629 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2630 page, 'uploader', default=None)
2631 mobj = re.search(
2632 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2633 page)
2634 if mobj:
2635 uploader_id = mobj.group('uploader_id')
2636 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2637 else:
2638 uploader_id = uploader_url = None
2639
2640 has_videos = True
2641
2642 if not playlist_title:
2643 try:
2644 # Some playlist URLs don't actually serve a playlist (e.g.
2645 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2646 next(self._entries(page, playlist_id))
2647 except StopIteration:
2648 has_videos = False
2649
2650 playlist = self.playlist_result(
2651 self._entries(page, playlist_id), playlist_id, playlist_title)
2652 playlist.update({
2653 'uploader': uploader,
2654 'uploader_id': uploader_id,
2655 'uploader_url': uploader_url,
2656 })
2657
2658 return has_videos, playlist
2659
2660 def _check_download_just_video(self, url, playlist_id):
2661 # Check if it's a video-specific URL
2662 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2663 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2664 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2665 'video id', default=None)
2666 if video_id:
2667 if self._downloader.params.get('noplaylist'):
2668 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2669 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2670 else:
2671 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2672 return video_id, None
2673 return None, None
2674
2675 def _real_extract(self, url):
2676 # Extract playlist id
2677 mobj = re.match(self._VALID_URL, url)
2678 if mobj is None:
2679 raise ExtractorError('Invalid URL: %s' % url)
2680 playlist_id = mobj.group(1) or mobj.group(2)
2681
2682 video_id, video = self._check_download_just_video(url, playlist_id)
2683 if video:
2684 return video
2685
2686 if playlist_id.startswith(('RD', 'UL', 'PU')):
2687 # Mixes require a custom extraction process
2688 return self._extract_mix(playlist_id)
2689
2690 has_videos, playlist = self._extract_playlist(playlist_id)
2691 if has_videos or not video_id:
2692 return playlist
2693
2694 # Some playlist URLs don't actually serve a playlist (see
2695 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2696 # Fallback to plain video extraction if there is a video id
2697 # along with playlist id.
2698 return self.url_result(video_id, 'Youtube', video_id=video_id)
2699
2700
2701 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2702 IE_DESC = 'YouTube.com channels'
2703 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2704 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2705 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2706 IE_NAME = 'youtube:channel'
2707 _TESTS = [{
2708 'note': 'paginated channel',
2709 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2710 'playlist_mincount': 91,
2711 'info_dict': {
2712 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2713 'title': 'Uploads from lex will',
2714 }
2715 }, {
2716 'note': 'Age restricted channel',
2717 # from https://www.youtube.com/user/DeusExOfficial
2718 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2719 'playlist_mincount': 64,
2720 'info_dict': {
2721 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2722 'title': 'Uploads from Deus Ex',
2723 },
2724 }, {
2725 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2726 'only_matching': True,
2727 }]
2728
2729 @classmethod
2730 def suitable(cls, url):
2731 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2732 else super(YoutubeChannelIE, cls).suitable(url))
2733
2734 def _build_template_url(self, url, channel_id):
2735 return self._TEMPLATE_URL % channel_id
2736
2737 def _real_extract(self, url):
2738 channel_id = self._match_id(url)
2739
2740 url = self._build_template_url(url, channel_id)
2741
2742 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2743 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2744 # otherwise fallback on channel by page extraction
2745 channel_page = self._download_webpage(
2746 url + '?view=57', channel_id,
2747 'Downloading channel page', fatal=False)
2748 if channel_page is False:
2749 channel_playlist_id = False
2750 else:
2751 channel_playlist_id = self._html_search_meta(
2752 'channelId', channel_page, 'channel id', default=None)
2753 if not channel_playlist_id:
2754 channel_url = self._html_search_meta(
2755 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2756 channel_page, 'channel url', default=None)
2757 if channel_url:
2758 channel_playlist_id = self._search_regex(
2759 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2760 channel_url, 'channel id', default=None)
2761 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2762 playlist_id = 'UU' + channel_playlist_id[2:]
2763 return self.url_result(
2764 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2765
2766 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2767 autogenerated = re.search(r'''(?x)
2768 class="[^"]*?(?:
2769 channel-header-autogenerated-label|
2770 yt-channel-title-autogenerated
2771 )[^"]*"''', channel_page) is not None
2772
2773 if autogenerated:
2774 # The videos are contained in a single page
2775 # the ajax pages can't be used, they are empty
2776 entries = [
2777 self.url_result(
2778 video_id, 'Youtube', video_id=video_id,
2779 video_title=video_title)
2780 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2781 return self.playlist_result(entries, channel_id)
2782
2783 try:
2784 next(self._entries(channel_page, channel_id))
2785 except StopIteration:
2786 alert_message = self._html_search_regex(
2787 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2788 channel_page, 'alert', default=None, group='alert')
2789 if alert_message:
2790 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2791
2792 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2793
2794
2795 class YoutubeUserIE(YoutubeChannelIE):
2796 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2797 _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_-]+)'
2798 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2799 IE_NAME = 'youtube:user'
2800
2801 _TESTS = [{
2802 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2803 'playlist_mincount': 320,
2804 'info_dict': {
2805 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2806 'title': 'Uploads from The Linux Foundation',
2807 }
2808 }, {
2809 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2810 # but not https://www.youtube.com/user/12minuteathlete/videos
2811 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2812 'playlist_mincount': 249,
2813 'info_dict': {
2814 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2815 'title': 'Uploads from 12 Minute Athlete',
2816 }
2817 }, {
2818 'url': 'ytuser:phihag',
2819 'only_matching': True,
2820 }, {
2821 'url': 'https://www.youtube.com/c/gametrailers',
2822 'only_matching': True,
2823 }, {
2824 'url': 'https://www.youtube.com/gametrailers',
2825 'only_matching': True,
2826 }, {
2827 # This channel is not available, geo restricted to JP
2828 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2829 'only_matching': True,
2830 }]
2831
2832 @classmethod
2833 def suitable(cls, url):
2834 # Don't return True if the url can be extracted with other youtube
2835 # extractor, the regex would is too permissive and it would match.
2836 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2837 if any(ie.suitable(url) for ie in other_yt_ies):
2838 return False
2839 else:
2840 return super(YoutubeUserIE, cls).suitable(url)
2841
2842 def _build_template_url(self, url, channel_id):
2843 mobj = re.match(self._VALID_URL, url)
2844 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2845
2846
2847 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2848 IE_DESC = 'YouTube.com live streams'
2849 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2850 IE_NAME = 'youtube:live'
2851
2852 _TESTS = [{
2853 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2854 'info_dict': {
2855 'id': 'a48o2S1cPoo',
2856 'ext': 'mp4',
2857 'title': 'The Young Turks - Live Main Show',
2858 'uploader': 'The Young Turks',
2859 'uploader_id': 'TheYoungTurks',
2860 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2861 'upload_date': '20150715',
2862 'license': 'Standard YouTube License',
2863 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2864 'categories': ['News & Politics'],
2865 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2866 'like_count': int,
2867 'dislike_count': int,
2868 },
2869 'params': {
2870 'skip_download': True,
2871 },
2872 }, {
2873 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2874 'only_matching': True,
2875 }, {
2876 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2877 'only_matching': True,
2878 }, {
2879 'url': 'https://www.youtube.com/TheYoungTurks/live',
2880 'only_matching': True,
2881 }]
2882
2883 def _real_extract(self, url):
2884 mobj = re.match(self._VALID_URL, url)
2885 channel_id = mobj.group('id')
2886 base_url = mobj.group('base_url')
2887 webpage = self._download_webpage(url, channel_id, fatal=False)
2888 if webpage:
2889 page_type = self._og_search_property(
2890 'type', webpage, 'page type', default='')
2891 video_id = self._html_search_meta(
2892 'videoId', webpage, 'video id', default=None)
2893 if page_type.startswith('video') and video_id and re.match(
2894 r'^[0-9A-Za-z_-]{11}$', video_id):
2895 return self.url_result(video_id, YoutubeIE.ie_key())
2896 return self.url_result(base_url)
2897
2898
2899 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2900 IE_DESC = 'YouTube.com user/channel playlists'
2901 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2902 IE_NAME = 'youtube:playlists'
2903
2904 _TESTS = [{
2905 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2906 'playlist_mincount': 4,
2907 'info_dict': {
2908 'id': 'ThirstForScience',
2909 'title': 'Thirst for Science',
2910 },
2911 }, {
2912 # with "Load more" button
2913 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2914 'playlist_mincount': 70,
2915 'info_dict': {
2916 'id': 'igorkle1',
2917 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
2918 },
2919 }, {
2920 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2921 'playlist_mincount': 17,
2922 'info_dict': {
2923 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2924 'title': 'Chem Player',
2925 },
2926 }]
2927
2928
2929 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2930 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2931
2932
2933 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2934 IE_DESC = 'YouTube.com searches'
2935 # there doesn't appear to be a real limit, for example if you search for
2936 # 'python' you get more than 8.000.000 results
2937 _MAX_RESULTS = float('inf')
2938 IE_NAME = 'youtube:search'
2939 _SEARCH_KEY = 'ytsearch'
2940 _EXTRA_QUERY_ARGS = {}
2941 _TESTS = []
2942
2943 def _get_n_results(self, query, n):
2944 """Get a specified number of results for a query"""
2945
2946 videos = []
2947 limit = n
2948
2949 url_query = {
2950 'search_query': query.encode('utf-8'),
2951 }
2952 url_query.update(self._EXTRA_QUERY_ARGS)
2953 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2954
2955 for pagenum in itertools.count(1):
2956 data = self._download_json(
2957 result_url, video_id='query "%s"' % query,
2958 note='Downloading page %s' % pagenum,
2959 errnote='Unable to download API page',
2960 query={'spf': 'navigate'})
2961 html_content = data[1]['body']['content']
2962
2963 if 'class="search-message' in html_content:
2964 raise ExtractorError(
2965 '[youtube] No video results', expected=True)
2966
2967 new_videos = list(self._process_page(html_content))
2968 videos += new_videos
2969 if not new_videos or len(videos) > limit:
2970 break
2971 next_link = self._html_search_regex(
2972 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2973 html_content, 'next link', default=None)
2974 if next_link is None:
2975 break
2976 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2977
2978 if len(videos) > n:
2979 videos = videos[:n]
2980 return self.playlist_result(videos, query)
2981
2982
2983 class YoutubeSearchDateIE(YoutubeSearchIE):
2984 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2985 _SEARCH_KEY = 'ytsearchdate'
2986 IE_DESC = 'YouTube.com searches, newest videos first'
2987 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2988
2989
2990 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2991 IE_DESC = 'YouTube.com search URLs'
2992 IE_NAME = 'youtube:search_url'
2993 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2994 _TESTS = [{
2995 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2996 'playlist_mincount': 5,
2997 'info_dict': {
2998 'title': 'youtube-dl test video',
2999 }
3000 }, {
3001 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3002 'only_matching': True,
3003 }]
3004
3005 def _real_extract(self, url):
3006 mobj = re.match(self._VALID_URL, url)
3007 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3008 webpage = self._download_webpage(url, query)
3009 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3010
3011
3012 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3013 IE_DESC = 'YouTube.com (multi-season) shows'
3014 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3015 IE_NAME = 'youtube:show'
3016 _TESTS = [{
3017 'url': 'https://www.youtube.com/show/airdisasters',
3018 'playlist_mincount': 5,
3019 'info_dict': {
3020 'id': 'airdisasters',
3021 'title': 'Air Disasters',
3022 }
3023 }]
3024
3025 def _real_extract(self, url):
3026 playlist_id = self._match_id(url)
3027 return super(YoutubeShowIE, self)._real_extract(
3028 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3029
3030
3031 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3032 """
3033 Base class for feed extractors
3034 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3035 """
3036 _LOGIN_REQUIRED = True
3037
3038 @property
3039 def IE_NAME(self):
3040 return 'youtube:%s' % self._FEED_NAME
3041
3042 def _real_initialize(self):
3043 self._login()
3044
3045 def _entries(self, page):
3046 # The extraction process is the same as for playlists, but the regex
3047 # for the video ids doesn't contain an index
3048 ids = []
3049 more_widget_html = content_html = page
3050 for page_num in itertools.count(1):
3051 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3052
3053 # 'recommended' feed has infinite 'load more' and each new portion spins
3054 # the same videos in (sometimes) slightly different order, so we'll check
3055 # for unicity and break when portion has no new videos
3056 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3057 if not new_ids:
3058 break
3059
3060 ids.extend(new_ids)
3061
3062 for entry in self._ids_to_results(new_ids):
3063 yield entry
3064
3065 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3066 if not mobj:
3067 break
3068
3069 more = self._download_json(
3070 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3071 'Downloading page #%s' % page_num,
3072 transform_source=uppercase_escape)
3073 content_html = more['content_html']
3074 more_widget_html = more['load_more_widget_html']
3075
3076 def _real_extract(self, url):
3077 page = self._download_webpage(
3078 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3079 self._PLAYLIST_TITLE)
3080 return self.playlist_result(
3081 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3082
3083
3084 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3085 IE_NAME = 'youtube:watchlater'
3086 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3087 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3088
3089 _TESTS = [{
3090 'url': 'https://www.youtube.com/playlist?list=WL',
3091 'only_matching': True,
3092 }, {
3093 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3094 'only_matching': True,
3095 }]
3096
3097 def _real_extract(self, url):
3098 _, video = self._check_download_just_video(url, 'WL')
3099 if video:
3100 return video
3101 _, playlist = self._extract_playlist('WL')
3102 return playlist
3103
3104
3105 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3106 IE_NAME = 'youtube:favorites'
3107 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3108 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3109 _LOGIN_REQUIRED = True
3110
3111 def _real_extract(self, url):
3112 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3113 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3114 return self.url_result(playlist_id, 'YoutubePlaylist')
3115
3116
3117 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3118 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3119 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3120 _FEED_NAME = 'recommended'
3121 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3122
3123
3124 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3125 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3126 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3127 _FEED_NAME = 'subscriptions'
3128 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3129
3130
3131 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3132 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3133 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3134 _FEED_NAME = 'history'
3135 _PLAYLIST_TITLE = 'Youtube History'
3136
3137
3138 class YoutubeTruncatedURLIE(InfoExtractor):
3139 IE_NAME = 'youtube:truncated_url'
3140 IE_DESC = False # Do not list
3141 _VALID_URL = r'''(?x)
3142 (?:https?://)?
3143 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3144 (?:watch\?(?:
3145 feature=[a-z_]+|
3146 annotation_id=annotation_[^&]+|
3147 x-yt-cl=[0-9]+|
3148 hl=[^&]*|
3149 t=[0-9]+
3150 )?
3151 |
3152 attribution_link\?a=[^&]+
3153 )
3154 $
3155 '''
3156
3157 _TESTS = [{
3158 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3159 'only_matching': True,
3160 }, {
3161 'url': 'https://www.youtube.com/watch?',
3162 'only_matching': True,
3163 }, {
3164 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3165 'only_matching': True,
3166 }, {
3167 'url': 'https://www.youtube.com/watch?feature=foo',
3168 'only_matching': True,
3169 }, {
3170 'url': 'https://www.youtube.com/watch?hl=en-GB',
3171 'only_matching': True,
3172 }, {
3173 'url': 'https://www.youtube.com/watch?t=2372',
3174 'only_matching': True,
3175 }]
3176
3177 def _real_extract(self, url):
3178 raise ExtractorError(
3179 'Did you forget to quote the URL? Remember that & is a meta '
3180 'character in most shells, so you want to put the URL in quotes, '
3181 'like youtube-dl '
3182 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3183 ' or simply youtube-dl BaW_jenozKc .',
3184 expected=True)
3185
3186
3187 class YoutubeTruncatedIDIE(InfoExtractor):
3188 IE_NAME = 'youtube:truncated_id'
3189 IE_DESC = False # Do not list
3190 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3191
3192 _TESTS = [{
3193 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3194 'only_matching': True,
3195 }]
3196
3197 def _real_extract(self, url):
3198 video_id = self._match_id(url)
3199 raise ExtractorError(
3200 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3201 expected=True)