]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
Import Debian changes 2020.01.24-0.1
[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 remove_quotes,
45 remove_start,
46 smuggle_url,
47 str_or_none,
48 str_to_int,
49 try_get,
50 unescapeHTML,
51 unified_strdate,
52 unsmuggle_url,
53 uppercase_escape,
54 url_or_none,
55 urlencode_postdata,
56 )
57
58
59 class YoutubeBaseInfoExtractor(InfoExtractor):
60 """Provide base functions for Youtube extractors"""
61 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
62 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
63
64 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
65 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
66 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
67
68 _NETRC_MACHINE = 'youtube'
69 # If True it will raise an error if no login info is provided
70 _LOGIN_REQUIRED = False
71
72 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}'
73
74 def _set_language(self):
75 self._set_cookie(
76 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
77 # YouTube sets the expire time to about two months
78 expire_time=time.time() + 2 * 30 * 24 * 3600)
79
80 def _ids_to_results(self, ids):
81 return [
82 self.url_result(vid_id, 'Youtube', video_id=vid_id)
83 for vid_id in ids]
84
85 def _login(self):
86 """
87 Attempt to log in to YouTube.
88 True is returned if successful or skipped.
89 False is returned if login failed.
90
91 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
92 """
93 username, password = self._get_login_info()
94 # No authentication to be performed
95 if username is None:
96 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
97 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
98 return True
99
100 login_page = self._download_webpage(
101 self._LOGIN_URL, None,
102 note='Downloading login page',
103 errnote='unable to fetch login page', fatal=False)
104 if login_page is False:
105 return
106
107 login_form = self._hidden_inputs(login_page)
108
109 def req(url, f_req, note, errnote):
110 data = login_form.copy()
111 data.update({
112 'pstMsg': 1,
113 'checkConnection': 'youtube',
114 'checkedDomains': 'youtube',
115 'hl': 'en',
116 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
117 'f.req': json.dumps(f_req),
118 'flowName': 'GlifWebSignIn',
119 'flowEntry': 'ServiceLogin',
120 # TODO: reverse actual botguard identifier generation algo
121 'bgRequest': '["identifier",""]',
122 })
123 return self._download_json(
124 url, None, note=note, errnote=errnote,
125 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
126 fatal=False,
127 data=urlencode_postdata(data), headers={
128 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
129 'Google-Accounts-XSRF': 1,
130 })
131
132 def warn(message):
133 self._downloader.report_warning(message)
134
135 lookup_req = [
136 username,
137 None, [], None, 'US', None, None, 2, False, True,
138 [
139 None, None,
140 [2, 1, None, 1,
141 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
142 None, [], 4],
143 1, [None, None, []], None, None, None, True
144 ],
145 username,
146 ]
147
148 lookup_results = req(
149 self._LOOKUP_URL, lookup_req,
150 'Looking up account info', 'Unable to look up account info')
151
152 if lookup_results is False:
153 return False
154
155 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
156 if not user_hash:
157 warn('Unable to extract user hash')
158 return False
159
160 challenge_req = [
161 user_hash,
162 None, 1, None, [1, None, None, None, [password, None, True]],
163 [
164 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
165 1, [None, None, []], None, None, None, True
166 ]]
167
168 challenge_results = req(
169 self._CHALLENGE_URL, challenge_req,
170 'Logging in', 'Unable to log in')
171
172 if challenge_results is False:
173 return
174
175 login_res = try_get(challenge_results, lambda x: x[0][5], list)
176 if login_res:
177 login_msg = try_get(login_res, lambda x: x[5], compat_str)
178 warn(
179 'Unable to login: %s' % 'Invalid password'
180 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
181 return False
182
183 res = try_get(challenge_results, lambda x: x[0][-1], list)
184 if not res:
185 warn('Unable to extract result entry')
186 return False
187
188 login_challenge = try_get(res, lambda x: x[0][0], list)
189 if login_challenge:
190 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
191 if challenge_str == 'TWO_STEP_VERIFICATION':
192 # SEND_SUCCESS - TFA code has been successfully sent to phone
193 # QUOTA_EXCEEDED - reached the limit of TFA codes
194 status = try_get(login_challenge, lambda x: x[5], compat_str)
195 if status == 'QUOTA_EXCEEDED':
196 warn('Exceeded the limit of TFA codes, try later')
197 return False
198
199 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
200 if not tl:
201 warn('Unable to extract TL')
202 return False
203
204 tfa_code = self._get_tfa_info('2-step verification code')
205
206 if not tfa_code:
207 warn(
208 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
209 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
210 return False
211
212 tfa_code = remove_start(tfa_code, 'G-')
213
214 tfa_req = [
215 user_hash, None, 2, None,
216 [
217 9, None, None, None, None, None, None, None,
218 [None, tfa_code, True, 2]
219 ]]
220
221 tfa_results = req(
222 self._TFA_URL.format(tl), tfa_req,
223 'Submitting TFA code', 'Unable to submit TFA code')
224
225 if tfa_results is False:
226 return False
227
228 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
229 if tfa_res:
230 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
231 warn(
232 'Unable to finish TFA: %s' % 'Invalid TFA code'
233 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
234 return False
235
236 check_cookie_url = try_get(
237 tfa_results, lambda x: x[0][-1][2], compat_str)
238 else:
239 CHALLENGES = {
240 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
241 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
242 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
243 }
244 challenge = CHALLENGES.get(
245 challenge_str,
246 '%s returned error %s.' % (self.IE_NAME, challenge_str))
247 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
248 return False
249 else:
250 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
251
252 if not check_cookie_url:
253 warn('Unable to extract CheckCookie URL')
254 return False
255
256 check_cookie_results = self._download_webpage(
257 check_cookie_url, None, 'Checking cookie', fatal=False)
258
259 if check_cookie_results is False:
260 return False
261
262 if 'https://myaccount.google.com/' not in check_cookie_results:
263 warn('Unable to log in')
264 return False
265
266 return True
267
268 def _download_webpage_handle(self, *args, **kwargs):
269 query = kwargs.get('query', {}).copy()
270 query['disable_polymer'] = 'true'
271 kwargs['query'] = query
272 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
273 *args, **compat_kwargs(kwargs))
274
275 def _real_initialize(self):
276 if self._downloader is None:
277 return
278 self._set_language()
279 if not self._login():
280 return
281
282
283 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
284 # Extract entries from page with "Load more" button
285 def _entries(self, page, playlist_id):
286 more_widget_html = content_html = page
287 for page_num in itertools.count(1):
288 for entry in self._process_page(content_html):
289 yield entry
290
291 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
292 if not mobj:
293 break
294
295 count = 0
296 retries = 3
297 while count <= retries:
298 try:
299 # Downloading page may result in intermittent 5xx HTTP error
300 # that is usually worked around with a retry
301 more = self._download_json(
302 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
303 'Downloading page #%s%s'
304 % (page_num, ' (retry #%d)' % count if count else ''),
305 transform_source=uppercase_escape)
306 break
307 except ExtractorError as e:
308 if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
309 count += 1
310 if count <= retries:
311 continue
312 raise
313
314 content_html = more['content_html']
315 if not content_html.strip():
316 # Some webpages show a "Load more" button but they don't
317 # have more videos
318 break
319 more_widget_html = more['load_more_widget_html']
320
321
322 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
323 def _process_page(self, content):
324 for video_id, video_title in self.extract_videos_from_page(content):
325 yield self.url_result(video_id, 'Youtube', video_id, video_title)
326
327 def extract_videos_from_page_impl(self, video_re, page, ids_in_page, titles_in_page):
328 for mobj in re.finditer(video_re, page):
329 # The link with index 0 is not the first video of the playlist (not sure if still actual)
330 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
331 continue
332 video_id = mobj.group('id')
333 video_title = unescapeHTML(
334 mobj.group('title')) if 'title' in mobj.groupdict() else None
335 if video_title:
336 video_title = video_title.strip()
337 if video_title == 'ā–ŗ Play all':
338 video_title = None
339 try:
340 idx = ids_in_page.index(video_id)
341 if video_title and not titles_in_page[idx]:
342 titles_in_page[idx] = video_title
343 except ValueError:
344 ids_in_page.append(video_id)
345 titles_in_page.append(video_title)
346
347 def extract_videos_from_page(self, page):
348 ids_in_page = []
349 titles_in_page = []
350 self.extract_videos_from_page_impl(
351 self._VIDEO_RE, page, ids_in_page, titles_in_page)
352 return zip(ids_in_page, titles_in_page)
353
354
355 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
356 def _process_page(self, content):
357 for playlist_id in orderedSet(re.findall(
358 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
359 content)):
360 yield self.url_result(
361 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
362
363 def _real_extract(self, url):
364 playlist_id = self._match_id(url)
365 webpage = self._download_webpage(url, playlist_id)
366 title = self._og_search_title(webpage, fatal=False)
367 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
368
369
370 class YoutubeIE(YoutubeBaseInfoExtractor):
371 IE_DESC = 'YouTube.com'
372 _VALID_URL = r"""(?x)^
373 (
374 (?:https?://|//) # http(s):// or protocol-independent URL
375 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/|
376 (?:www\.)?deturl\.com/www\.youtube\.com/|
377 (?:www\.)?pwnyoutube\.com/|
378 (?:www\.)?hooktube\.com/|
379 (?:www\.)?yourepeat\.com/|
380 tube\.majestyc\.net/|
381 # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances
382 (?:(?:www|dev)\.)?invidio\.us/|
383 (?:(?:www|no)\.)?invidiou\.sh/|
384 (?:(?:www|fi|de)\.)?invidious\.snopyta\.org/|
385 (?:www\.)?invidious\.kabi\.tk/|
386 (?:www\.)?invidious\.13ad\.de/|
387 (?:www\.)?invidious\.mastodon\.host/|
388 (?:www\.)?invidious\.nixnet\.xyz/|
389 (?:www\.)?invidious\.drycat\.fr/|
390 (?:www\.)?tube\.poal\.co/|
391 (?:www\.)?vid\.wxzm\.sx/|
392 (?:www\.)?yt\.elukerio\.org/|
393 (?:www\.)?yt\.lelux\.fi/|
394 (?:www\.)?kgg2m7yk5aybusll\.onion/|
395 (?:www\.)?qklhadlycap4cnod\.onion/|
396 (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/|
397 (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/|
398 (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/|
399 (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/|
400 (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/|
401 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
402 (?:.*?\#/)? # handle anchor (#/) redirect urls
403 (?: # the various things that can precede the ID:
404 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
405 |(?: # or the v= param in all its forms
406 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
407 (?:\?|\#!?) # the params delimiter ? or # or #!
408 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
409 v=
410 )
411 ))
412 |(?:
413 youtu\.be| # just youtu.be/xxxx
414 vid\.plus| # or vid.plus/xxxx
415 zwearz\.com/watch| # or zwearz.com/watch/xxxx
416 )/
417 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
418 )
419 )? # all until now is optional -> you can pass the naked ID
420 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
421 (?!.*?\blist=
422 (?:
423 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
424 WL # WL are handled by the watch later IE
425 )
426 )
427 (?(1).+)? # if we found the ID, everything can follow
428 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
429 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
430 _formats = {
431 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
432 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
433 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
434 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
435 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
436 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
437 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
438 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
439 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
440 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
441 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
442 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
443 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
444 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
445 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
446 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
447 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
448 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
449
450
451 # 3D videos
452 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
453 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
454 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
455 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
456 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
457 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
458 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
459
460 # Apple HTTP Live Streaming
461 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
462 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
463 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
464 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
465 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
466 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
467 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
468 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
469
470 # DASH mp4 video
471 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
472 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
473 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
474 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
475 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
476 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
477 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
478 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
479 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
480 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
481 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
482 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
483
484 # Dash mp4 audio
485 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
486 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
487 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
488 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
489 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
490 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
491 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
492
493 # Dash webm
494 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
495 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
496 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
497 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
498 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
499 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
500 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
501 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
502 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
503 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
504 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
505 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
506 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
507 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
508 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
509 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
510 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
511 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
512 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
513 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
514 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
515 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
516
517 # Dash webm audio
518 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
519 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
520
521 # Dash webm audio with opus inside
522 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
523 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
524 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
525
526 # RTMP (unnamed)
527 '_rtmp': {'protocol': 'rtmp'},
528
529 # av01 video only formats sometimes served with "unknown" codecs
530 '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
531 '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
532 '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
533 '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
534 }
535 _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
536
537 _GEO_BYPASS = False
538
539 IE_NAME = 'youtube'
540 _TESTS = [
541 {
542 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
543 'info_dict': {
544 'id': 'BaW_jenozKc',
545 'ext': 'mp4',
546 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
547 'uploader': 'Philipp Hagemeister',
548 'uploader_id': 'phihag',
549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
550 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
551 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
552 'upload_date': '20121002',
553 '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 .',
554 'categories': ['Science & Technology'],
555 'tags': ['youtube-dl'],
556 'duration': 10,
557 'view_count': int,
558 'like_count': int,
559 'dislike_count': int,
560 'start_time': 1,
561 'end_time': 9,
562 }
563 },
564 {
565 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
566 'note': 'Test generic use_cipher_signature video (#897)',
567 'info_dict': {
568 'id': 'UxxajLWwzqY',
569 'ext': 'mp4',
570 'upload_date': '20120506',
571 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
572 'alt_title': 'I Love It (feat. Charli XCX)',
573 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
574 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
575 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
576 'iconic ep', 'iconic', 'love', 'it'],
577 'duration': 180,
578 'uploader': 'Icona Pop',
579 'uploader_id': 'IconaPop',
580 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
581 'creator': 'Icona Pop',
582 'track': 'I Love It (feat. Charli XCX)',
583 'artist': 'Icona Pop',
584 }
585 },
586 {
587 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
588 'note': 'Test VEVO video with age protection (#956)',
589 'info_dict': {
590 'id': '07FYdnEawAQ',
591 'ext': 'mp4',
592 'upload_date': '20130703',
593 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
594 'alt_title': 'Tunnel Vision',
595 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
596 'duration': 419,
597 'uploader': 'justintimberlakeVEVO',
598 'uploader_id': 'justintimberlakeVEVO',
599 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
600 'creator': 'Justin Timberlake',
601 'track': 'Tunnel Vision',
602 'artist': 'Justin Timberlake',
603 'age_limit': 18,
604 }
605 },
606 {
607 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
608 'note': 'Embed-only video (#1746)',
609 'info_dict': {
610 'id': 'yZIXLfi8CZQ',
611 'ext': 'mp4',
612 'upload_date': '20120608',
613 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
614 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
615 'uploader': 'SET India',
616 'uploader_id': 'setindia',
617 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
618 'age_limit': 18,
619 }
620 },
621 {
622 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
623 'note': 'Use the first video ID in the URL',
624 'info_dict': {
625 'id': 'BaW_jenozKc',
626 'ext': 'mp4',
627 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
628 'uploader': 'Philipp Hagemeister',
629 'uploader_id': 'phihag',
630 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
631 'upload_date': '20121002',
632 '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 .',
633 'categories': ['Science & Technology'],
634 'tags': ['youtube-dl'],
635 'duration': 10,
636 'view_count': int,
637 'like_count': int,
638 'dislike_count': int,
639 },
640 'params': {
641 'skip_download': True,
642 },
643 },
644 {
645 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
646 'note': '256k DASH audio (format 141) via DASH manifest',
647 'info_dict': {
648 'id': 'a9LDPn-MO4I',
649 'ext': 'm4a',
650 'upload_date': '20121002',
651 'uploader_id': '8KVIDEO',
652 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
653 'description': '',
654 'uploader': '8KVIDEO',
655 'title': 'UHDTV TEST 8K VIDEO.mp4'
656 },
657 'params': {
658 'youtube_include_dash_manifest': True,
659 'format': '141',
660 },
661 'skip': 'format 141 not served anymore',
662 },
663 # DASH manifest with encrypted signature
664 {
665 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
666 'info_dict': {
667 'id': 'IB3lcPjvWLA',
668 'ext': 'm4a',
669 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
670 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
671 'duration': 244,
672 'uploader': 'AfrojackVEVO',
673 'uploader_id': 'AfrojackVEVO',
674 'upload_date': '20131011',
675 },
676 'params': {
677 'youtube_include_dash_manifest': True,
678 'format': '141/bestaudio[ext=m4a]',
679 },
680 },
681 # JS player signature function name containing $
682 {
683 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
684 'info_dict': {
685 'id': 'nfWlot6h_JM',
686 'ext': 'm4a',
687 'title': 'Taylor Swift - Shake It Off',
688 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
689 'duration': 242,
690 'uploader': 'TaylorSwiftVEVO',
691 'uploader_id': 'TaylorSwiftVEVO',
692 'upload_date': '20140818',
693 'creator': 'Taylor Swift',
694 },
695 'params': {
696 'youtube_include_dash_manifest': True,
697 'format': '141/bestaudio[ext=m4a]',
698 },
699 },
700 # Controversy video
701 {
702 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
703 'info_dict': {
704 'id': 'T4XJQO3qol8',
705 'ext': 'mp4',
706 'duration': 219,
707 'upload_date': '20100909',
708 'uploader': 'Amazing Atheist',
709 'uploader_id': 'TheAmazingAtheist',
710 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
711 'title': 'Burning Everyone\'s Koran',
712 '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',
713 }
714 },
715 # Normal age-gate video (No vevo, embed allowed)
716 {
717 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
718 'info_dict': {
719 'id': 'HtVdAasjOgU',
720 'ext': 'mp4',
721 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
722 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
723 'duration': 142,
724 'uploader': 'The Witcher',
725 'uploader_id': 'WitcherGame',
726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
727 'upload_date': '20140605',
728 'age_limit': 18,
729 },
730 },
731 # Age-gate video with encrypted signature
732 {
733 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
734 'info_dict': {
735 'id': '6kLq3WMV1nU',
736 'ext': 'mp4',
737 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
738 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
739 'duration': 246,
740 'uploader': 'LloydVEVO',
741 'uploader_id': 'LloydVEVO',
742 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
743 'upload_date': '20110629',
744 'age_limit': 18,
745 },
746 },
747 # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
748 # YouTube Red ad is not captured for creator
749 {
750 'url': '__2ABJjxzNo',
751 'info_dict': {
752 'id': '__2ABJjxzNo',
753 'ext': 'mp4',
754 'duration': 266,
755 'upload_date': '20100430',
756 'uploader_id': 'deadmau5',
757 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
758 'creator': 'deadmau5',
759 'description': 'md5:12c56784b8032162bb936a5f76d55360',
760 'uploader': 'deadmau5',
761 'title': 'Deadmau5 - Some Chords (HD)',
762 'alt_title': 'Some Chords',
763 },
764 'expected_warnings': [
765 'DASH manifest missing',
766 ]
767 },
768 # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
769 {
770 'url': 'lqQg6PlCWgI',
771 'info_dict': {
772 'id': 'lqQg6PlCWgI',
773 'ext': 'mp4',
774 'duration': 6085,
775 'upload_date': '20150827',
776 'uploader_id': 'olympic',
777 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
778 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
779 'uploader': 'Olympic',
780 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
781 },
782 'params': {
783 'skip_download': 'requires avconv',
784 }
785 },
786 # Non-square pixels
787 {
788 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
789 'info_dict': {
790 'id': '_b-2C3KPAM0',
791 'ext': 'mp4',
792 'stretched_ratio': 16 / 9.,
793 'duration': 85,
794 'upload_date': '20110310',
795 'uploader_id': 'AllenMeow',
796 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
797 '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',
798 'uploader': 'å­«į„‹į„…',
799 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
800 },
801 },
802 # url_encoded_fmt_stream_map is empty string
803 {
804 'url': 'qEJwOuvDf7I',
805 'info_dict': {
806 'id': 'qEJwOuvDf7I',
807 'ext': 'webm',
808 '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',
809 'description': '',
810 'upload_date': '20150404',
811 'uploader_id': 'spbelect',
812 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
813 },
814 'params': {
815 'skip_download': 'requires avconv',
816 },
817 'skip': 'This live event has ended.',
818 },
819 # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
820 {
821 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
822 'info_dict': {
823 'id': 'FIl7x6_3R5Y',
824 'ext': 'webm',
825 'title': 'md5:7b81415841e02ecd4313668cde88737a',
826 'description': 'md5:116377fd2963b81ec4ce64b542173306',
827 'duration': 220,
828 'upload_date': '20150625',
829 'uploader_id': 'dorappi2000',
830 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
831 'uploader': 'dorappi2000',
832 'formats': 'mincount:31',
833 },
834 'skip': 'not actual anymore',
835 },
836 # DASH manifest with segment_list
837 {
838 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
839 'md5': '8ce563a1d667b599d21064e982ab9e31',
840 'info_dict': {
841 'id': 'CsmdDsKjzN8',
842 'ext': 'mp4',
843 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
844 'uploader': 'Airtek',
845 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
846 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
847 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
848 },
849 'params': {
850 'youtube_include_dash_manifest': True,
851 'format': '135', # bestvideo
852 },
853 'skip': 'This live event has ended.',
854 },
855 {
856 # Multifeed videos (multiple cameras), URL is for Main Camera
857 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
858 'info_dict': {
859 'id': 'jqWvoWXjCVs',
860 'title': 'teamPGP: Rocket League Noob Stream',
861 'description': 'md5:dc7872fb300e143831327f1bae3af010',
862 },
863 'playlist': [{
864 'info_dict': {
865 'id': 'jqWvoWXjCVs',
866 'ext': 'mp4',
867 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
868 'description': 'md5:dc7872fb300e143831327f1bae3af010',
869 'duration': 7335,
870 'upload_date': '20150721',
871 'uploader': 'Beer Games Beer',
872 'uploader_id': 'beergamesbeer',
873 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
874 'license': 'Standard YouTube License',
875 },
876 }, {
877 'info_dict': {
878 'id': '6h8e8xoXJzg',
879 'ext': 'mp4',
880 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
881 'description': 'md5:dc7872fb300e143831327f1bae3af010',
882 'duration': 7337,
883 'upload_date': '20150721',
884 'uploader': 'Beer Games Beer',
885 'uploader_id': 'beergamesbeer',
886 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
887 'license': 'Standard YouTube License',
888 },
889 }, {
890 'info_dict': {
891 'id': 'PUOgX5z9xZw',
892 'ext': 'mp4',
893 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
894 'description': 'md5:dc7872fb300e143831327f1bae3af010',
895 'duration': 7337,
896 'upload_date': '20150721',
897 'uploader': 'Beer Games Beer',
898 'uploader_id': 'beergamesbeer',
899 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
900 'license': 'Standard YouTube License',
901 },
902 }, {
903 'info_dict': {
904 'id': 'teuwxikvS5k',
905 'ext': 'mp4',
906 'title': 'teamPGP: Rocket League Noob Stream (zim)',
907 'description': 'md5:dc7872fb300e143831327f1bae3af010',
908 'duration': 7334,
909 'upload_date': '20150721',
910 'uploader': 'Beer Games Beer',
911 'uploader_id': 'beergamesbeer',
912 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
913 'license': 'Standard YouTube License',
914 },
915 }],
916 'params': {
917 'skip_download': True,
918 },
919 'skip': 'This video is not available.',
920 },
921 {
922 # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
923 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
924 'info_dict': {
925 'id': 'gVfLd0zydlo',
926 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
927 },
928 'playlist_count': 2,
929 'skip': 'Not multifeed anymore',
930 },
931 {
932 'url': 'https://vid.plus/FlRa-iH7PGw',
933 'only_matching': True,
934 },
935 {
936 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
937 'only_matching': True,
938 },
939 {
940 # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
941 # Also tests cut-off URL expansion in video description (see
942 # https://github.com/ytdl-org/youtube-dl/issues/1892,
943 # https://github.com/ytdl-org/youtube-dl/issues/8164)
944 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
945 'info_dict': {
946 'id': 'lsguqyKfVQg',
947 'ext': 'mp4',
948 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
949 'alt_title': 'Dark Walk - Position Music',
950 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
951 'duration': 133,
952 'upload_date': '20151119',
953 'uploader_id': 'IronSoulElf',
954 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
955 'uploader': 'IronSoulElf',
956 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
957 'track': 'Dark Walk - Position Music',
958 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
959 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
960 },
961 'params': {
962 'skip_download': True,
963 },
964 },
965 {
966 # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
967 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
968 'only_matching': True,
969 },
970 {
971 # Video with yt:stretch=17:0
972 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
973 'info_dict': {
974 'id': 'Q39EVAstoRM',
975 'ext': 'mp4',
976 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
977 'description': 'md5:ee18a25c350637c8faff806845bddee9',
978 'upload_date': '20151107',
979 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
980 'uploader': 'CH GAMER DROID',
981 },
982 'params': {
983 'skip_download': True,
984 },
985 'skip': 'This video does not exist.',
986 },
987 {
988 # Video licensed under Creative Commons
989 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
990 'info_dict': {
991 'id': 'M4gD1WSo5mA',
992 'ext': 'mp4',
993 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
994 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
995 'duration': 721,
996 'upload_date': '20150127',
997 'uploader_id': 'BerkmanCenter',
998 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
999 'uploader': 'The Berkman Klein Center for Internet & Society',
1000 'license': 'Creative Commons Attribution license (reuse allowed)',
1001 },
1002 'params': {
1003 'skip_download': True,
1004 },
1005 },
1006 {
1007 # Channel-like uploader_url
1008 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
1009 'info_dict': {
1010 'id': 'eQcmzGIKrzg',
1011 'ext': 'mp4',
1012 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
1013 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
1014 'duration': 4060,
1015 'upload_date': '20151119',
1016 'uploader': 'Bernie Sanders',
1017 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
1018 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
1019 'license': 'Creative Commons Attribution license (reuse allowed)',
1020 },
1021 'params': {
1022 'skip_download': True,
1023 },
1024 },
1025 {
1026 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1027 'only_matching': True,
1028 },
1029 {
1030 # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1031 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1032 'only_matching': True,
1033 },
1034 {
1035 # Rental video preview
1036 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1037 'info_dict': {
1038 'id': 'uGpuVWrhIzE',
1039 'ext': 'mp4',
1040 'title': 'Piku - Trailer',
1041 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1042 'upload_date': '20150811',
1043 'uploader': 'FlixMatrix',
1044 'uploader_id': 'FlixMatrixKaravan',
1045 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1046 'license': 'Standard YouTube License',
1047 },
1048 'params': {
1049 'skip_download': True,
1050 },
1051 'skip': 'This video is not available.',
1052 },
1053 {
1054 # YouTube Red video with episode data
1055 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1056 'info_dict': {
1057 'id': 'iqKdEhx-dD4',
1058 'ext': 'mp4',
1059 'title': 'Isolation - Mind Field (Ep 1)',
1060 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1061 'duration': 2085,
1062 'upload_date': '20170118',
1063 'uploader': 'Vsauce',
1064 'uploader_id': 'Vsauce',
1065 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1066 'series': 'Mind Field',
1067 'season_number': 1,
1068 'episode_number': 1,
1069 },
1070 'params': {
1071 'skip_download': True,
1072 },
1073 'expected_warnings': [
1074 'Skipping DASH manifest',
1075 ],
1076 },
1077 {
1078 # The following content has been identified by the YouTube community
1079 # as inappropriate or offensive to some audiences.
1080 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1081 'info_dict': {
1082 'id': '6SJNVb0GnPI',
1083 'ext': 'mp4',
1084 'title': 'Race Differences in Intelligence',
1085 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1086 'duration': 965,
1087 'upload_date': '20140124',
1088 'uploader': 'New Century Foundation',
1089 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1090 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1091 },
1092 'params': {
1093 'skip_download': True,
1094 },
1095 },
1096 {
1097 # itag 212
1098 'url': '1t24XAntNCY',
1099 'only_matching': True,
1100 },
1101 {
1102 # geo restricted to JP
1103 'url': 'sJL6WA-aGkQ',
1104 'only_matching': True,
1105 },
1106 {
1107 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1108 'only_matching': True,
1109 },
1110 {
1111 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1112 'only_matching': True,
1113 },
1114 {
1115 # DRM protected
1116 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1117 'only_matching': True,
1118 },
1119 {
1120 # Video with unsupported adaptive stream type formats
1121 'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1122 'info_dict': {
1123 'id': 'Z4Vy8R84T1U',
1124 'ext': 'mp4',
1125 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1126 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1127 'duration': 433,
1128 'upload_date': '20130923',
1129 'uploader': 'Amelia Putri Harwita',
1130 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1131 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1132 'formats': 'maxcount:10',
1133 },
1134 'params': {
1135 'skip_download': True,
1136 'youtube_include_dash_manifest': False,
1137 },
1138 },
1139 {
1140 # Youtube Music Auto-generated description
1141 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1142 'info_dict': {
1143 'id': 'MgNrAu2pzNs',
1144 'ext': 'mp4',
1145 'title': 'Voyeur Girl',
1146 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1147 'upload_date': '20190312',
1148 'uploader': 'Various Artists - Topic',
1149 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1150 'artist': 'Stephen',
1151 'track': 'Voyeur Girl',
1152 'album': 'it\'s too much love to know my dear',
1153 'release_date': '20190313',
1154 'release_year': 2019,
1155 },
1156 'params': {
1157 'skip_download': True,
1158 },
1159 },
1160 {
1161 # Youtube Music Auto-generated description
1162 # Retrieve 'artist' field from 'Artist:' in video description
1163 # when it is present on youtube music video
1164 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1165 'info_dict': {
1166 'id': 'k0jLE7tTwjY',
1167 'ext': 'mp4',
1168 'title': 'Latch Feat. Sam Smith',
1169 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1170 'upload_date': '20150110',
1171 'uploader': 'Various Artists - Topic',
1172 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1173 'artist': 'Disclosure',
1174 'track': 'Latch Feat. Sam Smith',
1175 'album': 'Latch Featuring Sam Smith',
1176 'release_date': '20121008',
1177 'release_year': 2012,
1178 },
1179 'params': {
1180 'skip_download': True,
1181 },
1182 },
1183 {
1184 # Youtube Music Auto-generated description
1185 # handle multiple artists on youtube music video
1186 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1187 'info_dict': {
1188 'id': '74qn0eJSjpA',
1189 'ext': 'mp4',
1190 'title': 'Eastside',
1191 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1192 'upload_date': '20180710',
1193 'uploader': 'Benny Blanco - Topic',
1194 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1195 'artist': 'benny blanco, Halsey, Khalid',
1196 'track': 'Eastside',
1197 'album': 'Eastside',
1198 'release_date': '20180713',
1199 'release_year': 2018,
1200 },
1201 'params': {
1202 'skip_download': True,
1203 },
1204 },
1205 {
1206 # Youtube Music Auto-generated description
1207 # handle youtube music video with release_year and no release_date
1208 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1209 'info_dict': {
1210 'id': '-hcAI0g-f5M',
1211 'ext': 'mp4',
1212 'title': 'Put It On Me',
1213 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1214 'upload_date': '20180426',
1215 'uploader': 'Matt Maeson - Topic',
1216 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1217 'artist': 'Matt Maeson',
1218 'track': 'Put It On Me',
1219 'album': 'The Hearse',
1220 'release_date': None,
1221 'release_year': 2018,
1222 },
1223 'params': {
1224 'skip_download': True,
1225 },
1226 },
1227 {
1228 'url': 'https://www.youtubekids.com/watch?v=3b8nCWDgZ6Q',
1229 'only_matching': True,
1230 },
1231 ]
1232
1233 def __init__(self, *args, **kwargs):
1234 super(YoutubeIE, self).__init__(*args, **kwargs)
1235 self._player_cache = {}
1236
1237 def report_video_info_webpage_download(self, video_id):
1238 """Report attempt to download video info webpage."""
1239 self.to_screen('%s: Downloading video info webpage' % video_id)
1240
1241 def report_information_extraction(self, video_id):
1242 """Report attempt to extract video information."""
1243 self.to_screen('%s: Extracting video information' % video_id)
1244
1245 def report_unavailable_format(self, video_id, format):
1246 """Report extracted video URL."""
1247 self.to_screen('%s: Format %s not available' % (video_id, format))
1248
1249 def report_rtmp_download(self):
1250 """Indicate the download will use the RTMP protocol."""
1251 self.to_screen('RTMP download detected')
1252
1253 def _signature_cache_id(self, example_sig):
1254 """ Return a string representation of a signature """
1255 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1256
1257 def _extract_signature_function(self, video_id, player_url, example_sig):
1258 id_m = re.match(
1259 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1260 player_url)
1261 if not id_m:
1262 raise ExtractorError('Cannot identify player %r' % player_url)
1263 player_type = id_m.group('ext')
1264 player_id = id_m.group('id')
1265
1266 # Read from filesystem cache
1267 func_id = '%s_%s_%s' % (
1268 player_type, player_id, self._signature_cache_id(example_sig))
1269 assert os.path.basename(func_id) == func_id
1270
1271 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1272 if cache_spec is not None:
1273 return lambda s: ''.join(s[i] for i in cache_spec)
1274
1275 download_note = (
1276 'Downloading player %s' % player_url
1277 if self._downloader.params.get('verbose') else
1278 'Downloading %s player %s' % (player_type, player_id)
1279 )
1280 if player_type == 'js':
1281 code = self._download_webpage(
1282 player_url, video_id,
1283 note=download_note,
1284 errnote='Download of %s failed' % player_url)
1285 res = self._parse_sig_js(code)
1286 elif player_type == 'swf':
1287 urlh = self._request_webpage(
1288 player_url, video_id,
1289 note=download_note,
1290 errnote='Download of %s failed' % player_url)
1291 code = urlh.read()
1292 res = self._parse_sig_swf(code)
1293 else:
1294 assert False, 'Invalid player type %r' % player_type
1295
1296 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1297 cache_res = res(test_string)
1298 cache_spec = [ord(c) for c in cache_res]
1299
1300 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1301 return res
1302
1303 def _print_sig_code(self, func, example_sig):
1304 def gen_sig_code(idxs):
1305 def _genslice(start, end, step):
1306 starts = '' if start == 0 else str(start)
1307 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1308 steps = '' if step == 1 else (':%d' % step)
1309 return 's[%s%s%s]' % (starts, ends, steps)
1310
1311 step = None
1312 # Quelch pyflakes warnings - start will be set when step is set
1313 start = '(Never used)'
1314 for i, prev in zip(idxs[1:], idxs[:-1]):
1315 if step is not None:
1316 if i - prev == step:
1317 continue
1318 yield _genslice(start, prev, step)
1319 step = None
1320 continue
1321 if i - prev in [-1, 1]:
1322 step = i - prev
1323 start = prev
1324 continue
1325 else:
1326 yield 's[%d]' % prev
1327 if step is None:
1328 yield 's[%d]' % i
1329 else:
1330 yield _genslice(start, i, step)
1331
1332 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1333 cache_res = func(test_string)
1334 cache_spec = [ord(c) for c in cache_res]
1335 expr_code = ' + '.join(gen_sig_code(cache_spec))
1336 signature_id_tuple = '(%s)' % (
1337 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1338 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1339 ' return %s\n') % (signature_id_tuple, expr_code)
1340 self.to_screen('Extracted signature function:\n' + code)
1341
1342 def _parse_sig_js(self, jscode):
1343 funcname = self._search_regex(
1344 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1345 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1346 r'\b(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1347 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1348 # Obsolete patterns
1349 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1350 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1351 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1352 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1353 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1354 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1355 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1356 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1357 jscode, 'Initial JS player signature function name', group='sig')
1358
1359 jsi = JSInterpreter(jscode)
1360 initial_function = jsi.extract_function(funcname)
1361 return lambda s: initial_function([s])
1362
1363 def _parse_sig_swf(self, file_contents):
1364 swfi = SWFInterpreter(file_contents)
1365 TARGET_CLASSNAME = 'SignatureDecipher'
1366 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1367 initial_function = swfi.extract_function(searched_class, 'decipher')
1368 return lambda s: initial_function([s])
1369
1370 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1371 """Turn the encrypted s field into a working signature"""
1372
1373 if player_url is None:
1374 raise ExtractorError('Cannot decrypt signature without player_url')
1375
1376 if player_url.startswith('//'):
1377 player_url = 'https:' + player_url
1378 elif not re.match(r'https?://', player_url):
1379 player_url = compat_urlparse.urljoin(
1380 'https://www.youtube.com', player_url)
1381 try:
1382 player_id = (player_url, self._signature_cache_id(s))
1383 if player_id not in self._player_cache:
1384 func = self._extract_signature_function(
1385 video_id, player_url, s
1386 )
1387 self._player_cache[player_id] = func
1388 func = self._player_cache[player_id]
1389 if self._downloader.params.get('youtube_print_sig_code'):
1390 self._print_sig_code(func, s)
1391 return func(s)
1392 except Exception as e:
1393 tb = traceback.format_exc()
1394 raise ExtractorError(
1395 'Signature extraction failed: ' + tb, cause=e)
1396
1397 def _get_subtitles(self, video_id, webpage):
1398 try:
1399 subs_doc = self._download_xml(
1400 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1401 video_id, note=False)
1402 except ExtractorError as err:
1403 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1404 return {}
1405
1406 sub_lang_list = {}
1407 for track in subs_doc.findall('track'):
1408 lang = track.attrib['lang_code']
1409 if lang in sub_lang_list:
1410 continue
1411 sub_formats = []
1412 for ext in self._SUBTITLE_FORMATS:
1413 params = compat_urllib_parse_urlencode({
1414 'lang': lang,
1415 'v': video_id,
1416 'fmt': ext,
1417 'name': track.attrib['name'].encode('utf-8'),
1418 })
1419 sub_formats.append({
1420 'url': 'https://www.youtube.com/api/timedtext?' + params,
1421 'ext': ext,
1422 })
1423 sub_lang_list[lang] = sub_formats
1424 if not sub_lang_list:
1425 self._downloader.report_warning('video doesn\'t have subtitles')
1426 return {}
1427 return sub_lang_list
1428
1429 def _get_ytplayer_config(self, video_id, webpage):
1430 patterns = (
1431 # User data may contain arbitrary character sequences that may affect
1432 # JSON extraction with regex, e.g. when '};' is contained the second
1433 # regex won't capture the whole JSON. Yet working around by trying more
1434 # concrete regex first keeping in mind proper quoted string handling
1435 # to be implemented in future that will replace this workaround (see
1436 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1437 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1438 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1439 r';ytplayer\.config\s*=\s*({.+?});',
1440 )
1441 config = self._search_regex(
1442 patterns, webpage, 'ytplayer.config', default=None)
1443 if config:
1444 return self._parse_json(
1445 uppercase_escape(config), video_id, fatal=False)
1446
1447 def _get_automatic_captions(self, video_id, webpage):
1448 """We need the webpage for getting the captions url, pass it as an
1449 argument to speed up the process."""
1450 self.to_screen('%s: Looking for automatic captions' % video_id)
1451 player_config = self._get_ytplayer_config(video_id, webpage)
1452 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1453 if not player_config:
1454 self._downloader.report_warning(err_msg)
1455 return {}
1456 try:
1457 args = player_config['args']
1458 caption_url = args.get('ttsurl')
1459 if caption_url:
1460 timestamp = args['timestamp']
1461 # We get the available subtitles
1462 list_params = compat_urllib_parse_urlencode({
1463 'type': 'list',
1464 'tlangs': 1,
1465 'asrs': 1,
1466 })
1467 list_url = caption_url + '&' + list_params
1468 caption_list = self._download_xml(list_url, video_id)
1469 original_lang_node = caption_list.find('track')
1470 if original_lang_node is None:
1471 self._downloader.report_warning('Video doesn\'t have automatic captions')
1472 return {}
1473 original_lang = original_lang_node.attrib['lang_code']
1474 caption_kind = original_lang_node.attrib.get('kind', '')
1475
1476 sub_lang_list = {}
1477 for lang_node in caption_list.findall('target'):
1478 sub_lang = lang_node.attrib['lang_code']
1479 sub_formats = []
1480 for ext in self._SUBTITLE_FORMATS:
1481 params = compat_urllib_parse_urlencode({
1482 'lang': original_lang,
1483 'tlang': sub_lang,
1484 'fmt': ext,
1485 'ts': timestamp,
1486 'kind': caption_kind,
1487 })
1488 sub_formats.append({
1489 'url': caption_url + '&' + params,
1490 'ext': ext,
1491 })
1492 sub_lang_list[sub_lang] = sub_formats
1493 return sub_lang_list
1494
1495 def make_captions(sub_url, sub_langs):
1496 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1497 caption_qs = compat_parse_qs(parsed_sub_url.query)
1498 captions = {}
1499 for sub_lang in sub_langs:
1500 sub_formats = []
1501 for ext in self._SUBTITLE_FORMATS:
1502 caption_qs.update({
1503 'tlang': [sub_lang],
1504 'fmt': [ext],
1505 })
1506 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1507 query=compat_urllib_parse_urlencode(caption_qs, True)))
1508 sub_formats.append({
1509 'url': sub_url,
1510 'ext': ext,
1511 })
1512 captions[sub_lang] = sub_formats
1513 return captions
1514
1515 # New captions format as of 22.06.2017
1516 player_response = args.get('player_response')
1517 if player_response and isinstance(player_response, compat_str):
1518 player_response = self._parse_json(
1519 player_response, video_id, fatal=False)
1520 if player_response:
1521 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1522 base_url = renderer['captionTracks'][0]['baseUrl']
1523 sub_lang_list = []
1524 for lang in renderer['translationLanguages']:
1525 lang_code = lang.get('languageCode')
1526 if lang_code:
1527 sub_lang_list.append(lang_code)
1528 return make_captions(base_url, sub_lang_list)
1529
1530 # Some videos don't provide ttsurl but rather caption_tracks and
1531 # caption_translation_languages (e.g. 20LmZk1hakA)
1532 # Does not used anymore as of 22.06.2017
1533 caption_tracks = args['caption_tracks']
1534 caption_translation_languages = args['caption_translation_languages']
1535 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1536 sub_lang_list = []
1537 for lang in caption_translation_languages.split(','):
1538 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1539 sub_lang = lang_qs.get('lc', [None])[0]
1540 if sub_lang:
1541 sub_lang_list.append(sub_lang)
1542 return make_captions(caption_url, sub_lang_list)
1543 # An extractor error can be raise by the download process if there are
1544 # no automatic captions but there are subtitles
1545 except (KeyError, IndexError, ExtractorError):
1546 self._downloader.report_warning(err_msg)
1547 return {}
1548
1549 def _mark_watched(self, video_id, video_info, player_response):
1550 playback_url = url_or_none(try_get(
1551 player_response,
1552 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1553 video_info, lambda x: x['videostats_playback_base_url'][0]))
1554 if not playback_url:
1555 return
1556 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1557 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1558
1559 # cpn generation algorithm is reverse engineered from base.js.
1560 # In fact it works even with dummy cpn.
1561 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1562 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1563
1564 qs.update({
1565 'ver': ['2'],
1566 'cpn': [cpn],
1567 })
1568 playback_url = compat_urlparse.urlunparse(
1569 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1570
1571 self._download_webpage(
1572 playback_url, video_id, 'Marking watched',
1573 'Unable to mark watched', fatal=False)
1574
1575 @staticmethod
1576 def _extract_urls(webpage):
1577 # Embedded YouTube player
1578 entries = [
1579 unescapeHTML(mobj.group('url'))
1580 for mobj in re.finditer(r'''(?x)
1581 (?:
1582 <iframe[^>]+?src=|
1583 data-video-url=|
1584 <embed[^>]+?src=|
1585 embedSWF\(?:\s*|
1586 <object[^>]+data=|
1587 new\s+SWFObject\(
1588 )
1589 (["\'])
1590 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1591 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1592 \1''', webpage)]
1593
1594 # lazyYT YouTube embed
1595 entries.extend(list(map(
1596 unescapeHTML,
1597 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1598
1599 # Wordpress "YouTube Video Importer" plugin
1600 matches = re.findall(r'''(?x)<div[^>]+
1601 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1602 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1603 entries.extend(m[-1] for m in matches)
1604
1605 return entries
1606
1607 @staticmethod
1608 def _extract_url(webpage):
1609 urls = YoutubeIE._extract_urls(webpage)
1610 return urls[0] if urls else None
1611
1612 @classmethod
1613 def extract_id(cls, url):
1614 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1615 if mobj is None:
1616 raise ExtractorError('Invalid URL: %s' % url)
1617 video_id = mobj.group(2)
1618 return video_id
1619
1620 @staticmethod
1621 def _extract_chapters(description, duration):
1622 if not description:
1623 return None
1624 chapter_lines = re.findall(
1625 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1626 description)
1627 if not chapter_lines:
1628 return None
1629 chapters = []
1630 for next_num, (chapter_line, time_point) in enumerate(
1631 chapter_lines, start=1):
1632 start_time = parse_duration(time_point)
1633 if start_time is None:
1634 continue
1635 if start_time > duration:
1636 break
1637 end_time = (duration if next_num == len(chapter_lines)
1638 else parse_duration(chapter_lines[next_num][1]))
1639 if end_time is None:
1640 continue
1641 if end_time > duration:
1642 end_time = duration
1643 if start_time > end_time:
1644 break
1645 chapter_title = re.sub(
1646 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1647 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1648 chapters.append({
1649 'start_time': start_time,
1650 'end_time': end_time,
1651 'title': chapter_title,
1652 })
1653 return chapters
1654
1655 def _real_extract(self, url):
1656 url, smuggled_data = unsmuggle_url(url, {})
1657
1658 proto = (
1659 'http' if self._downloader.params.get('prefer_insecure', False)
1660 else 'https')
1661
1662 start_time = None
1663 end_time = None
1664 parsed_url = compat_urllib_parse_urlparse(url)
1665 for component in [parsed_url.fragment, parsed_url.query]:
1666 query = compat_parse_qs(component)
1667 if start_time is None and 't' in query:
1668 start_time = parse_duration(query['t'][0])
1669 if start_time is None and 'start' in query:
1670 start_time = parse_duration(query['start'][0])
1671 if end_time is None and 'end' in query:
1672 end_time = parse_duration(query['end'][0])
1673
1674 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1675 mobj = re.search(self._NEXT_URL_RE, url)
1676 if mobj:
1677 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1678 video_id = self.extract_id(url)
1679
1680 # Get video webpage
1681 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1682 video_webpage = self._download_webpage(url, video_id)
1683
1684 # Attempt to extract SWF player URL
1685 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1686 if mobj is not None:
1687 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1688 else:
1689 player_url = None
1690
1691 dash_mpds = []
1692
1693 def add_dash_mpd(video_info):
1694 dash_mpd = video_info.get('dashmpd')
1695 if dash_mpd and dash_mpd[0] not in dash_mpds:
1696 dash_mpds.append(dash_mpd[0])
1697
1698 def add_dash_mpd_pr(pl_response):
1699 dash_mpd = url_or_none(try_get(
1700 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1701 compat_str))
1702 if dash_mpd and dash_mpd not in dash_mpds:
1703 dash_mpds.append(dash_mpd)
1704
1705 is_live = None
1706 view_count = None
1707
1708 def extract_view_count(v_info):
1709 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1710
1711 def extract_token(v_info):
1712 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1713
1714 def extract_player_response(player_response, video_id):
1715 pl_response = str_or_none(player_response)
1716 if not pl_response:
1717 return
1718 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1719 if isinstance(pl_response, dict):
1720 add_dash_mpd_pr(pl_response)
1721 return pl_response
1722
1723 player_response = {}
1724
1725 # Get video info
1726 embed_webpage = None
1727 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1728 age_gate = True
1729 # We simulate the access to the video from www.youtube.com/v/{video_id}
1730 # this can be viewed without login into Youtube
1731 url = proto + '://www.youtube.com/embed/%s' % video_id
1732 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1733 data = compat_urllib_parse_urlencode({
1734 'video_id': video_id,
1735 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1736 'sts': self._search_regex(
1737 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1738 })
1739 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1740 video_info_webpage = self._download_webpage(
1741 video_info_url, video_id,
1742 note='Refetching age-gated info webpage',
1743 errnote='unable to download video info webpage')
1744 video_info = compat_parse_qs(video_info_webpage)
1745 pl_response = video_info.get('player_response', [None])[0]
1746 player_response = extract_player_response(pl_response, video_id)
1747 add_dash_mpd(video_info)
1748 view_count = extract_view_count(video_info)
1749 else:
1750 age_gate = False
1751 video_info = None
1752 sts = None
1753 # Try looking directly into the video webpage
1754 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1755 if ytplayer_config:
1756 args = ytplayer_config['args']
1757 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1758 # Convert to the same format returned by compat_parse_qs
1759 video_info = dict((k, [v]) for k, v in args.items())
1760 add_dash_mpd(video_info)
1761 # Rental video is not rented but preview is available (e.g.
1762 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1763 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1764 if not video_info and args.get('ypc_vid'):
1765 return self.url_result(
1766 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1767 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1768 is_live = True
1769 sts = ytplayer_config.get('sts')
1770 if not player_response:
1771 player_response = extract_player_response(args.get('player_response'), video_id)
1772 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1773 add_dash_mpd_pr(player_response)
1774 # We also try looking in get_video_info since it may contain different dashmpd
1775 # URL that points to a DASH manifest with possibly different itag set (some itags
1776 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1777 # manifest pointed by get_video_info's dashmpd).
1778 # The general idea is to take a union of itags of both DASH manifests (for example
1779 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1780 self.report_video_info_webpage_download(video_id)
1781 for el in ('embedded', 'detailpage', 'vevo', ''):
1782 query = {
1783 'video_id': video_id,
1784 'ps': 'default',
1785 'eurl': '',
1786 'gl': 'US',
1787 'hl': 'en',
1788 }
1789 if el:
1790 query['el'] = el
1791 if sts:
1792 query['sts'] = sts
1793 video_info_webpage = self._download_webpage(
1794 '%s://www.youtube.com/get_video_info' % proto,
1795 video_id, note=False,
1796 errnote='unable to download video info webpage',
1797 fatal=False, query=query)
1798 if not video_info_webpage:
1799 continue
1800 get_video_info = compat_parse_qs(video_info_webpage)
1801 if not player_response:
1802 pl_response = get_video_info.get('player_response', [None])[0]
1803 player_response = extract_player_response(pl_response, video_id)
1804 add_dash_mpd(get_video_info)
1805 if view_count is None:
1806 view_count = extract_view_count(get_video_info)
1807 if not video_info:
1808 video_info = get_video_info
1809 get_token = extract_token(get_video_info)
1810 if get_token:
1811 # Different get_video_info requests may report different results, e.g.
1812 # some may report video unavailability, but some may serve it without
1813 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1814 # the original webpage as well as el=info and el=embedded get_video_info
1815 # requests report video unavailability due to geo restriction while
1816 # el=detailpage succeeds and returns valid data). This is probably
1817 # due to YouTube measures against IP ranges of hosting providers.
1818 # Working around by preferring the first succeeded video_info containing
1819 # the token if no such video_info yet was found.
1820 token = extract_token(video_info)
1821 if not token:
1822 video_info = get_video_info
1823 break
1824
1825 def extract_unavailable_message():
1826 messages = []
1827 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1828 msg = self._html_search_regex(
1829 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1830 video_webpage, 'unavailable %s' % kind, default=None)
1831 if msg:
1832 messages.append(msg)
1833 if messages:
1834 return '\n'.join(messages)
1835
1836 if not video_info:
1837 unavailable_message = extract_unavailable_message()
1838 if not unavailable_message:
1839 unavailable_message = 'Unable to extract video data'
1840 raise ExtractorError(
1841 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1842
1843 video_details = try_get(
1844 player_response, lambda x: x['videoDetails'], dict) or {}
1845
1846 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1847 if not video_title:
1848 self._downloader.report_warning('Unable to extract video title')
1849 video_title = '_'
1850
1851 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1852 if video_description:
1853
1854 def replace_url(m):
1855 redir_url = compat_urlparse.urljoin(url, m.group(1))
1856 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1857 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1858 qs = compat_parse_qs(parsed_redir_url.query)
1859 q = qs.get('q')
1860 if q and q[0]:
1861 return q[0]
1862 return redir_url
1863
1864 description_original = video_description = re.sub(r'''(?x)
1865 <a\s+
1866 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1867 (?:title|href)="([^"]+)"\s+
1868 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1869 class="[^"]*"[^>]*>
1870 [^<]+\.{3}\s*
1871 </a>
1872 ''', replace_url, video_description)
1873 video_description = clean_html(video_description)
1874 else:
1875 video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
1876
1877 if not smuggled_data.get('force_singlefeed', False):
1878 if not self._downloader.params.get('noplaylist'):
1879 multifeed_metadata_list = try_get(
1880 player_response,
1881 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1882 compat_str) or try_get(
1883 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1884 if multifeed_metadata_list:
1885 entries = []
1886 feed_ids = []
1887 for feed in multifeed_metadata_list.split(','):
1888 # Unquote should take place before split on comma (,) since textual
1889 # fields may contain comma as well (see
1890 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1891 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1892 entries.append({
1893 '_type': 'url_transparent',
1894 'ie_key': 'Youtube',
1895 'url': smuggle_url(
1896 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1897 {'force_singlefeed': True}),
1898 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1899 })
1900 feed_ids.append(feed_data['id'][0])
1901 self.to_screen(
1902 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1903 % (', '.join(feed_ids), video_id))
1904 return self.playlist_result(entries, video_id, video_title, video_description)
1905 else:
1906 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1907
1908 if view_count is None:
1909 view_count = extract_view_count(video_info)
1910 if view_count is None and video_details:
1911 view_count = int_or_none(video_details.get('viewCount'))
1912
1913 if is_live is None:
1914 is_live = bool_or_none(video_details.get('isLive'))
1915
1916 # Check for "rental" videos
1917 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1918 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1919
1920 def _extract_filesize(media_url):
1921 return int_or_none(self._search_regex(
1922 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1923
1924 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
1925 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
1926
1927 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1928 self.report_rtmp_download()
1929 formats = [{
1930 'format_id': '_rtmp',
1931 'protocol': 'rtmp',
1932 'url': video_info['conn'][0],
1933 'player_url': player_url,
1934 }]
1935 elif not is_live and (streaming_formats or len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1936 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1937 if 'rtmpe%3Dyes' in encoded_url_map:
1938 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1939 formats = []
1940 formats_spec = {}
1941 fmt_list = video_info.get('fmt_list', [''])[0]
1942 if fmt_list:
1943 for fmt in fmt_list.split(','):
1944 spec = fmt.split('/')
1945 if len(spec) > 1:
1946 width_height = spec[1].split('x')
1947 if len(width_height) == 2:
1948 formats_spec[spec[0]] = {
1949 'resolution': spec[1],
1950 'width': int_or_none(width_height[0]),
1951 'height': int_or_none(width_height[1]),
1952 }
1953 for fmt in streaming_formats:
1954 itag = str_or_none(fmt.get('itag'))
1955 if not itag:
1956 continue
1957 quality = fmt.get('quality')
1958 quality_label = fmt.get('qualityLabel') or quality
1959 formats_spec[itag] = {
1960 'asr': int_or_none(fmt.get('audioSampleRate')),
1961 'filesize': int_or_none(fmt.get('contentLength')),
1962 'format_note': quality_label,
1963 'fps': int_or_none(fmt.get('fps')),
1964 'height': int_or_none(fmt.get('height')),
1965 # bitrate for itag 43 is always 2147483647
1966 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1967 'width': int_or_none(fmt.get('width')),
1968 }
1969
1970 for fmt in streaming_formats:
1971 if fmt.get('drm_families'):
1972 continue
1973 url = url_or_none(fmt.get('url'))
1974
1975 if not url:
1976 cipher = fmt.get('cipher')
1977 if not cipher:
1978 continue
1979 url_data = compat_parse_qs(cipher)
1980 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
1981 if not url:
1982 continue
1983 else:
1984 cipher = None
1985 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
1986
1987 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1988 # Unsupported FORMAT_STREAM_TYPE_OTF
1989 if stream_type == 3:
1990 continue
1991
1992 format_id = fmt.get('itag') or url_data['itag'][0]
1993 if not format_id:
1994 continue
1995 format_id = compat_str(format_id)
1996
1997 if cipher:
1998 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1999 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
2000 jsplayer_url_json = self._search_regex(
2001 ASSETS_RE,
2002 embed_webpage if age_gate else video_webpage,
2003 'JS player URL (1)', default=None)
2004 if not jsplayer_url_json and not age_gate:
2005 # We need the embed website after all
2006 if embed_webpage is None:
2007 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
2008 embed_webpage = self._download_webpage(
2009 embed_url, video_id, 'Downloading embed webpage')
2010 jsplayer_url_json = self._search_regex(
2011 ASSETS_RE, embed_webpage, 'JS player URL')
2012
2013 player_url = json.loads(jsplayer_url_json)
2014 if player_url is None:
2015 player_url_json = self._search_regex(
2016 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2017 video_webpage, 'age gate player URL')
2018 player_url = json.loads(player_url_json)
2019
2020 if 'sig' in url_data:
2021 url += '&signature=' + url_data['sig'][0]
2022 elif 's' in url_data:
2023 encrypted_sig = url_data['s'][0]
2024
2025 if self._downloader.params.get('verbose'):
2026 if player_url is None:
2027 player_version = 'unknown'
2028 player_desc = 'unknown'
2029 else:
2030 if player_url.endswith('swf'):
2031 player_version = self._search_regex(
2032 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
2033 'flash player', fatal=False)
2034 player_desc = 'flash player %s' % player_version
2035 else:
2036 player_version = self._search_regex(
2037 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
2038 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
2039 player_url,
2040 'html5 player', fatal=False)
2041 player_desc = 'html5 player %s' % player_version
2042
2043 parts_sizes = self._signature_cache_id(encrypted_sig)
2044 self.to_screen('{%s} signature length %s, %s' %
2045 (format_id, parts_sizes, player_desc))
2046
2047 signature = self._decrypt_signature(
2048 encrypted_sig, video_id, player_url, age_gate)
2049 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2050 url += '&%s=%s' % (sp, signature)
2051 if 'ratebypass' not in url:
2052 url += '&ratebypass=yes'
2053
2054 dct = {
2055 'format_id': format_id,
2056 'url': url,
2057 'player_url': player_url,
2058 }
2059 if format_id in self._formats:
2060 dct.update(self._formats[format_id])
2061 if format_id in formats_spec:
2062 dct.update(formats_spec[format_id])
2063
2064 # Some itags are not included in DASH manifest thus corresponding formats will
2065 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2066 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2067 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2068 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2069
2070 if width is None:
2071 width = int_or_none(fmt.get('width'))
2072 if height is None:
2073 height = int_or_none(fmt.get('height'))
2074
2075 filesize = int_or_none(url_data.get(
2076 'clen', [None])[0]) or _extract_filesize(url)
2077
2078 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2079 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2080
2081 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2082 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2083 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2084
2085 more_fields = {
2086 'filesize': filesize,
2087 'tbr': tbr,
2088 'width': width,
2089 'height': height,
2090 'fps': fps,
2091 'format_note': quality_label or quality,
2092 }
2093 for key, value in more_fields.items():
2094 if value:
2095 dct[key] = value
2096 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2097 if type_:
2098 type_split = type_.split(';')
2099 kind_ext = type_split[0].split('/')
2100 if len(kind_ext) == 2:
2101 kind, _ = kind_ext
2102 dct['ext'] = mimetype2ext(type_split[0])
2103 if kind in ('audio', 'video'):
2104 codecs = None
2105 for mobj in re.finditer(
2106 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2107 if mobj.group('key') == 'codecs':
2108 codecs = mobj.group('val')
2109 break
2110 if codecs:
2111 dct.update(parse_codecs(codecs))
2112 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2113 dct['downloader_options'] = {
2114 # Youtube throttles chunks >~10M
2115 'http_chunk_size': 10485760,
2116 }
2117 formats.append(dct)
2118 else:
2119 manifest_url = (
2120 url_or_none(try_get(
2121 player_response,
2122 lambda x: x['streamingData']['hlsManifestUrl'],
2123 compat_str))
2124 or url_or_none(try_get(
2125 video_info, lambda x: x['hlsvp'][0], compat_str)))
2126 if manifest_url:
2127 formats = []
2128 m3u8_formats = self._extract_m3u8_formats(
2129 manifest_url, video_id, 'mp4', fatal=False)
2130 for a_format in m3u8_formats:
2131 itag = self._search_regex(
2132 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2133 if itag:
2134 a_format['format_id'] = itag
2135 if itag in self._formats:
2136 dct = self._formats[itag].copy()
2137 dct.update(a_format)
2138 a_format = dct
2139 a_format['player_url'] = player_url
2140 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2141 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2142 formats.append(a_format)
2143 else:
2144 error_message = extract_unavailable_message()
2145 if not error_message:
2146 error_message = clean_html(try_get(
2147 player_response, lambda x: x['playabilityStatus']['reason'],
2148 compat_str))
2149 if not error_message:
2150 error_message = clean_html(
2151 try_get(video_info, lambda x: x['reason'][0], compat_str))
2152 if error_message:
2153 raise ExtractorError(error_message, expected=True)
2154 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2155
2156 # uploader
2157 video_uploader = try_get(
2158 video_info, lambda x: x['author'][0],
2159 compat_str) or str_or_none(video_details.get('author'))
2160 if video_uploader:
2161 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2162 else:
2163 self._downloader.report_warning('unable to extract uploader name')
2164
2165 # uploader_id
2166 video_uploader_id = None
2167 video_uploader_url = None
2168 mobj = re.search(
2169 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2170 video_webpage)
2171 if mobj is not None:
2172 video_uploader_id = mobj.group('uploader_id')
2173 video_uploader_url = mobj.group('uploader_url')
2174 else:
2175 self._downloader.report_warning('unable to extract uploader nickname')
2176
2177 channel_id = (
2178 str_or_none(video_details.get('channelId'))
2179 or self._html_search_meta(
2180 'channelId', video_webpage, 'channel id', default=None)
2181 or self._search_regex(
2182 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2183 video_webpage, 'channel id', default=None, group='id'))
2184 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2185
2186 # thumbnail image
2187 # We try first to get a high quality image:
2188 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2189 video_webpage, re.DOTALL)
2190 if m_thumb is not None:
2191 video_thumbnail = m_thumb.group(1)
2192 elif 'thumbnail_url' not in video_info:
2193 self._downloader.report_warning('unable to extract video thumbnail')
2194 video_thumbnail = None
2195 else: # don't panic if we can't find it
2196 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2197
2198 # upload date
2199 upload_date = self._html_search_meta(
2200 'datePublished', video_webpage, 'upload date', default=None)
2201 if not upload_date:
2202 upload_date = self._search_regex(
2203 [r'(?s)id="eow-date.*?>(.*?)</span>',
2204 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2205 video_webpage, 'upload date', default=None)
2206 upload_date = unified_strdate(upload_date)
2207
2208 video_license = self._html_search_regex(
2209 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2210 video_webpage, 'license', default=None)
2211
2212 m_music = re.search(
2213 r'''(?x)
2214 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2215 <ul[^>]*>\s*
2216 <li>(?P<title>.+?)
2217 by (?P<creator>.+?)
2218 (?:
2219 \(.+?\)|
2220 <a[^>]*
2221 (?:
2222 \bhref=["\']/red[^>]*>| # drop possible
2223 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2224 )
2225 .*?
2226 )?</li
2227 ''',
2228 video_webpage)
2229 if m_music:
2230 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2231 video_creator = clean_html(m_music.group('creator'))
2232 else:
2233 video_alt_title = video_creator = None
2234
2235 def extract_meta(field):
2236 return self._html_search_regex(
2237 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2238 video_webpage, field, default=None)
2239
2240 track = extract_meta('Song')
2241 artist = extract_meta('Artist')
2242 album = extract_meta('Album')
2243
2244 # Youtube Music Auto-generated description
2245 release_date = release_year = None
2246 if video_description:
2247 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)
2248 if mobj:
2249 if not track:
2250 track = mobj.group('track').strip()
2251 if not artist:
2252 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('Ā·'))
2253 if not album:
2254 album = mobj.group('album'.strip())
2255 release_year = mobj.group('release_year')
2256 release_date = mobj.group('release_date')
2257 if release_date:
2258 release_date = release_date.replace('-', '')
2259 if not release_year:
2260 release_year = int(release_date[:4])
2261 if release_year:
2262 release_year = int(release_year)
2263
2264 m_episode = re.search(
2265 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2266 video_webpage)
2267 if m_episode:
2268 series = unescapeHTML(m_episode.group('series'))
2269 season_number = int(m_episode.group('season'))
2270 episode_number = int(m_episode.group('episode'))
2271 else:
2272 series = season_number = episode_number = None
2273
2274 m_cat_container = self._search_regex(
2275 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2276 video_webpage, 'categories', default=None)
2277 if m_cat_container:
2278 category = self._html_search_regex(
2279 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2280 default=None)
2281 video_categories = None if category is None else [category]
2282 else:
2283 video_categories = None
2284
2285 video_tags = [
2286 unescapeHTML(m.group('content'))
2287 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2288
2289 def _extract_count(count_name):
2290 return str_to_int(self._search_regex(
2291 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2292 % re.escape(count_name),
2293 video_webpage, count_name, default=None))
2294
2295 like_count = _extract_count('like')
2296 dislike_count = _extract_count('dislike')
2297
2298 if view_count is None:
2299 view_count = str_to_int(self._search_regex(
2300 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2301 'view count', default=None))
2302
2303 average_rating = (
2304 float_or_none(video_details.get('averageRating'))
2305 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2306
2307 # subtitles
2308 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2309 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2310
2311 video_duration = try_get(
2312 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2313 if not video_duration:
2314 video_duration = int_or_none(video_details.get('lengthSeconds'))
2315 if not video_duration:
2316 video_duration = parse_duration(self._html_search_meta(
2317 'duration', video_webpage, 'video duration'))
2318
2319 # annotations
2320 video_annotations = None
2321 if self._downloader.params.get('writeannotations', False):
2322 xsrf_token = self._search_regex(
2323 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2324 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2325 invideo_url = try_get(
2326 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2327 if xsrf_token and invideo_url:
2328 xsrf_field_name = self._search_regex(
2329 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2330 video_webpage, 'xsrf field name',
2331 group='xsrf_field_name', default='session_token')
2332 video_annotations = self._download_webpage(
2333 self._proto_relative_url(invideo_url),
2334 video_id, note='Downloading annotations',
2335 errnote='Unable to download video annotations', fatal=False,
2336 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2337
2338 chapters = self._extract_chapters(description_original, video_duration)
2339
2340 # Look for the DASH manifest
2341 if self._downloader.params.get('youtube_include_dash_manifest', True):
2342 dash_mpd_fatal = True
2343 for mpd_url in dash_mpds:
2344 dash_formats = {}
2345 try:
2346 def decrypt_sig(mobj):
2347 s = mobj.group(1)
2348 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2349 return '/signature/%s' % dec_s
2350
2351 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2352
2353 for df in self._extract_mpd_formats(
2354 mpd_url, video_id, fatal=dash_mpd_fatal,
2355 formats_dict=self._formats):
2356 if not df.get('filesize'):
2357 df['filesize'] = _extract_filesize(df['url'])
2358 # Do not overwrite DASH format found in some previous DASH manifest
2359 if df['format_id'] not in dash_formats:
2360 dash_formats[df['format_id']] = df
2361 # Additional DASH manifests may end up in HTTP Error 403 therefore
2362 # allow them to fail without bug report message if we already have
2363 # some DASH manifest succeeded. This is temporary workaround to reduce
2364 # burst of bug reports until we figure out the reason and whether it
2365 # can be fixed at all.
2366 dash_mpd_fatal = False
2367 except (ExtractorError, KeyError) as e:
2368 self.report_warning(
2369 'Skipping DASH manifest: %r' % e, video_id)
2370 if dash_formats:
2371 # Remove the formats we found through non-DASH, they
2372 # contain less info and it can be wrong, because we use
2373 # fixed values (for example the resolution). See
2374 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2375 # example.
2376 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2377 formats.extend(dash_formats.values())
2378
2379 # Check for malformed aspect ratio
2380 stretched_m = re.search(
2381 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2382 video_webpage)
2383 if stretched_m:
2384 w = float(stretched_m.group('w'))
2385 h = float(stretched_m.group('h'))
2386 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2387 # We will only process correct ratios.
2388 if w > 0 and h > 0:
2389 ratio = w / h
2390 for f in formats:
2391 if f.get('vcodec') != 'none':
2392 f['stretched_ratio'] = ratio
2393
2394 if not formats:
2395 token = extract_token(video_info)
2396 if not token:
2397 if 'reason' in video_info:
2398 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2399 regions_allowed = self._html_search_meta(
2400 'regionsAllowed', video_webpage, default=None)
2401 countries = regions_allowed.split(',') if regions_allowed else None
2402 self.raise_geo_restricted(
2403 msg=video_info['reason'][0], countries=countries)
2404 reason = video_info['reason'][0]
2405 if 'Invalid parameters' in reason:
2406 unavailable_message = extract_unavailable_message()
2407 if unavailable_message:
2408 reason = unavailable_message
2409 raise ExtractorError(
2410 'YouTube said: %s' % reason,
2411 expected=True, video_id=video_id)
2412 else:
2413 raise ExtractorError(
2414 '"token" parameter not in video info for unknown reason',
2415 video_id=video_id)
2416
2417 if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2418 raise ExtractorError('This video is DRM protected.', expected=True)
2419
2420 self._sort_formats(formats)
2421
2422 self.mark_watched(video_id, video_info, player_response)
2423
2424 return {
2425 'id': video_id,
2426 'uploader': video_uploader,
2427 'uploader_id': video_uploader_id,
2428 'uploader_url': video_uploader_url,
2429 'channel_id': channel_id,
2430 'channel_url': channel_url,
2431 'upload_date': upload_date,
2432 'license': video_license,
2433 'creator': video_creator or artist,
2434 'title': video_title,
2435 'alt_title': video_alt_title or track,
2436 'thumbnail': video_thumbnail,
2437 'description': video_description,
2438 'categories': video_categories,
2439 'tags': video_tags,
2440 'subtitles': video_subtitles,
2441 'automatic_captions': automatic_captions,
2442 'duration': video_duration,
2443 'age_limit': 18 if age_gate else 0,
2444 'annotations': video_annotations,
2445 'chapters': chapters,
2446 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2447 'view_count': view_count,
2448 'like_count': like_count,
2449 'dislike_count': dislike_count,
2450 'average_rating': average_rating,
2451 'formats': formats,
2452 'is_live': is_live,
2453 'start_time': start_time,
2454 'end_time': end_time,
2455 'series': series,
2456 'season_number': season_number,
2457 'episode_number': episode_number,
2458 'track': track,
2459 'artist': artist,
2460 'album': album,
2461 'release_date': release_date,
2462 'release_year': release_year,
2463 }
2464
2465
2466 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2467 IE_DESC = 'YouTube.com playlists'
2468 _VALID_URL = r"""(?x)(?:
2469 (?:https?://)?
2470 (?:\w+\.)?
2471 (?:
2472 (?:
2473 youtube(?:kids)?\.com|
2474 invidio\.us
2475 )
2476 /
2477 (?:
2478 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2479 \? (?:.*?[&;])*? (?:p|a|list)=
2480 | p/
2481 )|
2482 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2483 )
2484 (
2485 (?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2486 # Top tracks, they can also include dots
2487 |(?:MC)[\w\.]*
2488 )
2489 .*
2490 |
2491 (%(playlist_id)s)
2492 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2493 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2494 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2495 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2496 IE_NAME = 'youtube:playlist'
2497 _TESTS = [{
2498 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2499 'info_dict': {
2500 'title': 'ytdl test PL',
2501 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2502 },
2503 'playlist_count': 3,
2504 }, {
2505 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2506 'info_dict': {
2507 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2508 'title': 'YDL_Empty_List',
2509 },
2510 'playlist_count': 0,
2511 'skip': 'This playlist is private',
2512 }, {
2513 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2514 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2515 'info_dict': {
2516 'title': '29C3: Not my department',
2517 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2518 'uploader': 'Christiaan008',
2519 'uploader_id': 'ChRiStIaAn008',
2520 },
2521 'playlist_count': 95,
2522 }, {
2523 'note': 'issue #673',
2524 'url': 'PLBB231211A4F62143',
2525 'info_dict': {
2526 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2527 'id': 'PLBB231211A4F62143',
2528 'uploader': 'Wickydoo',
2529 'uploader_id': 'Wickydoo',
2530 },
2531 'playlist_mincount': 26,
2532 }, {
2533 'note': 'Large playlist',
2534 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2535 'info_dict': {
2536 'title': 'Uploads from Cauchemar',
2537 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2538 'uploader': 'Cauchemar',
2539 'uploader_id': 'Cauchemar89',
2540 },
2541 'playlist_mincount': 799,
2542 }, {
2543 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2544 'info_dict': {
2545 'title': 'YDL_safe_search',
2546 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2547 },
2548 'playlist_count': 2,
2549 'skip': 'This playlist is private',
2550 }, {
2551 'note': 'embedded',
2552 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2553 'playlist_count': 4,
2554 'info_dict': {
2555 'title': 'JODA15',
2556 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2557 'uploader': 'milan',
2558 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2559 }
2560 }, {
2561 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2562 'playlist_mincount': 485,
2563 'info_dict': {
2564 'title': '2018 Chinese New Singles (11/6 updated)',
2565 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2566 'uploader': 'LBK',
2567 'uploader_id': 'sdragonfang',
2568 }
2569 }, {
2570 'note': 'Embedded SWF player',
2571 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2572 'playlist_count': 4,
2573 'info_dict': {
2574 'title': 'JODA7',
2575 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2576 },
2577 'skip': 'This playlist does not exist',
2578 }, {
2579 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2580 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2581 'info_dict': {
2582 'title': 'Uploads from Interstellar Movie',
2583 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2584 'uploader': 'Interstellar Movie',
2585 'uploader_id': 'InterstellarMovie1',
2586 },
2587 'playlist_mincount': 21,
2588 }, {
2589 # Playlist URL that does not actually serve a playlist
2590 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2591 'info_dict': {
2592 'id': 'FqZTN594JQw',
2593 'ext': 'webm',
2594 'title': "Smiley's People 01 detective, Adventure Series, Action",
2595 'uploader': 'STREEM',
2596 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2597 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2598 'upload_date': '20150526',
2599 'license': 'Standard YouTube License',
2600 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2601 'categories': ['People & Blogs'],
2602 'tags': list,
2603 'view_count': int,
2604 'like_count': int,
2605 'dislike_count': int,
2606 },
2607 'params': {
2608 'skip_download': True,
2609 },
2610 'skip': 'This video is not available.',
2611 'add_ie': [YoutubeIE.ie_key()],
2612 }, {
2613 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2614 'info_dict': {
2615 'id': 'yeWKywCrFtk',
2616 'ext': 'mp4',
2617 'title': 'Small Scale Baler and Braiding Rugs',
2618 'uploader': 'Backus-Page House Museum',
2619 'uploader_id': 'backuspagemuseum',
2620 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2621 'upload_date': '20161008',
2622 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2623 'categories': ['Nonprofits & Activism'],
2624 'tags': list,
2625 'like_count': int,
2626 'dislike_count': int,
2627 },
2628 'params': {
2629 'noplaylist': True,
2630 'skip_download': True,
2631 },
2632 }, {
2633 # https://github.com/ytdl-org/youtube-dl/issues/21844
2634 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2635 'info_dict': {
2636 'title': 'Data Analysis with Dr Mike Pound',
2637 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2638 'uploader_id': 'Computerphile',
2639 'uploader': 'Computerphile',
2640 },
2641 'playlist_mincount': 11,
2642 }, {
2643 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2644 'only_matching': True,
2645 }, {
2646 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2647 'only_matching': True,
2648 }, {
2649 # music album playlist
2650 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2651 'only_matching': True,
2652 }, {
2653 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2654 'only_matching': True,
2655 }, {
2656 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g',
2657 'only_matching': True,
2658 }]
2659
2660 def _real_initialize(self):
2661 self._login()
2662
2663 def extract_videos_from_page(self, page):
2664 ids_in_page = []
2665 titles_in_page = []
2666
2667 for item in re.findall(
2668 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2669 attrs = extract_attributes(item)
2670 video_id = attrs['data-video-id']
2671 video_title = unescapeHTML(attrs.get('data-title'))
2672 if video_title:
2673 video_title = video_title.strip()
2674 ids_in_page.append(video_id)
2675 titles_in_page.append(video_title)
2676
2677 # Fallback with old _VIDEO_RE
2678 self.extract_videos_from_page_impl(
2679 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2680
2681 # Relaxed fallbacks
2682 self.extract_videos_from_page_impl(
2683 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2684 ids_in_page, titles_in_page)
2685 self.extract_videos_from_page_impl(
2686 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2687 ids_in_page, titles_in_page)
2688
2689 return zip(ids_in_page, titles_in_page)
2690
2691 def _extract_mix(self, playlist_id):
2692 # The mixes are generated from a single video
2693 # the id of the playlist is just 'RD' + video_id
2694 ids = []
2695 last_id = playlist_id[-11:]
2696 for n in itertools.count(1):
2697 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2698 webpage = self._download_webpage(
2699 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2700 new_ids = orderedSet(re.findall(
2701 r'''(?xs)data-video-username=".*?".*?
2702 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2703 webpage))
2704 # Fetch new pages until all the videos are repeated, it seems that
2705 # there are always 51 unique videos.
2706 new_ids = [_id for _id in new_ids if _id not in ids]
2707 if not new_ids:
2708 break
2709 ids.extend(new_ids)
2710 last_id = ids[-1]
2711
2712 url_results = self._ids_to_results(ids)
2713
2714 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2715 title_span = (
2716 search_title('playlist-title')
2717 or search_title('title long-title')
2718 or search_title('title'))
2719 title = clean_html(title_span)
2720
2721 return self.playlist_result(url_results, playlist_id, title)
2722
2723 def _extract_playlist(self, playlist_id):
2724 url = self._TEMPLATE_URL % playlist_id
2725 page = self._download_webpage(url, playlist_id)
2726
2727 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2728 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2729 match = match.strip()
2730 # Check if the playlist exists or is private
2731 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2732 if mobj:
2733 reason = mobj.group('reason')
2734 message = 'This playlist %s' % reason
2735 if 'private' in reason:
2736 message += ', use --username or --netrc to access it'
2737 message += '.'
2738 raise ExtractorError(message, expected=True)
2739 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2740 raise ExtractorError(
2741 'Invalid parameters. Maybe URL is incorrect.',
2742 expected=True)
2743 elif re.match(r'[^<]*Choose your language[^<]*', match):
2744 continue
2745 else:
2746 self.report_warning('Youtube gives an alert message: ' + match)
2747
2748 playlist_title = self._html_search_regex(
2749 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2750 page, 'title', default=None)
2751
2752 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2753 uploader = self._html_search_regex(
2754 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2755 page, 'uploader', default=None)
2756 mobj = re.search(
2757 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2758 page)
2759 if mobj:
2760 uploader_id = mobj.group('uploader_id')
2761 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2762 else:
2763 uploader_id = uploader_url = None
2764
2765 has_videos = True
2766
2767 if not playlist_title:
2768 try:
2769 # Some playlist URLs don't actually serve a playlist (e.g.
2770 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2771 next(self._entries(page, playlist_id))
2772 except StopIteration:
2773 has_videos = False
2774
2775 playlist = self.playlist_result(
2776 self._entries(page, playlist_id), playlist_id, playlist_title)
2777 playlist.update({
2778 'uploader': uploader,
2779 'uploader_id': uploader_id,
2780 'uploader_url': uploader_url,
2781 })
2782
2783 return has_videos, playlist
2784
2785 def _check_download_just_video(self, url, playlist_id):
2786 # Check if it's a video-specific URL
2787 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2788 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2789 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2790 'video id', default=None)
2791 if video_id:
2792 if self._downloader.params.get('noplaylist'):
2793 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2794 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2795 else:
2796 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2797 return video_id, None
2798 return None, None
2799
2800 def _real_extract(self, url):
2801 # Extract playlist id
2802 mobj = re.match(self._VALID_URL, url)
2803 if mobj is None:
2804 raise ExtractorError('Invalid URL: %s' % url)
2805 playlist_id = mobj.group(1) or mobj.group(2)
2806
2807 video_id, video = self._check_download_just_video(url, playlist_id)
2808 if video:
2809 return video
2810
2811 if playlist_id.startswith(('RD', 'UL', 'PU')):
2812 # Mixes require a custom extraction process
2813 return self._extract_mix(playlist_id)
2814
2815 has_videos, playlist = self._extract_playlist(playlist_id)
2816 if has_videos or not video_id:
2817 return playlist
2818
2819 # Some playlist URLs don't actually serve a playlist (see
2820 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2821 # Fallback to plain video extraction if there is a video id
2822 # along with playlist id.
2823 return self.url_result(video_id, 'Youtube', video_id=video_id)
2824
2825
2826 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2827 IE_DESC = 'YouTube.com channels'
2828 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie|kids)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2829 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2830 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2831 IE_NAME = 'youtube:channel'
2832 _TESTS = [{
2833 'note': 'paginated channel',
2834 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2835 'playlist_mincount': 91,
2836 'info_dict': {
2837 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2838 'title': 'Uploads from lex will',
2839 'uploader': 'lex will',
2840 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2841 }
2842 }, {
2843 'note': 'Age restricted channel',
2844 # from https://www.youtube.com/user/DeusExOfficial
2845 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2846 'playlist_mincount': 64,
2847 'info_dict': {
2848 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2849 'title': 'Uploads from Deus Ex',
2850 'uploader': 'Deus Ex',
2851 'uploader_id': 'DeusExOfficial',
2852 },
2853 }, {
2854 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2855 'only_matching': True,
2856 }, {
2857 'url': 'https://www.youtubekids.com/channel/UCyu8StPfZWapR6rfW_JgqcA',
2858 'only_matching': True,
2859 }]
2860
2861 @classmethod
2862 def suitable(cls, url):
2863 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2864 else super(YoutubeChannelIE, cls).suitable(url))
2865
2866 def _build_template_url(self, url, channel_id):
2867 return self._TEMPLATE_URL % channel_id
2868
2869 def _real_extract(self, url):
2870 channel_id = self._match_id(url)
2871
2872 url = self._build_template_url(url, channel_id)
2873
2874 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2875 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2876 # otherwise fallback on channel by page extraction
2877 channel_page = self._download_webpage(
2878 url + '?view=57', channel_id,
2879 'Downloading channel page', fatal=False)
2880 if channel_page is False:
2881 channel_playlist_id = False
2882 else:
2883 channel_playlist_id = self._html_search_meta(
2884 'channelId', channel_page, 'channel id', default=None)
2885 if not channel_playlist_id:
2886 channel_url = self._html_search_meta(
2887 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2888 channel_page, 'channel url', default=None)
2889 if channel_url:
2890 channel_playlist_id = self._search_regex(
2891 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2892 channel_url, 'channel id', default=None)
2893 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2894 playlist_id = 'UU' + channel_playlist_id[2:]
2895 return self.url_result(
2896 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2897
2898 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2899 autogenerated = re.search(r'''(?x)
2900 class="[^"]*?(?:
2901 channel-header-autogenerated-label|
2902 yt-channel-title-autogenerated
2903 )[^"]*"''', channel_page) is not None
2904
2905 if autogenerated:
2906 # The videos are contained in a single page
2907 # the ajax pages can't be used, they are empty
2908 entries = [
2909 self.url_result(
2910 video_id, 'Youtube', video_id=video_id,
2911 video_title=video_title)
2912 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2913 return self.playlist_result(entries, channel_id)
2914
2915 try:
2916 next(self._entries(channel_page, channel_id))
2917 except StopIteration:
2918 alert_message = self._html_search_regex(
2919 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2920 channel_page, 'alert', default=None, group='alert')
2921 if alert_message:
2922 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2923
2924 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2925
2926
2927 class YoutubeUserIE(YoutubeChannelIE):
2928 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2929 _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_-]+)'
2930 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2931 IE_NAME = 'youtube:user'
2932
2933 _TESTS = [{
2934 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2935 'playlist_mincount': 320,
2936 'info_dict': {
2937 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2938 'title': 'Uploads from The Linux Foundation',
2939 'uploader': 'The Linux Foundation',
2940 'uploader_id': 'TheLinuxFoundation',
2941 }
2942 }, {
2943 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2944 # but not https://www.youtube.com/user/12minuteathlete/videos
2945 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2946 'playlist_mincount': 249,
2947 'info_dict': {
2948 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2949 'title': 'Uploads from 12 Minute Athlete',
2950 'uploader': '12 Minute Athlete',
2951 'uploader_id': 'the12minuteathlete',
2952 }
2953 }, {
2954 'url': 'ytuser:phihag',
2955 'only_matching': True,
2956 }, {
2957 'url': 'https://www.youtube.com/c/gametrailers',
2958 'only_matching': True,
2959 }, {
2960 'url': 'https://www.youtube.com/gametrailers',
2961 'only_matching': True,
2962 }, {
2963 # This channel is not available, geo restricted to JP
2964 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2965 'only_matching': True,
2966 }]
2967
2968 @classmethod
2969 def suitable(cls, url):
2970 # Don't return True if the url can be extracted with other youtube
2971 # extractor, the regex would is too permissive and it would match.
2972 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2973 if any(ie.suitable(url) for ie in other_yt_ies):
2974 return False
2975 else:
2976 return super(YoutubeUserIE, cls).suitable(url)
2977
2978 def _build_template_url(self, url, channel_id):
2979 mobj = re.match(self._VALID_URL, url)
2980 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2981
2982
2983 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2984 IE_DESC = 'YouTube.com live streams'
2985 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2986 IE_NAME = 'youtube:live'
2987
2988 _TESTS = [{
2989 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2990 'info_dict': {
2991 'id': 'a48o2S1cPoo',
2992 'ext': 'mp4',
2993 'title': 'The Young Turks - Live Main Show',
2994 'uploader': 'The Young Turks',
2995 'uploader_id': 'TheYoungTurks',
2996 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2997 'upload_date': '20150715',
2998 'license': 'Standard YouTube License',
2999 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
3000 'categories': ['News & Politics'],
3001 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
3002 'like_count': int,
3003 'dislike_count': int,
3004 },
3005 'params': {
3006 'skip_download': True,
3007 },
3008 }, {
3009 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
3010 'only_matching': True,
3011 }, {
3012 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3013 'only_matching': True,
3014 }, {
3015 'url': 'https://www.youtube.com/TheYoungTurks/live',
3016 'only_matching': True,
3017 }]
3018
3019 def _real_extract(self, url):
3020 mobj = re.match(self._VALID_URL, url)
3021 channel_id = mobj.group('id')
3022 base_url = mobj.group('base_url')
3023 webpage = self._download_webpage(url, channel_id, fatal=False)
3024 if webpage:
3025 page_type = self._og_search_property(
3026 'type', webpage, 'page type', default='')
3027 video_id = self._html_search_meta(
3028 'videoId', webpage, 'video id', default=None)
3029 if page_type.startswith('video') and video_id and re.match(
3030 r'^[0-9A-Za-z_-]{11}$', video_id):
3031 return self.url_result(video_id, YoutubeIE.ie_key())
3032 return self.url_result(base_url)
3033
3034
3035 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
3036 IE_DESC = 'YouTube.com user/channel playlists'
3037 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
3038 IE_NAME = 'youtube:playlists'
3039
3040 _TESTS = [{
3041 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3042 'playlist_mincount': 4,
3043 'info_dict': {
3044 'id': 'ThirstForScience',
3045 'title': 'ThirstForScience',
3046 },
3047 }, {
3048 # with "Load more" button
3049 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3050 'playlist_mincount': 70,
3051 'info_dict': {
3052 'id': 'igorkle1',
3053 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
3054 },
3055 }, {
3056 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3057 'playlist_mincount': 17,
3058 'info_dict': {
3059 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3060 'title': 'Chem Player',
3061 },
3062 'skip': 'Blocked',
3063 }]
3064
3065
3066 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3067 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3068
3069
3070 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3071 IE_DESC = 'YouTube.com searches'
3072 # there doesn't appear to be a real limit, for example if you search for
3073 # 'python' you get more than 8.000.000 results
3074 _MAX_RESULTS = float('inf')
3075 IE_NAME = 'youtube:search'
3076 _SEARCH_KEY = 'ytsearch'
3077 _EXTRA_QUERY_ARGS = {}
3078 _TESTS = []
3079
3080 def _get_n_results(self, query, n):
3081 """Get a specified number of results for a query"""
3082
3083 videos = []
3084 limit = n
3085
3086 url_query = {
3087 'search_query': query.encode('utf-8'),
3088 }
3089 url_query.update(self._EXTRA_QUERY_ARGS)
3090 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3091
3092 for pagenum in itertools.count(1):
3093 data = self._download_json(
3094 result_url, video_id='query "%s"' % query,
3095 note='Downloading page %s' % pagenum,
3096 errnote='Unable to download API page',
3097 query={'spf': 'navigate'})
3098 html_content = data[1]['body']['content']
3099
3100 if 'class="search-message' in html_content:
3101 raise ExtractorError(
3102 '[youtube] No video results', expected=True)
3103
3104 new_videos = list(self._process_page(html_content))
3105 videos += new_videos
3106 if not new_videos or len(videos) > limit:
3107 break
3108 next_link = self._html_search_regex(
3109 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3110 html_content, 'next link', default=None)
3111 if next_link is None:
3112 break
3113 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3114
3115 if len(videos) > n:
3116 videos = videos[:n]
3117 return self.playlist_result(videos, query)
3118
3119
3120 class YoutubeSearchDateIE(YoutubeSearchIE):
3121 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3122 _SEARCH_KEY = 'ytsearchdate'
3123 IE_DESC = 'YouTube.com searches, newest videos first'
3124 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3125
3126
3127 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3128 IE_DESC = 'YouTube.com search URLs'
3129 IE_NAME = 'youtube:search_url'
3130 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3131 _TESTS = [{
3132 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3133 'playlist_mincount': 5,
3134 'info_dict': {
3135 'title': 'youtube-dl test video',
3136 }
3137 }, {
3138 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3139 'only_matching': True,
3140 }]
3141
3142 def _real_extract(self, url):
3143 mobj = re.match(self._VALID_URL, url)
3144 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3145 webpage = self._download_webpage(url, query)
3146 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3147
3148
3149 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3150 IE_DESC = 'YouTube.com (multi-season) shows'
3151 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3152 IE_NAME = 'youtube:show'
3153 _TESTS = [{
3154 'url': 'https://www.youtube.com/show/airdisasters',
3155 'playlist_mincount': 5,
3156 'info_dict': {
3157 'id': 'airdisasters',
3158 'title': 'Air Disasters',
3159 }
3160 }]
3161
3162 def _real_extract(self, url):
3163 playlist_id = self._match_id(url)
3164 return super(YoutubeShowIE, self)._real_extract(
3165 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3166
3167
3168 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3169 """
3170 Base class for feed extractors
3171 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3172 """
3173 _LOGIN_REQUIRED = True
3174
3175 @property
3176 def IE_NAME(self):
3177 return 'youtube:%s' % self._FEED_NAME
3178
3179 def _real_initialize(self):
3180 self._login()
3181
3182 def _entries(self, page):
3183 # The extraction process is the same as for playlists, but the regex
3184 # for the video ids doesn't contain an index
3185 ids = []
3186 more_widget_html = content_html = page
3187 for page_num in itertools.count(1):
3188 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3189
3190 # 'recommended' feed has infinite 'load more' and each new portion spins
3191 # the same videos in (sometimes) slightly different order, so we'll check
3192 # for unicity and break when portion has no new videos
3193 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3194 if not new_ids:
3195 break
3196
3197 ids.extend(new_ids)
3198
3199 for entry in self._ids_to_results(new_ids):
3200 yield entry
3201
3202 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3203 if not mobj:
3204 break
3205
3206 more = self._download_json(
3207 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3208 'Downloading page #%s' % page_num,
3209 transform_source=uppercase_escape)
3210 content_html = more['content_html']
3211 more_widget_html = more['load_more_widget_html']
3212
3213 def _real_extract(self, url):
3214 page = self._download_webpage(
3215 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3216 self._PLAYLIST_TITLE)
3217 return self.playlist_result(
3218 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3219
3220
3221 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3222 IE_NAME = 'youtube:watchlater'
3223 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3224 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3225
3226 _TESTS = [{
3227 'url': 'https://www.youtube.com/playlist?list=WL',
3228 'only_matching': True,
3229 }, {
3230 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3231 'only_matching': True,
3232 }]
3233
3234 def _real_extract(self, url):
3235 _, video = self._check_download_just_video(url, 'WL')
3236 if video:
3237 return video
3238 _, playlist = self._extract_playlist('WL')
3239 return playlist
3240
3241
3242 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3243 IE_NAME = 'youtube:favorites'
3244 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3245 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3246 _LOGIN_REQUIRED = True
3247
3248 def _real_extract(self, url):
3249 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3250 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3251 return self.url_result(playlist_id, 'YoutubePlaylist')
3252
3253
3254 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3255 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3256 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3257 _FEED_NAME = 'recommended'
3258 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3259
3260
3261 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3262 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3263 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3264 _FEED_NAME = 'subscriptions'
3265 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3266
3267
3268 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3269 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3270 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3271 _FEED_NAME = 'history'
3272 _PLAYLIST_TITLE = 'Youtube History'
3273
3274
3275 class YoutubeTruncatedURLIE(InfoExtractor):
3276 IE_NAME = 'youtube:truncated_url'
3277 IE_DESC = False # Do not list
3278 _VALID_URL = r'''(?x)
3279 (?:https?://)?
3280 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3281 (?:watch\?(?:
3282 feature=[a-z_]+|
3283 annotation_id=annotation_[^&]+|
3284 x-yt-cl=[0-9]+|
3285 hl=[^&]*|
3286 t=[0-9]+
3287 )?
3288 |
3289 attribution_link\?a=[^&]+
3290 )
3291 $
3292 '''
3293
3294 _TESTS = [{
3295 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3296 'only_matching': True,
3297 }, {
3298 'url': 'https://www.youtube.com/watch?',
3299 'only_matching': True,
3300 }, {
3301 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3302 'only_matching': True,
3303 }, {
3304 'url': 'https://www.youtube.com/watch?feature=foo',
3305 'only_matching': True,
3306 }, {
3307 'url': 'https://www.youtube.com/watch?hl=en-GB',
3308 'only_matching': True,
3309 }, {
3310 'url': 'https://www.youtube.com/watch?t=2372',
3311 'only_matching': True,
3312 }]
3313
3314 def _real_extract(self, url):
3315 raise ExtractorError(
3316 'Did you forget to quote the URL? Remember that & is a meta '
3317 'character in most shells, so you want to put the URL in quotes, '
3318 'like youtube-dl '
3319 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3320 ' or simply youtube-dl BaW_jenozKc .',
3321 expected=True)
3322
3323
3324 class YoutubeTruncatedIDIE(InfoExtractor):
3325 IE_NAME = 'youtube:truncated_id'
3326 IE_DESC = False # Do not list
3327 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3328
3329 _TESTS = [{
3330 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3331 'only_matching': True,
3332 }]
3333
3334 def _real_extract(self, url):
3335 video_id = self._match_id(url)
3336 raise ExtractorError(
3337 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3338 expected=True)