]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
New upstream version 2019.09.01
[youtubedl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_HTTPError,
20 compat_kwargs,
21 compat_parse_qs,
22 compat_urllib_parse_unquote,
23 compat_urllib_parse_unquote_plus,
24 compat_urllib_parse_urlencode,
25 compat_urllib_parse_urlparse,
26 compat_urlparse,
27 compat_str,
28 )
29 from ..utils import (
30 bool_or_none,
31 clean_html,
32 dict_get,
33 error_to_compat_str,
34 extract_attributes,
35 ExtractorError,
36 float_or_none,
37 get_element_by_attribute,
38 get_element_by_id,
39 int_or_none,
40 mimetype2ext,
41 orderedSet,
42 parse_codecs,
43 parse_duration,
44 qualities,
45 remove_quotes,
46 remove_start,
47 smuggle_url,
48 str_or_none,
49 str_to_int,
50 try_get,
51 unescapeHTML,
52 unified_strdate,
53 unsmuggle_url,
54 uppercase_escape,
55 url_or_none,
56 urlencode_postdata,
57 )
58
59
60 class YoutubeBaseInfoExtractor(InfoExtractor):
61 """Provide base functions for Youtube extractors"""
62 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
63 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
64
65 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
66 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
67 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
68
69 _NETRC_MACHINE = 'youtube'
70 # If True it will raise an error if no login info is provided
71 _LOGIN_REQUIRED = False
72
73 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
74
75 def _set_language(self):
76 self._set_cookie(
77 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
78 # YouTube sets the expire time to about two months
79 expire_time=time.time() + 2 * 30 * 24 * 3600)
80
81 def _ids_to_results(self, ids):
82 return [
83 self.url_result(vid_id, 'Youtube', video_id=vid_id)
84 for vid_id in ids]
85
86 def _login(self):
87 """
88 Attempt to log in to YouTube.
89 True is returned if successful or skipped.
90 False is returned if login failed.
91
92 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
93 """
94 username, password = self._get_login_info()
95 # No authentication to be performed
96 if username is None:
97 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
98 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
99 return True
100
101 login_page = self._download_webpage(
102 self._LOGIN_URL, None,
103 note='Downloading login page',
104 errnote='unable to fetch login page', fatal=False)
105 if login_page is False:
106 return
107
108 login_form = self._hidden_inputs(login_page)
109
110 def req(url, f_req, note, errnote):
111 data = login_form.copy()
112 data.update({
113 'pstMsg': 1,
114 'checkConnection': 'youtube',
115 'checkedDomains': 'youtube',
116 'hl': 'en',
117 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
118 'f.req': json.dumps(f_req),
119 'flowName': 'GlifWebSignIn',
120 'flowEntry': 'ServiceLogin',
121 # TODO: reverse actual botguard identifier generation algo
122 'bgRequest': '["identifier",""]',
123 })
124 return self._download_json(
125 url, None, note=note, errnote=errnote,
126 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
127 fatal=False,
128 data=urlencode_postdata(data), headers={
129 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
130 'Google-Accounts-XSRF': 1,
131 })
132
133 def warn(message):
134 self._downloader.report_warning(message)
135
136 lookup_req = [
137 username,
138 None, [], None, 'US', None, None, 2, False, True,
139 [
140 None, None,
141 [2, 1, None, 1,
142 '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',
143 None, [], 4],
144 1, [None, None, []], None, None, None, True
145 ],
146 username,
147 ]
148
149 lookup_results = req(
150 self._LOOKUP_URL, lookup_req,
151 'Looking up account info', 'Unable to look up account info')
152
153 if lookup_results is False:
154 return False
155
156 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
157 if not user_hash:
158 warn('Unable to extract user hash')
159 return False
160
161 challenge_req = [
162 user_hash,
163 None, 1, None, [1, None, None, None, [password, None, True]],
164 [
165 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],
166 1, [None, None, []], None, None, None, True
167 ]]
168
169 challenge_results = req(
170 self._CHALLENGE_URL, challenge_req,
171 'Logging in', 'Unable to log in')
172
173 if challenge_results is False:
174 return
175
176 login_res = try_get(challenge_results, lambda x: x[0][5], list)
177 if login_res:
178 login_msg = try_get(login_res, lambda x: x[5], compat_str)
179 warn(
180 'Unable to login: %s' % 'Invalid password'
181 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
182 return False
183
184 res = try_get(challenge_results, lambda x: x[0][-1], list)
185 if not res:
186 warn('Unable to extract result entry')
187 return False
188
189 login_challenge = try_get(res, lambda x: x[0][0], list)
190 if login_challenge:
191 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
192 if challenge_str == 'TWO_STEP_VERIFICATION':
193 # SEND_SUCCESS - TFA code has been successfully sent to phone
194 # QUOTA_EXCEEDED - reached the limit of TFA codes
195 status = try_get(login_challenge, lambda x: x[5], compat_str)
196 if status == 'QUOTA_EXCEEDED':
197 warn('Exceeded the limit of TFA codes, try later')
198 return False
199
200 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
201 if not tl:
202 warn('Unable to extract TL')
203 return False
204
205 tfa_code = self._get_tfa_info('2-step verification code')
206
207 if not tfa_code:
208 warn(
209 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
210 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
211 return False
212
213 tfa_code = remove_start(tfa_code, 'G-')
214
215 tfa_req = [
216 user_hash, None, 2, None,
217 [
218 9, None, None, None, None, None, None, None,
219 [None, tfa_code, True, 2]
220 ]]
221
222 tfa_results = req(
223 self._TFA_URL.format(tl), tfa_req,
224 'Submitting TFA code', 'Unable to submit TFA code')
225
226 if tfa_results is False:
227 return False
228
229 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
230 if tfa_res:
231 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
232 warn(
233 'Unable to finish TFA: %s' % 'Invalid TFA code'
234 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
235 return False
236
237 check_cookie_url = try_get(
238 tfa_results, lambda x: x[0][-1][2], compat_str)
239 else:
240 CHALLENGES = {
241 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
242 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
243 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
244 }
245 challenge = CHALLENGES.get(
246 challenge_str,
247 '%s returned error %s.' % (self.IE_NAME, challenge_str))
248 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
249 return False
250 else:
251 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
252
253 if not check_cookie_url:
254 warn('Unable to extract CheckCookie URL')
255 return False
256
257 check_cookie_results = self._download_webpage(
258 check_cookie_url, None, 'Checking cookie', fatal=False)
259
260 if check_cookie_results is False:
261 return False
262
263 if 'https://myaccount.google.com/' not in check_cookie_results:
264 warn('Unable to log in')
265 return False
266
267 return True
268
269 def _download_webpage_handle(self, *args, **kwargs):
270 query = kwargs.get('query', {}).copy()
271 query['disable_polymer'] = 'true'
272 kwargs['query'] = query
273 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
274 *args, **compat_kwargs(kwargs))
275
276 def _real_initialize(self):
277 if self._downloader is None:
278 return
279 self._set_language()
280 if not self._login():
281 return
282
283
284 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
285 # Extract entries from page with "Load more" button
286 def _entries(self, page, playlist_id):
287 more_widget_html = content_html = page
288 for page_num in itertools.count(1):
289 for entry in self._process_page(content_html):
290 yield entry
291
292 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
293 if not mobj:
294 break
295
296 count = 0
297 retries = 3
298 while count <= retries:
299 try:
300 # Downloading page may result in intermittent 5xx HTTP error
301 # that is usually worked around with a retry
302 more = self._download_json(
303 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
304 'Downloading page #%s%s'
305 % (page_num, ' (retry #%d)' % count if count else ''),
306 transform_source=uppercase_escape)
307 break
308 except ExtractorError as e:
309 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
310 count += 1
311 if count <= retries:
312 continue
313 raise
314
315 content_html = more['content_html']
316 if not content_html.strip():
317 # Some webpages show a "Load more" button but they don't
318 # have more videos
319 break
320 more_widget_html = more['load_more_widget_html']
321
322
323 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
324 def _process_page(self, content):
325 for video_id, video_title in self.extract_videos_from_page(content):
326 yield self.url_result(video_id, 'Youtube', video_id, video_title)
327
328 def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page):
329 for mobj in re.finditer(video_re, page):
330 # The link with index 0 is not the first video of the playlist (not sure if still actual)
331 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
332 continue
333 video_id = mobj.group('id')
334 video_title = unescapeHTML(
335 mobj.group('title')) if 'title' in mobj.groupdict() else None
336 if video_title:
337 video_title = video_title.strip()
338 if video_title == 'ā–ŗ Play all':
339 video_title = None
340 try:
341 idx = ids_in_page.index(video_id)
342 if video_title and not titles_in_page[idx]:
343 titles_in_page[idx] = video_title
344 except ValueError:
345 ids_in_page.append(video_id)
346 titles_in_page.append(video_title)
347
348 def extract_videos_from_page(self, page):
349 ids_in_page = []
350 titles_in_page = []
351 self.extract_videos_from_page_impl(
352 self._VIDEO_RE, page, ids_in_page, titles_in_page)
353 return zip(ids_in_page, titles_in_page)
354
355
356 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
357 def _process_page(self, content):
358 for playlist_id in orderedSet(re.findall(
359 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
360 content)):
361 yield self.url_result(
362 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
363
364 def _real_extract(self, url):
365 playlist_id = self._match_id(url)
366 webpage = self._download_webpage(url, playlist_id)
367 title = self._og_search_title(webpage, fatal=False)
368 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
369
370
371 class YoutubeIE(YoutubeBaseInfoExtractor):
372 IE_DESC = 'YouTube.com'
373 _VALID_URL = r"""(?x)^
374 (
375 (?:https?://|//) # http(s):// or protocol-independent URL
376 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
377 (?:www\.)?deturl\.com/www\.youtube\.com/|
378 (?:www\.)?pwnyoutube\.com/|
379 (?:www\.)?hooktube\.com/|
380 (?:www\.)?yourepeat\.com/|
381 tube\.majestyc\.net/|
382 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
383 (?:(?:www|dev)\.)?invidio\.us/|
384 (?:(?:www|no)\.)?invidiou\.sh/|
385 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
386 (?:www\.)?invidious\.kabi\.tk/|
387 (?:www\.)?invidious\.enkirton\.net/|
388 (?:www\.)?invidious\.13ad\.de/|
389 (?:www\.)?invidious\.mastodon\.host/|
390 (?:www\.)?invidious\.nixnet\.xyz/|
391 (?:www\.)?tube\.poal\.co/|
392 (?:www\.)?vid\.wxzm\.sx/|
393 (?:www\.)?yt\.elukerio\.org/|
394 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
395 (?:.*?\#/)? # handle anchor (#/) redirect urls
396 (?: # the various things that can precede the ID:
397 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
398 |(?: # or the v= param in all its forms
399 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
400 (?:\?|\#!?) # the params delimiter ? or # or #!
401 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
402 v=
403 )
404 ))
405 |(?:
406 youtu\.be| # just youtu.be/xxxx
407 vid\.plus| # or vid.plus/xxxx
408 zwearz\.com/watch| # or zwearz.com/watch/xxxx
409 )/
410 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
411 )
412 )? # all until now is optional -> you can pass the naked ID
413 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
414 (?!.*?\blist=
415 (?:
416 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
417 WL # WL are handled by the watch later IE
418 )
419 )
420 (?(1).+)? # if we found the ID, everything can follow
421 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
422 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
423 _formats = {
424 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
425 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
426 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
427 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
428 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
429 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
430 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
431 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
432 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
433 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
434 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
435 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
436 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
437 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
438 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
439 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
440 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
441 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
442
443
444 # 3D videos
445 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
446 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
447 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
448 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
449 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
450 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
451 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
452
453 # Apple HTTP Live Streaming
454 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
455 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
456 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
457 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
458 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
459 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
460 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
461 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
462
463 # DASH mp4 video
464 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
465 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
466 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
467 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
468 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
469 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
470 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
471 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
472 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
473 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
474 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
475 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
476
477 # Dash mp4 audio
478 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
479 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
480 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
481 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
482 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
483 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
484 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
485
486 # Dash webm
487 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
488 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
489 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
490 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
491 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
492 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
493 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
494 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
495 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
496 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
497 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
498 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
499 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
500 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
501 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
502 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
503 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
504 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
505 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
506 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
507 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
508 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
509
510 # Dash webm audio
511 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
512 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
513
514 # Dash webm audio with opus inside
515 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
516 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
517 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
518
519 # RTMP (unnamed)
520 '_rtmp': {'protocol': 'rtmp'},
521
522 # av01 video only formats sometimes served with "unknown" codecs
523 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
524 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
525 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
526 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
527 }
528 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
529
530 _GEO_BYPASS = False
531
532 IE_NAME = 'youtube'
533 _TESTS = [
534 {
535 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
536 'info_dict': {
537 'id': 'BaW_jenozKc',
538 'ext': 'mp4',
539 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
540 'uploader': 'Philipp Hagemeister',
541 'uploader_id': 'phihag',
542 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
543 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
544 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
545 'upload_date': '20121002',
546 '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 .',
547 'categories': ['Science & Technology'],
548 'tags': ['youtube-dl'],
549 'duration': 10,
550 'view_count': int,
551 'like_count': int,
552 'dislike_count': int,
553 'start_time': 1,
554 'end_time': 9,
555 }
556 },
557 {
558 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
559 'note': 'Test generic use_cipher_signature video (#897)',
560 'info_dict': {
561 'id': 'UxxajLWwzqY',
562 'ext': 'mp4',
563 'upload_date': '20120506',
564 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
565 'alt_title': 'I Love It (feat. Charli XCX)',
566 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
567 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
568 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
569 'iconic ep', 'iconic', 'love', 'it'],
570 'duration': 180,
571 'uploader': 'Icona Pop',
572 'uploader_id': 'IconaPop',
573 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
574 'creator': 'Icona Pop',
575 'track': 'I Love It (feat. Charli XCX)',
576 'artist': 'Icona Pop',
577 }
578 },
579 {
580 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
581 'note': 'Test VEVO video with age protection (#956)',
582 'info_dict': {
583 'id': '07FYdnEawAQ',
584 'ext': 'mp4',
585 'upload_date': '20130703',
586 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
587 'alt_title': 'Tunnel Vision',
588 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
589 'duration': 419,
590 'uploader': 'justintimberlakeVEVO',
591 'uploader_id': 'justintimberlakeVEVO',
592 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
593 'creator': 'Justin Timberlake',
594 'track': 'Tunnel Vision',
595 'artist': 'Justin Timberlake',
596 'age_limit': 18,
597 }
598 },
599 {
600 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
601 'note': 'Embed-only video (#1746)',
602 'info_dict': {
603 'id': 'yZIXLfi8CZQ',
604 'ext': 'mp4',
605 'upload_date': '20120608',
606 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
607 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
608 'uploader': 'SET India',
609 'uploader_id': 'setindia',
610 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
611 'age_limit': 18,
612 }
613 },
614 {
615 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
616 'note': 'Use the first video ID in the URL',
617 'info_dict': {
618 'id': 'BaW_jenozKc',
619 'ext': 'mp4',
620 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
621 'uploader': 'Philipp Hagemeister',
622 'uploader_id': 'phihag',
623 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
624 'upload_date': '20121002',
625 '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 .',
626 'categories': ['Science & Technology'],
627 'tags': ['youtube-dl'],
628 'duration': 10,
629 'view_count': int,
630 'like_count': int,
631 'dislike_count': int,
632 },
633 'params': {
634 'skip_download': True,
635 },
636 },
637 {
638 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
639 'note': '256k DASH audio (format 141) via DASH manifest',
640 'info_dict': {
641 'id': 'a9LDPn-MO4I',
642 'ext': 'm4a',
643 'upload_date': '20121002',
644 'uploader_id': '8KVIDEO',
645 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
646 'description': '',
647 'uploader': '8KVIDEO',
648 'title': 'UHDTV TEST 8K VIDEO.mp4'
649 },
650 'params': {
651 'youtube_include_dash_manifest': True,
652 'format': '141',
653 },
654 'skip': 'format 141 not served anymore',
655 },
656 # DASH manifest with encrypted signature
657 {
658 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
659 'info_dict': {
660 'id': 'IB3lcPjvWLA',
661 'ext': 'm4a',
662 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
663 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
664 'duration': 244,
665 'uploader': 'AfrojackVEVO',
666 'uploader_id': 'AfrojackVEVO',
667 'upload_date': '20131011',
668 },
669 'params': {
670 'youtube_include_dash_manifest': True,
671 'format': '141/bestaudio[ext=m4a]',
672 },
673 },
674 # JS player signature function name containing $
675 {
676 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
677 'info_dict': {
678 'id': 'nfWlot6h_JM',
679 'ext': 'm4a',
680 'title': 'Taylor Swift - Shake It Off',
681 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
682 'duration': 242,
683 'uploader': 'TaylorSwiftVEVO',
684 'uploader_id': 'TaylorSwiftVEVO',
685 'upload_date': '20140818',
686 'creator': 'Taylor Swift',
687 },
688 'params': {
689 'youtube_include_dash_manifest': True,
690 'format': '141/bestaudio[ext=m4a]',
691 },
692 },
693 # Controversy video
694 {
695 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
696 'info_dict': {
697 'id': 'T4XJQO3qol8',
698 'ext': 'mp4',
699 'duration': 219,
700 'upload_date': '20100909',
701 'uploader': 'Amazing Atheist',
702 'uploader_id': 'TheAmazingAtheist',
703 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
704 'title': 'Burning Everyone\'s Koran',
705 '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',
706 }
707 },
708 # Normal age-gate video (No vevo, embed allowed)
709 {
710 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
711 'info_dict': {
712 'id': 'HtVdAasjOgU',
713 'ext': 'mp4',
714 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
715 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
716 'duration': 142,
717 'uploader': 'The Witcher',
718 'uploader_id': 'WitcherGame',
719 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
720 'upload_date': '20140605',
721 'age_limit': 18,
722 },
723 },
724 # Age-gate video with encrypted signature
725 {
726 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
727 'info_dict': {
728 'id': '6kLq3WMV1nU',
729 'ext': 'mp4',
730 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
731 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
732 'duration': 246,
733 'uploader': 'LloydVEVO',
734 'uploader_id': 'LloydVEVO',
735 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
736 'upload_date': '20110629',
737 'age_limit': 18,
738 },
739 },
740 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
741 # YouTube Red ad is not captured for creator
742 {
743 'url': '__2ABJjxzNo',
744 'info_dict': {
745 'id': '__2ABJjxzNo',
746 'ext': 'mp4',
747 'duration': 266,
748 'upload_date': '20100430',
749 'uploader_id': 'deadmau5',
750 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
751 'creator': 'deadmau5',
752 'description': 'md5:12c56784b8032162bb936a5f76d55360',
753 'uploader': 'deadmau5',
754 'title': 'Deadmau5 - Some Chords (HD)',
755 'alt_title': 'Some Chords',
756 },
757 'expected_warnings': [
758 'DASH manifest missing',
759 ]
760 },
761 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
762 {
763 'url': 'lqQg6PlCWgI',
764 'info_dict': {
765 'id': 'lqQg6PlCWgI',
766 'ext': 'mp4',
767 'duration': 6085,
768 'upload_date': '20150827',
769 'uploader_id': 'olympic',
770 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
771 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
772 'uploader': 'Olympic',
773 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
774 },
775 'params': {
776 'skip_download': 'requires avconv',
777 }
778 },
779 # Non-square pixels
780 {
781 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
782 'info_dict': {
783 'id': '_b-2C3KPAM0',
784 'ext': 'mp4',
785 'stretched_ratio': 16 / 9.,
786 'duration': 85,
787 'upload_date': '20110310',
788 'uploader_id': 'AllenMeow',
789 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
790 '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',
791 'uploader': 'å­«į„‹į„…',
792 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
793 },
794 },
795 # url_encoded_fmt_stream_map is empty string
796 {
797 'url': 'qEJwOuvDf7I',
798 'info_dict': {
799 'id': 'qEJwOuvDf7I',
800 'ext': 'webm',
801 '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',
802 'description': '',
803 'upload_date': '20150404',
804 'uploader_id': 'spbelect',
805 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
806 },
807 'params': {
808 'skip_download': 'requires avconv',
809 },
810 'skip': 'This live event has ended.',
811 },
812 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
813 {
814 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
815 'info_dict': {
816 'id': 'FIl7x6_3R5Y',
817 'ext': 'webm',
818 'title': 'md5:7b81415841e02ecd4313668cde88737a',
819 'description': 'md5:116377fd2963b81ec4ce64b542173306',
820 'duration': 220,
821 'upload_date': '20150625',
822 'uploader_id': 'dorappi2000',
823 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
824 'uploader': 'dorappi2000',
825 'formats': 'mincount:31',
826 },
827 'skip': 'not actual anymore',
828 },
829 # DASH manifest with segment_list
830 {
831 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
832 'md5': '8ce563a1d667b599d21064e982ab9e31',
833 'info_dict': {
834 'id': 'CsmdDsKjzN8',
835 'ext': 'mp4',
836 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
837 'uploader': 'Airtek',
838 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
839 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
840 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
841 },
842 'params': {
843 'youtube_include_dash_manifest': True,
844 'format': '135', # bestvideo
845 },
846 'skip': 'This live event has ended.',
847 },
848 {
849 # Multifeed videos (multiple cameras), URL is for Main Camera
850 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
851 'info_dict': {
852 'id': 'jqWvoWXjCVs',
853 'title': 'teamPGP: Rocket League Noob Stream',
854 'description': 'md5:dc7872fb300e143831327f1bae3af010',
855 },
856 'playlist': [{
857 'info_dict': {
858 'id': 'jqWvoWXjCVs',
859 'ext': 'mp4',
860 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
861 'description': 'md5:dc7872fb300e143831327f1bae3af010',
862 'duration': 7335,
863 'upload_date': '20150721',
864 'uploader': 'Beer Games Beer',
865 'uploader_id': 'beergamesbeer',
866 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
867 'license': 'Standard YouTube License',
868 },
869 }, {
870 'info_dict': {
871 'id': '6h8e8xoXJzg',
872 'ext': 'mp4',
873 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
874 'description': 'md5:dc7872fb300e143831327f1bae3af010',
875 'duration': 7337,
876 'upload_date': '20150721',
877 'uploader': 'Beer Games Beer',
878 'uploader_id': 'beergamesbeer',
879 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
880 'license': 'Standard YouTube License',
881 },
882 }, {
883 'info_dict': {
884 'id': 'PUOgX5z9xZw',
885 'ext': 'mp4',
886 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
887 'description': 'md5:dc7872fb300e143831327f1bae3af010',
888 'duration': 7337,
889 'upload_date': '20150721',
890 'uploader': 'Beer Games Beer',
891 'uploader_id': 'beergamesbeer',
892 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
893 'license': 'Standard YouTube License',
894 },
895 }, {
896 'info_dict': {
897 'id': 'teuwxikvS5k',
898 'ext': 'mp4',
899 'title': 'teamPGP: Rocket League Noob Stream (zim)',
900 'description': 'md5:dc7872fb300e143831327f1bae3af010',
901 'duration': 7334,
902 'upload_date': '20150721',
903 'uploader': 'Beer Games Beer',
904 'uploader_id': 'beergamesbeer',
905 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
906 'license': 'Standard YouTube License',
907 },
908 }],
909 'params': {
910 'skip_download': True,
911 },
912 'skip': 'This video is not available.',
913 },
914 {
915 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
916 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
917 'info_dict': {
918 'id': 'gVfLd0zydlo',
919 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
920 },
921 'playlist_count': 2,
922 'skip': 'Not multifeed anymore',
923 },
924 {
925 'url': 'https://vid.plus/FlRa-iH7PGw',
926 'only_matching': True,
927 },
928 {
929 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
930 'only_matching': True,
931 },
932 {
933 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
934 # Also tests cut-off URL expansion in video description (see
935 # https://github.com/ytdl-org/youtube-dl/issues/1892,
936 # https://github.com/ytdl-org/youtube-dl/issues/8164)
937 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
938 'info_dict': {
939 'id': 'lsguqyKfVQg',
940 'ext': 'mp4',
941 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
942 'alt_title': 'Dark Walk - Position Music',
943 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
944 'duration': 133,
945 'upload_date': '20151119',
946 'uploader_id': 'IronSoulElf',
947 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
948 'uploader': 'IronSoulElf',
949 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
950 'track': 'Dark Walk - Position Music',
951 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
952 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
953 },
954 'params': {
955 'skip_download': True,
956 },
957 },
958 {
959 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
960 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
961 'only_matching': True,
962 },
963 {
964 # Video with yt:stretch=17:0
965 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
966 'info_dict': {
967 'id': 'Q39EVAstoRM',
968 'ext': 'mp4',
969 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
970 'description': 'md5:ee18a25c350637c8faff806845bddee9',
971 'upload_date': '20151107',
972 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
973 'uploader': 'CH GAMER DROID',
974 },
975 'params': {
976 'skip_download': True,
977 },
978 'skip': 'This video does not exist.',
979 },
980 {
981 # Video licensed under Creative Commons
982 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
983 'info_dict': {
984 'id': 'M4gD1WSo5mA',
985 'ext': 'mp4',
986 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
987 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
988 'duration': 721,
989 'upload_date': '20150127',
990 'uploader_id': 'BerkmanCenter',
991 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
992 'uploader': 'The Berkman Klein Center for Internet & Society',
993 'license': 'Creative Commons Attribution license (reuse allowed)',
994 },
995 'params': {
996 'skip_download': True,
997 },
998 },
999 {
1000 # Channel-like uploader_url
1001 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1002 'info_dict': {
1003 'id': 'eQcmzGIKrzg',
1004 'ext': 'mp4',
1005 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1006 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
1007 'duration': 4060,
1008 'upload_date': '20151119',
1009 'uploader': 'Bernie Sanders',
1010 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1011 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1012 'license': 'Creative Commons Attribution license (reuse allowed)',
1013 },
1014 'params': {
1015 'skip_download': True,
1016 },
1017 },
1018 {
1019 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1020 'only_matching': True,
1021 },
1022 {
1023 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1024 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1025 'only_matching': True,
1026 },
1027 {
1028 # Rental video preview
1029 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1030 'info_dict': {
1031 'id': 'uGpuVWrhIzE',
1032 'ext': 'mp4',
1033 'title': 'Piku - Trailer',
1034 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1035 'upload_date': '20150811',
1036 'uploader': 'FlixMatrix',
1037 'uploader_id': 'FlixMatrixKaravan',
1038 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1039 'license': 'Standard YouTube License',
1040 },
1041 'params': {
1042 'skip_download': True,
1043 },
1044 'skip': 'This video is not available.',
1045 },
1046 {
1047 # YouTube Red video with episode data
1048 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1049 'info_dict': {
1050 'id': 'iqKdEhx-dD4',
1051 'ext': 'mp4',
1052 'title': 'Isolation - Mind Field (Ep 1)',
1053 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1054 'duration': 2085,
1055 'upload_date': '20170118',
1056 'uploader': 'Vsauce',
1057 'uploader_id': 'Vsauce',
1058 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1059 'series': 'Mind Field',
1060 'season_number': 1,
1061 'episode_number': 1,
1062 },
1063 'params': {
1064 'skip_download': True,
1065 },
1066 'expected_warnings': [
1067 'Skipping DASH manifest',
1068 ],
1069 },
1070 {
1071 # The following content has been identified by the YouTube community
1072 # as inappropriate or offensive to some audiences.
1073 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1074 'info_dict': {
1075 'id': '6SJNVb0GnPI',
1076 'ext': 'mp4',
1077 'title': 'Race Differences in Intelligence',
1078 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1079 'duration': 965,
1080 'upload_date': '20140124',
1081 'uploader': 'New Century Foundation',
1082 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1083 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1084 },
1085 'params': {
1086 'skip_download': True,
1087 },
1088 },
1089 {
1090 # itag 212
1091 'url': '1t24XAntNCY',
1092 'only_matching': True,
1093 },
1094 {
1095 # geo restricted to JP
1096 'url': 'sJL6WA-aGkQ',
1097 'only_matching': True,
1098 },
1099 {
1100 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1101 'only_matching': True,
1102 },
1103 {
1104 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1105 'only_matching': True,
1106 },
1107 {
1108 # DRM protected
1109 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1110 'only_matching': True,
1111 },
1112 {
1113 # Video with unsupported adaptive stream type formats
1114 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1115 'info_dict': {
1116 'id': 'Z4Vy8R84T1U',
1117 'ext': 'mp4',
1118 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1119 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1120 'duration': 433,
1121 'upload_date': '20130923',
1122 'uploader': 'Amelia Putri Harwita',
1123 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1124 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1125 'formats': 'maxcount:10',
1126 },
1127 'params': {
1128 'skip_download': True,
1129 'youtube_include_dash_manifest': False,
1130 },
1131 },
1132 {
1133 # Youtube Music Auto-generated description
1134 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1135 'info_dict': {
1136 'id': 'MgNrAu2pzNs',
1137 'ext': 'mp4',
1138 'title': 'Voyeur Girl',
1139 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1140 'upload_date': '20190312',
1141 'uploader': 'Various Artists - Topic',
1142 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1143 'artist': 'Stephen',
1144 'track': 'Voyeur Girl',
1145 'album': 'it\'s too much love to know my dear',
1146 'release_date': '20190313',
1147 'release_year': 2019,
1148 },
1149 'params': {
1150 'skip_download': True,
1151 },
1152 },
1153 {
1154 # Youtube Music Auto-generated description
1155 # Retrieve 'artist' field from 'Artist:' in video description
1156 # when it is present on youtube music video
1157 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1158 'info_dict': {
1159 'id': 'k0jLE7tTwjY',
1160 'ext': 'mp4',
1161 'title': 'Latch Feat. Sam Smith',
1162 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1163 'upload_date': '20150110',
1164 'uploader': 'Various Artists - Topic',
1165 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1166 'artist': 'Disclosure',
1167 'track': 'Latch Feat. Sam Smith',
1168 'album': 'Latch Featuring Sam Smith',
1169 'release_date': '20121008',
1170 'release_year': 2012,
1171 },
1172 'params': {
1173 'skip_download': True,
1174 },
1175 },
1176 {
1177 # Youtube Music Auto-generated description
1178 # handle multiple artists on youtube music video
1179 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1180 'info_dict': {
1181 'id': '74qn0eJSjpA',
1182 'ext': 'mp4',
1183 'title': 'Eastside',
1184 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1185 'upload_date': '20180710',
1186 'uploader': 'Benny Blanco - Topic',
1187 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1188 'artist': 'benny blanco, Halsey, Khalid',
1189 'track': 'Eastside',
1190 'album': 'Eastside',
1191 'release_date': '20180713',
1192 'release_year': 2018,
1193 },
1194 'params': {
1195 'skip_download': True,
1196 },
1197 },
1198 {
1199 # Youtube Music Auto-generated description
1200 # handle youtube music video with release_year and no release_date
1201 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1202 'info_dict': {
1203 'id': '-hcAI0g-f5M',
1204 'ext': 'mp4',
1205 'title': 'Put It On Me',
1206 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1207 'upload_date': '20180426',
1208 'uploader': 'Matt Maeson - Topic',
1209 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1210 'artist': 'Matt Maeson',
1211 'track': 'Put It On Me',
1212 'album': 'The Hearse',
1213 'release_date': None,
1214 'release_year': 2018,
1215 },
1216 'params': {
1217 'skip_download': True,
1218 },
1219 },
1220 ]
1221
1222 def __init__(self, *args, **kwargs):
1223 super(YoutubeIE, self).__init__(*args, **kwargs)
1224 self._player_cache = {}
1225
1226 def report_video_info_webpage_download(self, video_id):
1227 """Report attempt to download video info webpage."""
1228 self.to_screen('%s: Downloading video info webpage' % video_id)
1229
1230 def report_information_extraction(self, video_id):
1231 """Report attempt to extract video information."""
1232 self.to_screen('%s: Extracting video information' % video_id)
1233
1234 def report_unavailable_format(self, video_id, format):
1235 """Report extracted video URL."""
1236 self.to_screen('%s: Format %s not available' % (video_id, format))
1237
1238 def report_rtmp_download(self):
1239 """Indicate the download will use the RTMP protocol."""
1240 self.to_screen('RTMP download detected')
1241
1242 def _signature_cache_id(self, example_sig):
1243 """ Return a string representation of a signature """
1244 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1245
1246 def _extract_signature_function(self, video_id, player_url, example_sig):
1247 id_m = re.match(
1248 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1249 player_url)
1250 if not id_m:
1251 raise ExtractorError('Cannot identify player %r' % player_url)
1252 player_type = id_m.group('ext')
1253 player_id = id_m.group('id')
1254
1255 # Read from filesystem cache
1256 func_id = '%s_%s_%s' % (
1257 player_type, player_id, self._signature_cache_id(example_sig))
1258 assert os.path.basename(func_id) == func_id
1259
1260 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1261 if cache_spec is not None:
1262 return lambda s: ''.join(s[i] for i in cache_spec)
1263
1264 download_note = (
1265 'Downloading player %s' % player_url
1266 if self._downloader.params.get('verbose') else
1267 'Downloading %s player %s' % (player_type, player_id)
1268 )
1269 if player_type == 'js':
1270 code = self._download_webpage(
1271 player_url, video_id,
1272 note=download_note,
1273 errnote='Download of %s failed' % player_url)
1274 res = self._parse_sig_js(code)
1275 elif player_type == 'swf':
1276 urlh = self._request_webpage(
1277 player_url, video_id,
1278 note=download_note,
1279 errnote='Download of %s failed' % player_url)
1280 code = urlh.read()
1281 res = self._parse_sig_swf(code)
1282 else:
1283 assert False, 'Invalid player type %r' % player_type
1284
1285 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1286 cache_res = res(test_string)
1287 cache_spec = [ord(c) for c in cache_res]
1288
1289 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1290 return res
1291
1292 def _print_sig_code(self, func, example_sig):
1293 def gen_sig_code(idxs):
1294 def _genslice(start, end, step):
1295 starts = '' if start == 0 else str(start)
1296 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1297 steps = '' if step == 1 else (':%d' % step)
1298 return 's[%s%s%s]' % (starts, ends, steps)
1299
1300 step = None
1301 # Quelch pyflakes warnings - start will be set when step is set
1302 start = '(Never used)'
1303 for i, prev in zip(idxs[1:], idxs[:-1]):
1304 if step is not None:
1305 if i - prev == step:
1306 continue
1307 yield _genslice(start, prev, step)
1308 step = None
1309 continue
1310 if i - prev in [-1, 1]:
1311 step = i - prev
1312 start = prev
1313 continue
1314 else:
1315 yield 's[%d]' % prev
1316 if step is None:
1317 yield 's[%d]' % i
1318 else:
1319 yield _genslice(start, i, step)
1320
1321 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1322 cache_res = func(test_string)
1323 cache_spec = [ord(c) for c in cache_res]
1324 expr_code = ' + '.join(gen_sig_code(cache_spec))
1325 signature_id_tuple = '(%s)' % (
1326 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1327 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1328 ' return %s\n') % (signature_id_tuple, expr_code)
1329 self.to_screen('Extracted signature function:\n' + code)
1330
1331 def _parse_sig_js(self, jscode):
1332 funcname = self._search_regex(
1333 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1334 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1335 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1336 # Obsolete patterns
1337 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1338 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1339 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1340 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1341 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1342 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1343 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1344 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1345 jscode, 'Initial JS player signature function name', group='sig')
1346
1347 jsi = JSInterpreter(jscode)
1348 initial_function = jsi.extract_function(funcname)
1349 return lambda s: initial_function([s])
1350
1351 def _parse_sig_swf(self, file_contents):
1352 swfi = SWFInterpreter(file_contents)
1353 TARGET_CLASSNAME = 'SignatureDecipher'
1354 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1355 initial_function = swfi.extract_function(searched_class, 'decipher')
1356 return lambda s: initial_function([s])
1357
1358 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1359 """Turn the encrypted s field into a working signature"""
1360
1361 if player_url is None:
1362 raise ExtractorError('Cannot decrypt signature without player_url')
1363
1364 if player_url.startswith('//'):
1365 player_url = 'https:' + player_url
1366 elif not re.match(r'https?://', player_url):
1367 player_url = compat_urlparse.urljoin(
1368 'https://www.youtube.com', player_url)
1369 try:
1370 player_id = (player_url, self._signature_cache_id(s))
1371 if player_id not in self._player_cache:
1372 func = self._extract_signature_function(
1373 video_id, player_url, s
1374 )
1375 self._player_cache[player_id] = func
1376 func = self._player_cache[player_id]
1377 if self._downloader.params.get('youtube_print_sig_code'):
1378 self._print_sig_code(func, s)
1379 return func(s)
1380 except Exception as e:
1381 tb = traceback.format_exc()
1382 raise ExtractorError(
1383 'Signature extraction failed: ' + tb, cause=e)
1384
1385 def _get_subtitles(self, video_id, webpage):
1386 try:
1387 subs_doc = self._download_xml(
1388 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1389 video_id, note=False)
1390 except ExtractorError as err:
1391 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1392 return {}
1393
1394 sub_lang_list = {}
1395 for track in subs_doc.findall('track'):
1396 lang = track.attrib['lang_code']
1397 if lang in sub_lang_list:
1398 continue
1399 sub_formats = []
1400 for ext in self._SUBTITLE_FORMATS:
1401 params = compat_urllib_parse_urlencode({
1402 'lang': lang,
1403 'v': video_id,
1404 'fmt': ext,
1405 'name': track.attrib['name'].encode('utf-8'),
1406 })
1407 sub_formats.append({
1408 'url': 'https://www.youtube.com/api/timedtext?' + params,
1409 'ext': ext,
1410 })
1411 sub_lang_list[lang] = sub_formats
1412 if not sub_lang_list:
1413 self._downloader.report_warning('video doesn\'t have subtitles')
1414 return {}
1415 return sub_lang_list
1416
1417 def _get_ytplayer_config(self, video_id, webpage):
1418 patterns = (
1419 # User data may contain arbitrary character sequences that may affect
1420 # JSON extraction with regex, e.g. when '};' is contained the second
1421 # regex won't capture the whole JSON. Yet working around by trying more
1422 # concrete regex first keeping in mind proper quoted string handling
1423 # to be implemented in future that will replace this workaround (see
1424 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1425 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1426 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1427 r';ytplayer\.config\s*=\s*({.+?});',
1428 )
1429 config = self._search_regex(
1430 patterns, webpage, 'ytplayer.config', default=None)
1431 if config:
1432 return self._parse_json(
1433 uppercase_escape(config), video_id, fatal=False)
1434
1435 def _get_automatic_captions(self, video_id, webpage):
1436 """We need the webpage for getting the captions url, pass it as an
1437 argument to speed up the process."""
1438 self.to_screen('%s: Looking for automatic captions' % video_id)
1439 player_config = self._get_ytplayer_config(video_id, webpage)
1440 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1441 if not player_config:
1442 self._downloader.report_warning(err_msg)
1443 return {}
1444 try:
1445 args = player_config['args']
1446 caption_url = args.get('ttsurl')
1447 if caption_url:
1448 timestamp = args['timestamp']
1449 # We get the available subtitles
1450 list_params = compat_urllib_parse_urlencode({
1451 'type': 'list',
1452 'tlangs': 1,
1453 'asrs': 1,
1454 })
1455 list_url = caption_url + '&' + list_params
1456 caption_list = self._download_xml(list_url, video_id)
1457 original_lang_node = caption_list.find('track')
1458 if original_lang_node is None:
1459 self._downloader.report_warning('Video doesn\'t have automatic captions')
1460 return {}
1461 original_lang = original_lang_node.attrib['lang_code']
1462 caption_kind = original_lang_node.attrib.get('kind', '')
1463
1464 sub_lang_list = {}
1465 for lang_node in caption_list.findall('target'):
1466 sub_lang = lang_node.attrib['lang_code']
1467 sub_formats = []
1468 for ext in self._SUBTITLE_FORMATS:
1469 params = compat_urllib_parse_urlencode({
1470 'lang': original_lang,
1471 'tlang': sub_lang,
1472 'fmt': ext,
1473 'ts': timestamp,
1474 'kind': caption_kind,
1475 })
1476 sub_formats.append({
1477 'url': caption_url + '&' + params,
1478 'ext': ext,
1479 })
1480 sub_lang_list[sub_lang] = sub_formats
1481 return sub_lang_list
1482
1483 def make_captions(sub_url, sub_langs):
1484 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1485 caption_qs = compat_parse_qs(parsed_sub_url.query)
1486 captions = {}
1487 for sub_lang in sub_langs:
1488 sub_formats = []
1489 for ext in self._SUBTITLE_FORMATS:
1490 caption_qs.update({
1491 'tlang': [sub_lang],
1492 'fmt': [ext],
1493 })
1494 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1495 query=compat_urllib_parse_urlencode(caption_qs, True)))
1496 sub_formats.append({
1497 'url': sub_url,
1498 'ext': ext,
1499 })
1500 captions[sub_lang] = sub_formats
1501 return captions
1502
1503 # New captions format as of 22.06.2017
1504 player_response = args.get('player_response')
1505 if player_response and isinstance(player_response, compat_str):
1506 player_response = self._parse_json(
1507 player_response, video_id, fatal=False)
1508 if player_response:
1509 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1510 base_url = renderer['captionTracks'][0]['baseUrl']
1511 sub_lang_list = []
1512 for lang in renderer['translationLanguages']:
1513 lang_code = lang.get('languageCode')
1514 if lang_code:
1515 sub_lang_list.append(lang_code)
1516 return make_captions(base_url, sub_lang_list)
1517
1518 # Some videos don't provide ttsurl but rather caption_tracks and
1519 # caption_translation_languages (e.g. 20LmZk1hakA)
1520 # Does not used anymore as of 22.06.2017
1521 caption_tracks = args['caption_tracks']
1522 caption_translation_languages = args['caption_translation_languages']
1523 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1524 sub_lang_list = []
1525 for lang in caption_translation_languages.split(','):
1526 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1527 sub_lang = lang_qs.get('lc', [None])[0]
1528 if sub_lang:
1529 sub_lang_list.append(sub_lang)
1530 return make_captions(caption_url, sub_lang_list)
1531 # An extractor error can be raise by the download process if there are
1532 # no automatic captions but there are subtitles
1533 except (KeyError, IndexError, ExtractorError):
1534 self._downloader.report_warning(err_msg)
1535 return {}
1536
1537 def _mark_watched(self, video_id, video_info, player_response):
1538 playback_url = url_or_none(try_get(
1539 player_response,
1540 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1541 video_info, lambda x: x['videostats_playback_base_url'][0]))
1542 if not playback_url:
1543 return
1544 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1545 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1546
1547 # cpn generation algorithm is reverse engineered from base.js.
1548 # In fact it works even with dummy cpn.
1549 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1550 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1551
1552 qs.update({
1553 'ver': ['2'],
1554 'cpn': [cpn],
1555 })
1556 playback_url = compat_urlparse.urlunparse(
1557 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1558
1559 self._download_webpage(
1560 playback_url, video_id, 'Marking watched',
1561 'Unable to mark watched', fatal=False)
1562
1563 @staticmethod
1564 def _extract_urls(webpage):
1565 # Embedded YouTube player
1566 entries = [
1567 unescapeHTML(mobj.group('url'))
1568 for mobj in re.finditer(r'''(?x)
1569 (?:
1570 <iframe[^>]+?src=|
1571 data-video-url=|
1572 <embed[^>]+?src=|
1573 embedSWF\(?:\s*|
1574 <object[^>]+data=|
1575 new\s+SWFObject\(
1576 )
1577 (["\'])
1578 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1579 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1580 \1''', webpage)]
1581
1582 # lazyYT YouTube embed
1583 entries.extend(list(map(
1584 unescapeHTML,
1585 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1586
1587 # Wordpress "YouTube Video Importer" plugin
1588 matches = re.findall(r'''(?x)<div[^>]+
1589 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1590 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1591 entries.extend(m[-1] for m in matches)
1592
1593 return entries
1594
1595 @staticmethod
1596 def _extract_url(webpage):
1597 urls = YoutubeIE._extract_urls(webpage)
1598 return urls[0] if urls else None
1599
1600 @classmethod
1601 def extract_id(cls, url):
1602 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1603 if mobj is None:
1604 raise ExtractorError('Invalid URL: %s' % url)
1605 video_id = mobj.group(2)
1606 return video_id
1607
1608 @staticmethod
1609 def _extract_chapters(description, duration):
1610 if not description:
1611 return None
1612 chapter_lines = re.findall(
1613 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1614 description)
1615 if not chapter_lines:
1616 return None
1617 chapters = []
1618 for next_num, (chapter_line, time_point) in enumerate(
1619 chapter_lines, start=1):
1620 start_time = parse_duration(time_point)
1621 if start_time is None:
1622 continue
1623 if start_time > duration:
1624 break
1625 end_time = (duration if next_num == len(chapter_lines)
1626 else parse_duration(chapter_lines[next_num][1]))
1627 if end_time is None:
1628 continue
1629 if end_time > duration:
1630 end_time = duration
1631 if start_time > end_time:
1632 break
1633 chapter_title = re.sub(
1634 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1635 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1636 chapters.append({
1637 'start_time': start_time,
1638 'end_time': end_time,
1639 'title': chapter_title,
1640 })
1641 return chapters
1642
1643 def _real_extract(self, url):
1644 url, smuggled_data = unsmuggle_url(url, {})
1645
1646 proto = (
1647 'http' if self._downloader.params.get('prefer_insecure', False)
1648 else 'https')
1649
1650 start_time = None
1651 end_time = None
1652 parsed_url = compat_urllib_parse_urlparse(url)
1653 for component in [parsed_url.fragment, parsed_url.query]:
1654 query = compat_parse_qs(component)
1655 if start_time is None and 't' in query:
1656 start_time = parse_duration(query['t'][0])
1657 if start_time is None and 'start' in query:
1658 start_time = parse_duration(query['start'][0])
1659 if end_time is None and 'end' in query:
1660 end_time = parse_duration(query['end'][0])
1661
1662 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1663 mobj = re.search(self._NEXT_URL_RE, url)
1664 if mobj:
1665 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1666 video_id = self.extract_id(url)
1667
1668 # Get video webpage
1669 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1670 video_webpage = self._download_webpage(url, video_id)
1671
1672 # Attempt to extract SWF player URL
1673 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1674 if mobj is not None:
1675 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1676 else:
1677 player_url = None
1678
1679 dash_mpds = []
1680
1681 def add_dash_mpd(video_info):
1682 dash_mpd = video_info.get('dashmpd')
1683 if dash_mpd and dash_mpd[0] not in dash_mpds:
1684 dash_mpds.append(dash_mpd[0])
1685
1686 def add_dash_mpd_pr(pl_response):
1687 dash_mpd = url_or_none(try_get(
1688 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1689 compat_str))
1690 if dash_mpd and dash_mpd not in dash_mpds:
1691 dash_mpds.append(dash_mpd)
1692
1693 is_live = None
1694 view_count = None
1695
1696 def extract_view_count(v_info):
1697 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1698
1699 def extract_token(v_info):
1700 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1701
1702 def extract_player_response(player_response, video_id):
1703 pl_response = str_or_none(player_response)
1704 if not pl_response:
1705 return
1706 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1707 if isinstance(pl_response, dict):
1708 add_dash_mpd_pr(pl_response)
1709 return pl_response
1710
1711 player_response = {}
1712
1713 # Get video info
1714 embed_webpage = None
1715 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1716 age_gate = True
1717 # We simulate the access to the video from www.youtube.com/v/{video_id}
1718 # this can be viewed without login into Youtube
1719 url = proto + '://www.youtube.com/embed/%s' % video_id
1720 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1721 data = compat_urllib_parse_urlencode({
1722 'video_id': video_id,
1723 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1724 'sts': self._search_regex(
1725 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1726 })
1727 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1728 video_info_webpage = self._download_webpage(
1729 video_info_url, video_id,
1730 note='Refetching age-gated info webpage',
1731 errnote='unable to download video info webpage')
1732 video_info = compat_parse_qs(video_info_webpage)
1733 pl_response = video_info.get('player_response', [None])[0]
1734 player_response = extract_player_response(pl_response, video_id)
1735 add_dash_mpd(video_info)
1736 view_count = extract_view_count(video_info)
1737 else:
1738 age_gate = False
1739 video_info = None
1740 sts = None
1741 # Try looking directly into the video webpage
1742 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1743 if ytplayer_config:
1744 args = ytplayer_config['args']
1745 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1746 # Convert to the same format returned by compat_parse_qs
1747 video_info = dict((k, [v]) for k, v in args.items())
1748 add_dash_mpd(video_info)
1749 # Rental video is not rented but preview is available (e.g.
1750 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1751 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1752 if not video_info and args.get('ypc_vid'):
1753 return self.url_result(
1754 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1755 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1756 is_live = True
1757 sts = ytplayer_config.get('sts')
1758 if not player_response:
1759 player_response = extract_player_response(args.get('player_response'), video_id)
1760 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1761 add_dash_mpd_pr(player_response)
1762 # We also try looking in get_video_info since it may contain different dashmpd
1763 # URL that points to a DASH manifest with possibly different itag set (some itags
1764 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1765 # manifest pointed by get_video_info's dashmpd).
1766 # The general idea is to take a union of itags of both DASH manifests (for example
1767 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1768 self.report_video_info_webpage_download(video_id)
1769 for el in ('embedded', 'detailpage', 'vevo', ''):
1770 query = {
1771 'video_id': video_id,
1772 'ps': 'default',
1773 'eurl': '',
1774 'gl': 'US',
1775 'hl': 'en',
1776 }
1777 if el:
1778 query['el'] = el
1779 if sts:
1780 query['sts'] = sts
1781 video_info_webpage = self._download_webpage(
1782 '%s://www.youtube.com/get_video_info' % proto,
1783 video_id, note=False,
1784 errnote='unable to download video info webpage',
1785 fatal=False, query=query)
1786 if not video_info_webpage:
1787 continue
1788 get_video_info = compat_parse_qs(video_info_webpage)
1789 if not player_response:
1790 pl_response = get_video_info.get('player_response', [None])[0]
1791 player_response = extract_player_response(pl_response, video_id)
1792 add_dash_mpd(get_video_info)
1793 if view_count is None:
1794 view_count = extract_view_count(get_video_info)
1795 if not video_info:
1796 video_info = get_video_info
1797 get_token = extract_token(get_video_info)
1798 if get_token:
1799 # Different get_video_info requests may report different results, e.g.
1800 # some may report video unavailability, but some may serve it without
1801 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1802 # the original webpage as well as el=info and el=embedded get_video_info
1803 # requests report video unavailability due to geo restriction while
1804 # el=detailpage succeeds and returns valid data). This is probably
1805 # due to YouTube measures against IP ranges of hosting providers.
1806 # Working around by preferring the first succeeded video_info containing
1807 # the token if no such video_info yet was found.
1808 token = extract_token(video_info)
1809 if not token:
1810 video_info = get_video_info
1811 break
1812
1813 def extract_unavailable_message():
1814 messages = []
1815 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1816 msg = self._html_search_regex(
1817 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1818 video_webpage, 'unavailable %s' % kind, default=None)
1819 if msg:
1820 messages.append(msg)
1821 if messages:
1822 return '\n'.join(messages)
1823
1824 if not video_info:
1825 unavailable_message = extract_unavailable_message()
1826 if not unavailable_message:
1827 unavailable_message = 'Unable to extract video data'
1828 raise ExtractorError(
1829 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1830
1831 video_details = try_get(
1832 player_response, lambda x: x['videoDetails'], dict) or {}
1833
1834 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1835 if not video_title:
1836 self._downloader.report_warning('Unable to extract video title')
1837 video_title = '_'
1838
1839 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1840 if video_description:
1841
1842 def replace_url(m):
1843 redir_url = compat_urlparse.urljoin(url, m.group(1))
1844 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1845 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1846 qs = compat_parse_qs(parsed_redir_url.query)
1847 q = qs.get('q')
1848 if q and q[0]:
1849 return q[0]
1850 return redir_url
1851
1852 description_original = video_description = re.sub(r'''(?x)
1853 <a\s+
1854 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1855 (?:title|href)="([^"]+)"\s+
1856 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1857 class="[^"]*"[^>]*>
1858 [^<]+\.{3}\s*
1859 </a>
1860 ''', replace_url, video_description)
1861 video_description = clean_html(video_description)
1862 else:
1863 video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
1864
1865 if not smuggled_data.get('force_singlefeed', False):
1866 if not self._downloader.params.get('noplaylist'):
1867 multifeed_metadata_list = try_get(
1868 player_response,
1869 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1870 compat_str) or try_get(
1871 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1872 if multifeed_metadata_list:
1873 entries = []
1874 feed_ids = []
1875 for feed in multifeed_metadata_list.split(','):
1876 # Unquote should take place before split on comma (,) since textual
1877 # fields may contain comma as well (see
1878 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1879 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1880 entries.append({
1881 '_type': 'url_transparent',
1882 'ie_key': 'Youtube',
1883 'url': smuggle_url(
1884 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1885 {'force_singlefeed': True}),
1886 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1887 })
1888 feed_ids.append(feed_data['id'][0])
1889 self.to_screen(
1890 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1891 % (', '.join(feed_ids), video_id))
1892 return self.playlist_result(entries, video_id, video_title, video_description)
1893 else:
1894 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1895
1896 if view_count is None:
1897 view_count = extract_view_count(video_info)
1898 if view_count is None and video_details:
1899 view_count = int_or_none(video_details.get('viewCount'))
1900
1901 if is_live is None:
1902 is_live = bool_or_none(video_details.get('isLive'))
1903
1904 # Check for "rental" videos
1905 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1906 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1907
1908 def _extract_filesize(media_url):
1909 return int_or_none(self._search_regex(
1910 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1911
1912 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1913 self.report_rtmp_download()
1914 formats = [{
1915 'format_id': '_rtmp',
1916 'protocol': 'rtmp',
1917 'url': video_info['conn'][0],
1918 'player_url': player_url,
1919 }]
1920 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):
1921 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1922 if 'rtmpe%3Dyes' in encoded_url_map:
1923 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1924 formats_spec = {}
1925 fmt_list = video_info.get('fmt_list', [''])[0]
1926 if fmt_list:
1927 for fmt in fmt_list.split(','):
1928 spec = fmt.split('/')
1929 if len(spec) > 1:
1930 width_height = spec[1].split('x')
1931 if len(width_height) == 2:
1932 formats_spec[spec[0]] = {
1933 'resolution': spec[1],
1934 'width': int_or_none(width_height[0]),
1935 'height': int_or_none(width_height[1]),
1936 }
1937 q = qualities(['small', 'medium', 'hd720'])
1938 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1939 if streaming_formats:
1940 for fmt in streaming_formats:
1941 itag = str_or_none(fmt.get('itag'))
1942 if not itag:
1943 continue
1944 quality = fmt.get('quality')
1945 quality_label = fmt.get('qualityLabel') or quality
1946 formats_spec[itag] = {
1947 'asr': int_or_none(fmt.get('audioSampleRate')),
1948 'filesize': int_or_none(fmt.get('contentLength')),
1949 'format_note': quality_label,
1950 'fps': int_or_none(fmt.get('fps')),
1951 'height': int_or_none(fmt.get('height')),
1952 'quality': q(quality),
1953 # bitrate for itag 43 is always 2147483647
1954 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1955 'width': int_or_none(fmt.get('width')),
1956 }
1957 formats = []
1958 for url_data_str in encoded_url_map.split(','):
1959 url_data = compat_parse_qs(url_data_str)
1960 if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
1961 continue
1962 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1963 # Unsupported FORMAT_STREAM_TYPE_OTF
1964 if stream_type == 3:
1965 continue
1966 format_id = url_data['itag'][0]
1967 url = url_data['url'][0]
1968
1969 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1970 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1971 jsplayer_url_json = self._search_regex(
1972 ASSETS_RE,
1973 embed_webpage if age_gate else video_webpage,
1974 'JS player URL (1)', default=None)
1975 if not jsplayer_url_json and not age_gate:
1976 # We need the embed website after all
1977 if embed_webpage is None:
1978 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1979 embed_webpage = self._download_webpage(
1980 embed_url, video_id, 'Downloading embed webpage')
1981 jsplayer_url_json = self._search_regex(
1982 ASSETS_RE, embed_webpage, 'JS player URL')
1983
1984 player_url = json.loads(jsplayer_url_json)
1985 if player_url is None:
1986 player_url_json = self._search_regex(
1987 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1988 video_webpage, 'age gate player URL')
1989 player_url = json.loads(player_url_json)
1990
1991 if 'sig' in url_data:
1992 url += '&signature=' + url_data['sig'][0]
1993 elif 's' in url_data:
1994 encrypted_sig = url_data['s'][0]
1995
1996 if self._downloader.params.get('verbose'):
1997 if player_url is None:
1998 player_version = 'unknown'
1999 player_desc = 'unknown'
2000 else:
2001 if player_url.endswith('swf'):
2002 player_version = self._search_regex(
2003 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
2004 'flash player', fatal=False)
2005 player_desc = 'flash player %s' % player_version
2006 else:
2007 player_version = self._search_regex(
2008 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
2009 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
2010 player_url,
2011 'html5 player', fatal=False)
2012 player_desc = 'html5 player %s' % player_version
2013
2014 parts_sizes = self._signature_cache_id(encrypted_sig)
2015 self.to_screen('{%s} signature length %s, %s' %
2016 (format_id, parts_sizes, player_desc))
2017
2018 signature = self._decrypt_signature(
2019 encrypted_sig, video_id, player_url, age_gate)
2020 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2021 url += '&%s=%s' % (sp, signature)
2022 if 'ratebypass' not in url:
2023 url += '&ratebypass=yes'
2024
2025 dct = {
2026 'format_id': format_id,
2027 'url': url,
2028 'player_url': player_url,
2029 }
2030 if format_id in self._formats:
2031 dct.update(self._formats[format_id])
2032 if format_id in formats_spec:
2033 dct.update(formats_spec[format_id])
2034
2035 # Some itags are not included in DASH manifest thus corresponding formats will
2036 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2037 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2038 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2039 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2040
2041 filesize = int_or_none(url_data.get(
2042 'clen', [None])[0]) or _extract_filesize(url)
2043
2044 quality = url_data.get('quality', [None])[0]
2045
2046 more_fields = {
2047 'filesize': filesize,
2048 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2049 'width': width,
2050 'height': height,
2051 'fps': int_or_none(url_data.get('fps', [None])[0]),
2052 'format_note': url_data.get('quality_label', [None])[0] or quality,
2053 'quality': q(quality),
2054 }
2055 for key, value in more_fields.items():
2056 if value:
2057 dct[key] = value
2058 type_ = url_data.get('type', [None])[0]
2059 if type_:
2060 type_split = type_.split(';')
2061 kind_ext = type_split[0].split('/')
2062 if len(kind_ext) == 2:
2063 kind, _ = kind_ext
2064 dct['ext'] = mimetype2ext(type_split[0])
2065 if kind in ('audio', 'video'):
2066 codecs = None
2067 for mobj in re.finditer(
2068 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2069 if mobj.group('key') == 'codecs':
2070 codecs = mobj.group('val')
2071 break
2072 if codecs:
2073 dct.update(parse_codecs(codecs))
2074 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2075 dct['downloader_options'] = {
2076 # Youtube throttles chunks >~10M
2077 'http_chunk_size': 10485760,
2078 }
2079 formats.append(dct)
2080 else:
2081 manifest_url = (
2082 url_or_none(try_get(
2083 player_response,
2084 lambda x: x['streamingData']['hlsManifestUrl'],
2085 compat_str))
2086 or url_or_none(try_get(
2087 video_info, lambda x: x['hlsvp'][0], compat_str)))
2088 if manifest_url:
2089 formats = []
2090 m3u8_formats = self._extract_m3u8_formats(
2091 manifest_url, video_id, 'mp4', fatal=False)
2092 for a_format in m3u8_formats:
2093 itag = self._search_regex(
2094 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2095 if itag:
2096 a_format['format_id'] = itag
2097 if itag in self._formats:
2098 dct = self._formats[itag].copy()
2099 dct.update(a_format)
2100 a_format = dct
2101 a_format['player_url'] = player_url
2102 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2103 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2104 formats.append(a_format)
2105 else:
2106 error_message = extract_unavailable_message()
2107 if not error_message:
2108 error_message = clean_html(try_get(
2109 player_response, lambda x: x['playabilityStatus']['reason'],
2110 compat_str))
2111 if not error_message:
2112 error_message = clean_html(
2113 try_get(video_info, lambda x: x['reason'][0], compat_str))
2114 if error_message:
2115 raise ExtractorError(error_message, expected=True)
2116 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2117
2118 # uploader
2119 video_uploader = try_get(
2120 video_info, lambda x: x['author'][0],
2121 compat_str) or str_or_none(video_details.get('author'))
2122 if video_uploader:
2123 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2124 else:
2125 self._downloader.report_warning('unable to extract uploader name')
2126
2127 # uploader_id
2128 video_uploader_id = None
2129 video_uploader_url = None
2130 mobj = re.search(
2131 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2132 video_webpage)
2133 if mobj is not None:
2134 video_uploader_id = mobj.group('uploader_id')
2135 video_uploader_url = mobj.group('uploader_url')
2136 else:
2137 self._downloader.report_warning('unable to extract uploader nickname')
2138
2139 channel_id = (
2140 str_or_none(video_details.get('channelId'))
2141 or self._html_search_meta(
2142 'channelId', video_webpage, 'channel id', default=None)
2143 or self._search_regex(
2144 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2145 video_webpage, 'channel id', default=None, group='id'))
2146 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2147
2148 # thumbnail image
2149 # We try first to get a high quality image:
2150 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2151 video_webpage, re.DOTALL)
2152 if m_thumb is not None:
2153 video_thumbnail = m_thumb.group(1)
2154 elif 'thumbnail_url' not in video_info:
2155 self._downloader.report_warning('unable to extract video thumbnail')
2156 video_thumbnail = None
2157 else: # don't panic if we can't find it
2158 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2159
2160 # upload date
2161 upload_date = self._html_search_meta(
2162 'datePublished', video_webpage, 'upload date', default=None)
2163 if not upload_date:
2164 upload_date = self._search_regex(
2165 [r'(?s)id="eow-date.*?>(.*?)</span>',
2166 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2167 video_webpage, 'upload date', default=None)
2168 upload_date = unified_strdate(upload_date)
2169
2170 video_license = self._html_search_regex(
2171 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2172 video_webpage, 'license', default=None)
2173
2174 m_music = re.search(
2175 r'''(?x)
2176 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2177 <ul[^>]*>\s*
2178 <li>(?P<title>.+?)
2179 by (?P<creator>.+?)
2180 (?:
2181 \(.+?\)|
2182 <a[^>]*
2183 (?:
2184 \bhref=["\']/red[^>]*>| # drop possible
2185 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2186 )
2187 .*?
2188 )?</li
2189 ''',
2190 video_webpage)
2191 if m_music:
2192 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2193 video_creator = clean_html(m_music.group('creator'))
2194 else:
2195 video_alt_title = video_creator = None
2196
2197 def extract_meta(field):
2198 return self._html_search_regex(
2199 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2200 video_webpage, field, default=None)
2201
2202 track = extract_meta('Song')
2203 artist = extract_meta('Artist')
2204 album = extract_meta('Album')
2205
2206 # Youtube Music Auto-generated description
2207 release_date = release_year = None
2208 if video_description:
2209 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)
2210 if mobj:
2211 if not track:
2212 track = mobj.group('track').strip()
2213 if not artist:
2214 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('Ā·'))
2215 if not album:
2216 album = mobj.group('album'.strip())
2217 release_year = mobj.group('release_year')
2218 release_date = mobj.group('release_date')
2219 if release_date:
2220 release_date = release_date.replace('-', '')
2221 if not release_year:
2222 release_year = int(release_date[:4])
2223 if release_year:
2224 release_year = int(release_year)
2225
2226 m_episode = re.search(
2227 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2228 video_webpage)
2229 if m_episode:
2230 series = unescapeHTML(m_episode.group('series'))
2231 season_number = int(m_episode.group('season'))
2232 episode_number = int(m_episode.group('episode'))
2233 else:
2234 series = season_number = episode_number = None
2235
2236 m_cat_container = self._search_regex(
2237 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2238 video_webpage, 'categories', default=None)
2239 if m_cat_container:
2240 category = self._html_search_regex(
2241 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2242 default=None)
2243 video_categories = None if category is None else [category]
2244 else:
2245 video_categories = None
2246
2247 video_tags = [
2248 unescapeHTML(m.group('content'))
2249 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2250
2251 def _extract_count(count_name):
2252 return str_to_int(self._search_regex(
2253 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2254 % re.escape(count_name),
2255 video_webpage, count_name, default=None))
2256
2257 like_count = _extract_count('like')
2258 dislike_count = _extract_count('dislike')
2259
2260 if view_count is None:
2261 view_count = str_to_int(self._search_regex(
2262 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2263 'view count', default=None))
2264
2265 average_rating = (
2266 float_or_none(video_details.get('averageRating'))
2267 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2268
2269 # subtitles
2270 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2271 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2272
2273 video_duration = try_get(
2274 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2275 if not video_duration:
2276 video_duration = int_or_none(video_details.get('lengthSeconds'))
2277 if not video_duration:
2278 video_duration = parse_duration(self._html_search_meta(
2279 'duration', video_webpage, 'video duration'))
2280
2281 # annotations
2282 video_annotations = None
2283 if self._downloader.params.get('writeannotations', False):
2284 xsrf_token = self._search_regex(
2285 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2286 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2287 invideo_url = try_get(
2288 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2289 if xsrf_token and invideo_url:
2290 xsrf_field_name = self._search_regex(
2291 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2292 video_webpage, 'xsrf field name',
2293 group='xsrf_field_name', default='session_token')
2294 video_annotations = self._download_webpage(
2295 self._proto_relative_url(invideo_url),
2296 video_id, note='Downloading annotations',
2297 errnote='Unable to download video annotations', fatal=False,
2298 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2299
2300 chapters = self._extract_chapters(description_original, video_duration)
2301
2302 # Look for the DASH manifest
2303 if self._downloader.params.get('youtube_include_dash_manifest', True):
2304 dash_mpd_fatal = True
2305 for mpd_url in dash_mpds:
2306 dash_formats = {}
2307 try:
2308 def decrypt_sig(mobj):
2309 s = mobj.group(1)
2310 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2311 return '/signature/%s' % dec_s
2312
2313 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2314
2315 for df in self._extract_mpd_formats(
2316 mpd_url, video_id, fatal=dash_mpd_fatal,
2317 formats_dict=self._formats):
2318 if not df.get('filesize'):
2319 df['filesize'] = _extract_filesize(df['url'])
2320 # Do not overwrite DASH format found in some previous DASH manifest
2321 if df['format_id'] not in dash_formats:
2322 dash_formats[df['format_id']] = df
2323 # Additional DASH manifests may end up in HTTP Error 403 therefore
2324 # allow them to fail without bug report message if we already have
2325 # some DASH manifest succeeded. This is temporary workaround to reduce
2326 # burst of bug reports until we figure out the reason and whether it
2327 # can be fixed at all.
2328 dash_mpd_fatal = False
2329 except (ExtractorError, KeyError) as e:
2330 self.report_warning(
2331 'Skipping DASH manifest: %r' % e, video_id)
2332 if dash_formats:
2333 # Remove the formats we found through non-DASH, they
2334 # contain less info and it can be wrong, because we use
2335 # fixed values (for example the resolution). See
2336 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2337 # example.
2338 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2339 formats.extend(dash_formats.values())
2340
2341 # Check for malformed aspect ratio
2342 stretched_m = re.search(
2343 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2344 video_webpage)
2345 if stretched_m:
2346 w = float(stretched_m.group('w'))
2347 h = float(stretched_m.group('h'))
2348 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2349 # We will only process correct ratios.
2350 if w > 0 and h > 0:
2351 ratio = w / h
2352 for f in formats:
2353 if f.get('vcodec') != 'none':
2354 f['stretched_ratio'] = ratio
2355
2356 if not formats:
2357 token = extract_token(video_info)
2358 if not token:
2359 if 'reason' in video_info:
2360 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2361 regions_allowed = self._html_search_meta(
2362 'regionsAllowed', video_webpage, default=None)
2363 countries = regions_allowed.split(',') if regions_allowed else None
2364 self.raise_geo_restricted(
2365 msg=video_info['reason'][0], countries=countries)
2366 reason = video_info['reason'][0]
2367 if 'Invalid parameters' in reason:
2368 unavailable_message = extract_unavailable_message()
2369 if unavailable_message:
2370 reason = unavailable_message
2371 raise ExtractorError(
2372 'YouTube said: %s' % reason,
2373 expected=True, video_id=video_id)
2374 else:
2375 raise ExtractorError(
2376 '"token" parameter not in video info for unknown reason',
2377 video_id=video_id)
2378
2379 if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2380 raise ExtractorError('This video is DRM protected.', expected=True)
2381
2382 self._sort_formats(formats)
2383
2384 self.mark_watched(video_id, video_info, player_response)
2385
2386 return {
2387 'id': video_id,
2388 'uploader': video_uploader,
2389 'uploader_id': video_uploader_id,
2390 'uploader_url': video_uploader_url,
2391 'channel_id': channel_id,
2392 'channel_url': channel_url,
2393 'upload_date': upload_date,
2394 'license': video_license,
2395 'creator': video_creator or artist,
2396 'title': video_title,
2397 'alt_title': video_alt_title or track,
2398 'thumbnail': video_thumbnail,
2399 'description': video_description,
2400 'categories': video_categories,
2401 'tags': video_tags,
2402 'subtitles': video_subtitles,
2403 'automatic_captions': automatic_captions,
2404 'duration': video_duration,
2405 'age_limit': 18 if age_gate else 0,
2406 'annotations': video_annotations,
2407 'chapters': chapters,
2408 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2409 'view_count': view_count,
2410 'like_count': like_count,
2411 'dislike_count': dislike_count,
2412 'average_rating': average_rating,
2413 'formats': formats,
2414 'is_live': is_live,
2415 'start_time': start_time,
2416 'end_time': end_time,
2417 'series': series,
2418 'season_number': season_number,
2419 'episode_number': episode_number,
2420 'track': track,
2421 'artist': artist,
2422 'album': album,
2423 'release_date': release_date,
2424 'release_year': release_year,
2425 }
2426
2427
2428 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2429 IE_DESC = 'YouTube.com playlists'
2430 _VALID_URL = r"""(?x)(?:
2431 (?:https?://)?
2432 (?:\w+\.)?
2433 (?:
2434 (?:
2435 youtube\.com|
2436 invidio\.us
2437 )
2438 /
2439 (?:
2440 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2441 \? (?:.*?[&;])*? (?:p|a|list)=
2442 | p/
2443 )|
2444 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2445 )
2446 (
2447 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2448 # Top tracks, they can also include dots
2449 |(?:MC)[\w\.]*
2450 )
2451 .*
2452 |
2453 (%(playlist_id)s)
2454 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2455 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2456 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2457 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2458 IE_NAME = 'youtube:playlist'
2459 _TESTS = [{
2460 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2461 'info_dict': {
2462 'title': 'ytdl test PL',
2463 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2464 },
2465 'playlist_count': 3,
2466 }, {
2467 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2468 'info_dict': {
2469 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2470 'title': 'YDL_Empty_List',
2471 },
2472 'playlist_count': 0,
2473 'skip': 'This playlist is private',
2474 }, {
2475 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2476 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2477 'info_dict': {
2478 'title': '29C3: Not my department',
2479 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2480 'uploader': 'Christiaan008',
2481 'uploader_id': 'ChRiStIaAn008',
2482 },
2483 'playlist_count': 95,
2484 }, {
2485 'note': 'issue #673',
2486 'url': 'PLBB231211A4F62143',
2487 'info_dict': {
2488 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2489 'id': 'PLBB231211A4F62143',
2490 'uploader': 'Wickydoo',
2491 'uploader_id': 'Wickydoo',
2492 },
2493 'playlist_mincount': 26,
2494 }, {
2495 'note': 'Large playlist',
2496 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2497 'info_dict': {
2498 'title': 'Uploads from Cauchemar',
2499 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2500 'uploader': 'Cauchemar',
2501 'uploader_id': 'Cauchemar89',
2502 },
2503 'playlist_mincount': 799,
2504 }, {
2505 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2506 'info_dict': {
2507 'title': 'YDL_safe_search',
2508 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2509 },
2510 'playlist_count': 2,
2511 'skip': 'This playlist is private',
2512 }, {
2513 'note': 'embedded',
2514 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2515 'playlist_count': 4,
2516 'info_dict': {
2517 'title': 'JODA15',
2518 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2519 'uploader': 'milan',
2520 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2521 }
2522 }, {
2523 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2524 'playlist_mincount': 485,
2525 'info_dict': {
2526 'title': '2018 Chinese New Singles (11/6 updated)',
2527 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2528 'uploader': 'LBK',
2529 'uploader_id': 'sdragonfang',
2530 }
2531 }, {
2532 'note': 'Embedded SWF player',
2533 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2534 'playlist_count': 4,
2535 'info_dict': {
2536 'title': 'JODA7',
2537 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2538 },
2539 'skip': 'This playlist does not exist',
2540 }, {
2541 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2542 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2543 'info_dict': {
2544 'title': 'Uploads from Interstellar Movie',
2545 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2546 'uploader': 'Interstellar Movie',
2547 'uploader_id': 'InterstellarMovie1',
2548 },
2549 'playlist_mincount': 21,
2550 }, {
2551 # Playlist URL that does not actually serve a playlist
2552 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2553 'info_dict': {
2554 'id': 'FqZTN594JQw',
2555 'ext': 'webm',
2556 'title': "Smiley's People 01 detective, Adventure Series, Action",
2557 'uploader': 'STREEM',
2558 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2559 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2560 'upload_date': '20150526',
2561 'license': 'Standard YouTube License',
2562 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2563 'categories': ['People & Blogs'],
2564 'tags': list,
2565 'view_count': int,
2566 'like_count': int,
2567 'dislike_count': int,
2568 },
2569 'params': {
2570 'skip_download': True,
2571 },
2572 'skip': 'This video is not available.',
2573 'add_ie': [YoutubeIE.ie_key()],
2574 }, {
2575 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2576 'info_dict': {
2577 'id': 'yeWKywCrFtk',
2578 'ext': 'mp4',
2579 'title': 'Small Scale Baler and Braiding Rugs',
2580 'uploader': 'Backus-Page House Museum',
2581 'uploader_id': 'backuspagemuseum',
2582 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2583 'upload_date': '20161008',
2584 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2585 'categories': ['Nonprofits & Activism'],
2586 'tags': list,
2587 'like_count': int,
2588 'dislike_count': int,
2589 },
2590 'params': {
2591 'noplaylist': True,
2592 'skip_download': True,
2593 },
2594 }, {
2595 # https://github.com/ytdl-org/youtube-dl/issues/21844
2596 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2597 'info_dict': {
2598 'title': 'Data Analysis with Dr Mike Pound',
2599 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2600 'uploader_id': 'Computerphile',
2601 'uploader': 'Computerphile',
2602 },
2603 'playlist_mincount': 11,
2604 }, {
2605 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2606 'only_matching': True,
2607 }, {
2608 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2609 'only_matching': True,
2610 }, {
2611 # music album playlist
2612 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2613 'only_matching': True,
2614 }, {
2615 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2616 'only_matching': True,
2617 }]
2618
2619 def _real_initialize(self):
2620 self._login()
2621
2622 def extract_videos_from_page(self, page):
2623 ids_in_page = []
2624 titles_in_page = []
2625
2626 for item in re.findall(
2627 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2628 attrs = extract_attributes(item)
2629 video_id = attrs['data-video-id']
2630 video_title = unescapeHTML(attrs.get('data-title'))
2631 if video_title:
2632 video_title = video_title.strip()
2633 ids_in_page.append(video_id)
2634 titles_in_page.append(video_title)
2635
2636 # Fallback with old _VIDEO_RE
2637 self.extract_videos_from_page_impl(
2638 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2639
2640 # Relaxed fallbacks
2641 self.extract_videos_from_page_impl(
2642 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2643 ids_in_page, titles_in_page)
2644 self.extract_videos_from_page_impl(
2645 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2646 ids_in_page, titles_in_page)
2647
2648 return zip(ids_in_page, titles_in_page)
2649
2650 def _extract_mix(self, playlist_id):
2651 # The mixes are generated from a single video
2652 # the id of the playlist is just 'RD' + video_id
2653 ids = []
2654 last_id = playlist_id[-11:]
2655 for n in itertools.count(1):
2656 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2657 webpage = self._download_webpage(
2658 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2659 new_ids = orderedSet(re.findall(
2660 r'''(?xs)data-video-username=".*?".*?
2661 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2662 webpage))
2663 # Fetch new pages until all the videos are repeated, it seems that
2664 # there are always 51 unique videos.
2665 new_ids = [_id for _id in new_ids if _id not in ids]
2666 if not new_ids:
2667 break
2668 ids.extend(new_ids)
2669 last_id = ids[-1]
2670
2671 url_results = self._ids_to_results(ids)
2672
2673 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2674 title_span = (
2675 search_title('playlist-title')
2676 or search_title('title long-title')
2677 or search_title('title'))
2678 title = clean_html(title_span)
2679
2680 return self.playlist_result(url_results, playlist_id, title)
2681
2682 def _extract_playlist(self, playlist_id):
2683 url = self._TEMPLATE_URL % playlist_id
2684 page = self._download_webpage(url, playlist_id)
2685
2686 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2687 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2688 match = match.strip()
2689 # Check if the playlist exists or is private
2690 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2691 if mobj:
2692 reason = mobj.group('reason')
2693 message = 'This playlist %s' % reason
2694 if 'private' in reason:
2695 message += ', use --username or --netrc to access it'
2696 message += '.'
2697 raise ExtractorError(message, expected=True)
2698 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2699 raise ExtractorError(
2700 'Invalid parameters. Maybe URL is incorrect.',
2701 expected=True)
2702 elif re.match(r'[^<]*Choose your language[^<]*', match):
2703 continue
2704 else:
2705 self.report_warning('Youtube gives an alert message: ' + match)
2706
2707 playlist_title = self._html_search_regex(
2708 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2709 page, 'title', default=None)
2710
2711 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2712 uploader = self._search_regex(
2713 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2714 page, 'uploader', default=None)
2715 mobj = re.search(
2716 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2717 page)
2718 if mobj:
2719 uploader_id = mobj.group('uploader_id')
2720 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2721 else:
2722 uploader_id = uploader_url = None
2723
2724 has_videos = True
2725
2726 if not playlist_title:
2727 try:
2728 # Some playlist URLs don't actually serve a playlist (e.g.
2729 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2730 next(self._entries(page, playlist_id))
2731 except StopIteration:
2732 has_videos = False
2733
2734 playlist = self.playlist_result(
2735 self._entries(page, playlist_id), playlist_id, playlist_title)
2736 playlist.update({
2737 'uploader': uploader,
2738 'uploader_id': uploader_id,
2739 'uploader_url': uploader_url,
2740 })
2741
2742 return has_videos, playlist
2743
2744 def _check_download_just_video(self, url, playlist_id):
2745 # Check if it's a video-specific URL
2746 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2747 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2748 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2749 'video id', default=None)
2750 if video_id:
2751 if self._downloader.params.get('noplaylist'):
2752 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2753 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2754 else:
2755 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2756 return video_id, None
2757 return None, None
2758
2759 def _real_extract(self, url):
2760 # Extract playlist id
2761 mobj = re.match(self._VALID_URL, url)
2762 if mobj is None:
2763 raise ExtractorError('Invalid URL: %s' % url)
2764 playlist_id = mobj.group(1) or mobj.group(2)
2765
2766 video_id, video = self._check_download_just_video(url, playlist_id)
2767 if video:
2768 return video
2769
2770 if playlist_id.startswith(('RD', 'UL', 'PU')):
2771 # Mixes require a custom extraction process
2772 return self._extract_mix(playlist_id)
2773
2774 has_videos, playlist = self._extract_playlist(playlist_id)
2775 if has_videos or not video_id:
2776 return playlist
2777
2778 # Some playlist URLs don't actually serve a playlist (see
2779 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2780 # Fallback to plain video extraction if there is a video id
2781 # along with playlist id.
2782 return self.url_result(video_id, 'Youtube', video_id=video_id)
2783
2784
2785 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2786 IE_DESC = 'YouTube.com channels'
2787 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2788 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2789 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2790 IE_NAME = 'youtube:channel'
2791 _TESTS = [{
2792 'note': 'paginated channel',
2793 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2794 'playlist_mincount': 91,
2795 'info_dict': {
2796 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2797 'title': 'Uploads from lex will',
2798 'uploader': 'lex will',
2799 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2800 }
2801 }, {
2802 'note': 'Age restricted channel',
2803 # from https://www.youtube.com/user/DeusExOfficial
2804 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2805 'playlist_mincount': 64,
2806 'info_dict': {
2807 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2808 'title': 'Uploads from Deus Ex',
2809 'uploader': 'Deus Ex',
2810 'uploader_id': 'DeusExOfficial',
2811 },
2812 }, {
2813 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2814 'only_matching': True,
2815 }]
2816
2817 @classmethod
2818 def suitable(cls, url):
2819 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2820 else super(YoutubeChannelIE, cls).suitable(url))
2821
2822 def _build_template_url(self, url, channel_id):
2823 return self._TEMPLATE_URL % channel_id
2824
2825 def _real_extract(self, url):
2826 channel_id = self._match_id(url)
2827
2828 url = self._build_template_url(url, channel_id)
2829
2830 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2831 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2832 # otherwise fallback on channel by page extraction
2833 channel_page = self._download_webpage(
2834 url + '?view=57', channel_id,
2835 'Downloading channel page', fatal=False)
2836 if channel_page is False:
2837 channel_playlist_id = False
2838 else:
2839 channel_playlist_id = self._html_search_meta(
2840 'channelId', channel_page, 'channel id', default=None)
2841 if not channel_playlist_id:
2842 channel_url = self._html_search_meta(
2843 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2844 channel_page, 'channel url', default=None)
2845 if channel_url:
2846 channel_playlist_id = self._search_regex(
2847 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2848 channel_url, 'channel id', default=None)
2849 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2850 playlist_id = 'UU' + channel_playlist_id[2:]
2851 return self.url_result(
2852 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2853
2854 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2855 autogenerated = re.search(r'''(?x)
2856 class="[^"]*?(?:
2857 channel-header-autogenerated-label|
2858 yt-channel-title-autogenerated
2859 )[^"]*"''', channel_page) is not None
2860
2861 if autogenerated:
2862 # The videos are contained in a single page
2863 # the ajax pages can't be used, they are empty
2864 entries = [
2865 self.url_result(
2866 video_id, 'Youtube', video_id=video_id,
2867 video_title=video_title)
2868 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2869 return self.playlist_result(entries, channel_id)
2870
2871 try:
2872 next(self._entries(channel_page, channel_id))
2873 except StopIteration:
2874 alert_message = self._html_search_regex(
2875 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2876 channel_page, 'alert', default=None, group='alert')
2877 if alert_message:
2878 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2879
2880 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2881
2882
2883 class YoutubeUserIE(YoutubeChannelIE):
2884 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2885 _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_-]+)'
2886 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2887 IE_NAME = 'youtube:user'
2888
2889 _TESTS = [{
2890 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2891 'playlist_mincount': 320,
2892 'info_dict': {
2893 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2894 'title': 'Uploads from The Linux Foundation',
2895 'uploader': 'The Linux Foundation',
2896 'uploader_id': 'TheLinuxFoundation',
2897 }
2898 }, {
2899 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2900 # but not https://www.youtube.com/user/12minuteathlete/videos
2901 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2902 'playlist_mincount': 249,
2903 'info_dict': {
2904 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2905 'title': 'Uploads from 12 Minute Athlete',
2906 'uploader': '12 Minute Athlete',
2907 'uploader_id': 'the12minuteathlete',
2908 }
2909 }, {
2910 'url': 'ytuser:phihag',
2911 'only_matching': True,
2912 }, {
2913 'url': 'https://www.youtube.com/c/gametrailers',
2914 'only_matching': True,
2915 }, {
2916 'url': 'https://www.youtube.com/gametrailers',
2917 'only_matching': True,
2918 }, {
2919 # This channel is not available, geo restricted to JP
2920 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2921 'only_matching': True,
2922 }]
2923
2924 @classmethod
2925 def suitable(cls, url):
2926 # Don't return True if the url can be extracted with other youtube
2927 # extractor, the regex would is too permissive and it would match.
2928 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2929 if any(ie.suitable(url) for ie in other_yt_ies):
2930 return False
2931 else:
2932 return super(YoutubeUserIE, cls).suitable(url)
2933
2934 def _build_template_url(self, url, channel_id):
2935 mobj = re.match(self._VALID_URL, url)
2936 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2937
2938
2939 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2940 IE_DESC = 'YouTube.com live streams'
2941 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2942 IE_NAME = 'youtube:live'
2943
2944 _TESTS = [{
2945 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2946 'info_dict': {
2947 'id': 'a48o2S1cPoo',
2948 'ext': 'mp4',
2949 'title': 'The Young Turks - Live Main Show',
2950 'uploader': 'The Young Turks',
2951 'uploader_id': 'TheYoungTurks',
2952 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2953 'upload_date': '20150715',
2954 'license': 'Standard YouTube License',
2955 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2956 'categories': ['News & Politics'],
2957 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2958 'like_count': int,
2959 'dislike_count': int,
2960 },
2961 'params': {
2962 'skip_download': True,
2963 },
2964 }, {
2965 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2966 'only_matching': True,
2967 }, {
2968 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2969 'only_matching': True,
2970 }, {
2971 'url': 'https://www.youtube.com/TheYoungTurks/live',
2972 'only_matching': True,
2973 }]
2974
2975 def _real_extract(self, url):
2976 mobj = re.match(self._VALID_URL, url)
2977 channel_id = mobj.group('id')
2978 base_url = mobj.group('base_url')
2979 webpage = self._download_webpage(url, channel_id, fatal=False)
2980 if webpage:
2981 page_type = self._og_search_property(
2982 'type', webpage, 'page type', default='')
2983 video_id = self._html_search_meta(
2984 'videoId', webpage, 'video id', default=None)
2985 if page_type.startswith('video') and video_id and re.match(
2986 r'^[0-9A-Za-z_-]{11}$', video_id):
2987 return self.url_result(video_id, YoutubeIE.ie_key())
2988 return self.url_result(base_url)
2989
2990
2991 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2992 IE_DESC = 'YouTube.com user/channel playlists'
2993 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2994 IE_NAME = 'youtube:playlists'
2995
2996 _TESTS = [{
2997 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2998 'playlist_mincount': 4,
2999 'info_dict': {
3000 'id': 'ThirstForScience',
3001 'title': 'ThirstForScience',
3002 },
3003 }, {
3004 # with "Load more" button
3005 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3006 'playlist_mincount': 70,
3007 'info_dict': {
3008 'id': 'igorkle1',
3009 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
3010 },
3011 }, {
3012 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3013 'playlist_mincount': 17,
3014 'info_dict': {
3015 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3016 'title': 'Chem Player',
3017 },
3018 'skip': 'Blocked',
3019 }]
3020
3021
3022 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3023 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3024
3025
3026 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3027 IE_DESC = 'YouTube.com searches'
3028 # there doesn't appear to be a real limit, for example if you search for
3029 # 'python' you get more than 8.000.000 results
3030 _MAX_RESULTS = float('inf')
3031 IE_NAME = 'youtube:search'
3032 _SEARCH_KEY = 'ytsearch'
3033 _EXTRA_QUERY_ARGS = {}
3034 _TESTS = []
3035
3036 def _get_n_results(self, query, n):
3037 """Get a specified number of results for a query"""
3038
3039 videos = []
3040 limit = n
3041
3042 url_query = {
3043 'search_query': query.encode('utf-8'),
3044 }
3045 url_query.update(self._EXTRA_QUERY_ARGS)
3046 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3047
3048 for pagenum in itertools.count(1):
3049 data = self._download_json(
3050 result_url, video_id='query "%s"' % query,
3051 note='Downloading page %s' % pagenum,
3052 errnote='Unable to download API page',
3053 query={'spf': 'navigate'})
3054 html_content = data[1]['body']['content']
3055
3056 if 'class="search-message' in html_content:
3057 raise ExtractorError(
3058 '[youtube] No video results', expected=True)
3059
3060 new_videos = list(self._process_page(html_content))
3061 videos += new_videos
3062 if not new_videos or len(videos) > limit:
3063 break
3064 next_link = self._html_search_regex(
3065 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3066 html_content, 'next link', default=None)
3067 if next_link is None:
3068 break
3069 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3070
3071 if len(videos) > n:
3072 videos = videos[:n]
3073 return self.playlist_result(videos, query)
3074
3075
3076 class YoutubeSearchDateIE(YoutubeSearchIE):
3077 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3078 _SEARCH_KEY = 'ytsearchdate'
3079 IE_DESC = 'YouTube.com searches, newest videos first'
3080 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3081
3082
3083 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3084 IE_DESC = 'YouTube.com search URLs'
3085 IE_NAME = 'youtube:search_url'
3086 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3087 _TESTS = [{
3088 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3089 'playlist_mincount': 5,
3090 'info_dict': {
3091 'title': 'youtube-dl test video',
3092 }
3093 }, {
3094 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3095 'only_matching': True,
3096 }]
3097
3098 def _real_extract(self, url):
3099 mobj = re.match(self._VALID_URL, url)
3100 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3101 webpage = self._download_webpage(url, query)
3102 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3103
3104
3105 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3106 IE_DESC = 'YouTube.com (multi-season) shows'
3107 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3108 IE_NAME = 'youtube:show'
3109 _TESTS = [{
3110 'url': 'https://www.youtube.com/show/airdisasters',
3111 'playlist_mincount': 5,
3112 'info_dict': {
3113 'id': 'airdisasters',
3114 'title': 'Air Disasters',
3115 }
3116 }]
3117
3118 def _real_extract(self, url):
3119 playlist_id = self._match_id(url)
3120 return super(YoutubeShowIE, self)._real_extract(
3121 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3122
3123
3124 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3125 """
3126 Base class for feed extractors
3127 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3128 """
3129 _LOGIN_REQUIRED = True
3130
3131 @property
3132 def IE_NAME(self):
3133 return 'youtube:%s' % self._FEED_NAME
3134
3135 def _real_initialize(self):
3136 self._login()
3137
3138 def _entries(self, page):
3139 # The extraction process is the same as for playlists, but the regex
3140 # for the video ids doesn't contain an index
3141 ids = []
3142 more_widget_html = content_html = page
3143 for page_num in itertools.count(1):
3144 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3145
3146 # 'recommended' feed has infinite 'load more' and each new portion spins
3147 # the same videos in (sometimes) slightly different order, so we'll check
3148 # for unicity and break when portion has no new videos
3149 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3150 if not new_ids:
3151 break
3152
3153 ids.extend(new_ids)
3154
3155 for entry in self._ids_to_results(new_ids):
3156 yield entry
3157
3158 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3159 if not mobj:
3160 break
3161
3162 more = self._download_json(
3163 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3164 'Downloading page #%s' % page_num,
3165 transform_source=uppercase_escape)
3166 content_html = more['content_html']
3167 more_widget_html = more['load_more_widget_html']
3168
3169 def _real_extract(self, url):
3170 page = self._download_webpage(
3171 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3172 self._PLAYLIST_TITLE)
3173 return self.playlist_result(
3174 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3175
3176
3177 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3178 IE_NAME = 'youtube:watchlater'
3179 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3180 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3181
3182 _TESTS = [{
3183 'url': 'https://www.youtube.com/playlist?list=WL',
3184 'only_matching': True,
3185 }, {
3186 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3187 'only_matching': True,
3188 }]
3189
3190 def _real_extract(self, url):
3191 _, video = self._check_download_just_video(url, 'WL')
3192 if video:
3193 return video
3194 _, playlist = self._extract_playlist('WL')
3195 return playlist
3196
3197
3198 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3199 IE_NAME = 'youtube:favorites'
3200 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3201 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3202 _LOGIN_REQUIRED = True
3203
3204 def _real_extract(self, url):
3205 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3206 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3207 return self.url_result(playlist_id, 'YoutubePlaylist')
3208
3209
3210 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3211 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3212 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3213 _FEED_NAME = 'recommended'
3214 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3215
3216
3217 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3218 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3219 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3220 _FEED_NAME = 'subscriptions'
3221 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3222
3223
3224 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3225 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3226 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3227 _FEED_NAME = 'history'
3228 _PLAYLIST_TITLE = 'Youtube History'
3229
3230
3231 class YoutubeTruncatedURLIE(InfoExtractor):
3232 IE_NAME = 'youtube:truncated_url'
3233 IE_DESC = False # Do not list
3234 _VALID_URL = r'''(?x)
3235 (?:https?://)?
3236 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3237 (?:watch\?(?:
3238 feature=[a-z_]+|
3239 annotation_id=annotation_[^&]+|
3240 x-yt-cl=[0-9]+|
3241 hl=[^&]*|
3242 t=[0-9]+
3243 )?
3244 |
3245 attribution_link\?a=[^&]+
3246 )
3247 $
3248 '''
3249
3250 _TESTS = [{
3251 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3252 'only_matching': True,
3253 }, {
3254 'url': 'https://www.youtube.com/watch?',
3255 'only_matching': True,
3256 }, {
3257 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3258 'only_matching': True,
3259 }, {
3260 'url': 'https://www.youtube.com/watch?feature=foo',
3261 'only_matching': True,
3262 }, {
3263 'url': 'https://www.youtube.com/watch?hl=en-GB',
3264 'only_matching': True,
3265 }, {
3266 'url': 'https://www.youtube.com/watch?t=2372',
3267 'only_matching': True,
3268 }]
3269
3270 def _real_extract(self, url):
3271 raise ExtractorError(
3272 'Did you forget to quote the URL? Remember that & is a meta '
3273 'character in most shells, so you want to put the URL in quotes, '
3274 'like youtube-dl '
3275 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3276 ' or simply youtube-dl BaW_jenozKc .',
3277 expected=True)
3278
3279
3280 class YoutubeTruncatedIDIE(InfoExtractor):
3281 IE_NAME = 'youtube:truncated_id'
3282 IE_DESC = False # Do not list
3283 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3284
3285 _TESTS = [{
3286 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3287 'only_matching': True,
3288 }]
3289
3290 def _real_extract(self, url):
3291 video_id = self._match_id(url)
3292 raise ExtractorError(
3293 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3294 expected=True)