]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
New upstream version 2019.09.28
[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|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)?\.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\.enkirton\.net/|
387 (?:www\.)?invidious\.13ad\.de/|
388 (?:www\.)?invidious\.mastodon\.host/|
389 (?:www\.)?invidious\.nixnet\.xyz/|
390 (?:www\.)?invidious\.drycat\.fr/|
391 (?:www\.)?tube\.poal\.co/|
392 (?:www\.)?vid\.wxzm\.sx/|
393 (?:www\.)?yt\.elukerio\.org/|
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
1229 def __init__(self, *args, **kwargs):
1230 super(YoutubeIE, self).__init__(*args, **kwargs)
1231 self._player_cache = {}
1232
1233 def report_video_info_webpage_download(self, video_id):
1234 """Report attempt to download video info webpage."""
1235 self.to_screen('%s: Downloading video info webpage' % video_id)
1236
1237 def report_information_extraction(self, video_id):
1238 """Report attempt to extract video information."""
1239 self.to_screen('%s: Extracting video information' % video_id)
1240
1241 def report_unavailable_format(self, video_id, format):
1242 """Report extracted video URL."""
1243 self.to_screen('%s: Format %s not available' % (video_id, format))
1244
1245 def report_rtmp_download(self):
1246 """Indicate the download will use the RTMP protocol."""
1247 self.to_screen('RTMP download detected')
1248
1249 def _signature_cache_id(self, example_sig):
1250 """ Return a string representation of a signature """
1251 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1252
1253 def _extract_signature_function(self, video_id, player_url, example_sig):
1254 id_m = re.match(
1255 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1256 player_url)
1257 if not id_m:
1258 raise ExtractorError('Cannot identify player %r' % player_url)
1259 player_type = id_m.group('ext')
1260 player_id = id_m.group('id')
1261
1262 # Read from filesystem cache
1263 func_id = '%s_%s_%s' % (
1264 player_type, player_id, self._signature_cache_id(example_sig))
1265 assert os.path.basename(func_id) == func_id
1266
1267 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1268 if cache_spec is not None:
1269 return lambda s: ''.join(s[i] for i in cache_spec)
1270
1271 download_note = (
1272 'Downloading player %s' % player_url
1273 if self._downloader.params.get('verbose') else
1274 'Downloading %s player %s' % (player_type, player_id)
1275 )
1276 if player_type == 'js':
1277 code = self._download_webpage(
1278 player_url, video_id,
1279 note=download_note,
1280 errnote='Download of %s failed' % player_url)
1281 res = self._parse_sig_js(code)
1282 elif player_type == 'swf':
1283 urlh = self._request_webpage(
1284 player_url, video_id,
1285 note=download_note,
1286 errnote='Download of %s failed' % player_url)
1287 code = urlh.read()
1288 res = self._parse_sig_swf(code)
1289 else:
1290 assert False, 'Invalid player type %r' % player_type
1291
1292 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1293 cache_res = res(test_string)
1294 cache_spec = [ord(c) for c in cache_res]
1295
1296 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1297 return res
1298
1299 def _print_sig_code(self, func, example_sig):
1300 def gen_sig_code(idxs):
1301 def _genslice(start, end, step):
1302 starts = '' if start == 0 else str(start)
1303 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1304 steps = '' if step == 1 else (':%d' % step)
1305 return 's[%s%s%s]' % (starts, ends, steps)
1306
1307 step = None
1308 # Quelch pyflakes warnings - start will be set when step is set
1309 start = '(Never used)'
1310 for i, prev in zip(idxs[1:], idxs[:-1]):
1311 if step is not None:
1312 if i - prev == step:
1313 continue
1314 yield _genslice(start, prev, step)
1315 step = None
1316 continue
1317 if i - prev in [-1, 1]:
1318 step = i - prev
1319 start = prev
1320 continue
1321 else:
1322 yield 's[%d]' % prev
1323 if step is None:
1324 yield 's[%d]' % i
1325 else:
1326 yield _genslice(start, i, step)
1327
1328 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1329 cache_res = func(test_string)
1330 cache_spec = [ord(c) for c in cache_res]
1331 expr_code = ' + '.join(gen_sig_code(cache_spec))
1332 signature_id_tuple = '(%s)' % (
1333 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1334 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1335 ' return %s\n') % (signature_id_tuple, expr_code)
1336 self.to_screen('Extracted signature function:\n' + code)
1337
1338 def _parse_sig_js(self, jscode):
1339 funcname = self._search_regex(
1340 (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1341 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1342 r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
1343 # Obsolete patterns
1344 r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1345 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1346 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1347 r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1348 r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1349 r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1350 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1351 r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1352 jscode, 'Initial JS player signature function name', group='sig')
1353
1354 jsi = JSInterpreter(jscode)
1355 initial_function = jsi.extract_function(funcname)
1356 return lambda s: initial_function([s])
1357
1358 def _parse_sig_swf(self, file_contents):
1359 swfi = SWFInterpreter(file_contents)
1360 TARGET_CLASSNAME = 'SignatureDecipher'
1361 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1362 initial_function = swfi.extract_function(searched_class, 'decipher')
1363 return lambda s: initial_function([s])
1364
1365 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1366 """Turn the encrypted s field into a working signature"""
1367
1368 if player_url is None:
1369 raise ExtractorError('Cannot decrypt signature without player_url')
1370
1371 if player_url.startswith('//'):
1372 player_url = 'https:' + player_url
1373 elif not re.match(r'https?://', player_url):
1374 player_url = compat_urlparse.urljoin(
1375 'https://www.youtube.com', player_url)
1376 try:
1377 player_id = (player_url, self._signature_cache_id(s))
1378 if player_id not in self._player_cache:
1379 func = self._extract_signature_function(
1380 video_id, player_url, s
1381 )
1382 self._player_cache[player_id] = func
1383 func = self._player_cache[player_id]
1384 if self._downloader.params.get('youtube_print_sig_code'):
1385 self._print_sig_code(func, s)
1386 return func(s)
1387 except Exception as e:
1388 tb = traceback.format_exc()
1389 raise ExtractorError(
1390 'Signature extraction failed: ' + tb, cause=e)
1391
1392 def _get_subtitles(self, video_id, webpage):
1393 try:
1394 subs_doc = self._download_xml(
1395 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1396 video_id, note=False)
1397 except ExtractorError as err:
1398 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1399 return {}
1400
1401 sub_lang_list = {}
1402 for track in subs_doc.findall('track'):
1403 lang = track.attrib['lang_code']
1404 if lang in sub_lang_list:
1405 continue
1406 sub_formats = []
1407 for ext in self._SUBTITLE_FORMATS:
1408 params = compat_urllib_parse_urlencode({
1409 'lang': lang,
1410 'v': video_id,
1411 'fmt': ext,
1412 'name': track.attrib['name'].encode('utf-8'),
1413 })
1414 sub_formats.append({
1415 'url': 'https://www.youtube.com/api/timedtext?' + params,
1416 'ext': ext,
1417 })
1418 sub_lang_list[lang] = sub_formats
1419 if not sub_lang_list:
1420 self._downloader.report_warning('video doesn\'t have subtitles')
1421 return {}
1422 return sub_lang_list
1423
1424 def _get_ytplayer_config(self, video_id, webpage):
1425 patterns = (
1426 # User data may contain arbitrary character sequences that may affect
1427 # JSON extraction with regex, e.g. when '};' is contained the second
1428 # regex won't capture the whole JSON. Yet working around by trying more
1429 # concrete regex first keeping in mind proper quoted string handling
1430 # to be implemented in future that will replace this workaround (see
1431 # https://github.com/ytdl-org/youtube-dl/issues/7468,
1432 # https://github.com/ytdl-org/youtube-dl/pull/7599)
1433 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1434 r';ytplayer\.config\s*=\s*({.+?});',
1435 )
1436 config = self._search_regex(
1437 patterns, webpage, 'ytplayer.config', default=None)
1438 if config:
1439 return self._parse_json(
1440 uppercase_escape(config), video_id, fatal=False)
1441
1442 def _get_automatic_captions(self, video_id, webpage):
1443 """We need the webpage for getting the captions url, pass it as an
1444 argument to speed up the process."""
1445 self.to_screen('%s: Looking for automatic captions' % video_id)
1446 player_config = self._get_ytplayer_config(video_id, webpage)
1447 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1448 if not player_config:
1449 self._downloader.report_warning(err_msg)
1450 return {}
1451 try:
1452 args = player_config['args']
1453 caption_url = args.get('ttsurl')
1454 if caption_url:
1455 timestamp = args['timestamp']
1456 # We get the available subtitles
1457 list_params = compat_urllib_parse_urlencode({
1458 'type': 'list',
1459 'tlangs': 1,
1460 'asrs': 1,
1461 })
1462 list_url = caption_url + '&' + list_params
1463 caption_list = self._download_xml(list_url, video_id)
1464 original_lang_node = caption_list.find('track')
1465 if original_lang_node is None:
1466 self._downloader.report_warning('Video doesn\'t have automatic captions')
1467 return {}
1468 original_lang = original_lang_node.attrib['lang_code']
1469 caption_kind = original_lang_node.attrib.get('kind', '')
1470
1471 sub_lang_list = {}
1472 for lang_node in caption_list.findall('target'):
1473 sub_lang = lang_node.attrib['lang_code']
1474 sub_formats = []
1475 for ext in self._SUBTITLE_FORMATS:
1476 params = compat_urllib_parse_urlencode({
1477 'lang': original_lang,
1478 'tlang': sub_lang,
1479 'fmt': ext,
1480 'ts': timestamp,
1481 'kind': caption_kind,
1482 })
1483 sub_formats.append({
1484 'url': caption_url + '&' + params,
1485 'ext': ext,
1486 })
1487 sub_lang_list[sub_lang] = sub_formats
1488 return sub_lang_list
1489
1490 def make_captions(sub_url, sub_langs):
1491 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1492 caption_qs = compat_parse_qs(parsed_sub_url.query)
1493 captions = {}
1494 for sub_lang in sub_langs:
1495 sub_formats = []
1496 for ext in self._SUBTITLE_FORMATS:
1497 caption_qs.update({
1498 'tlang': [sub_lang],
1499 'fmt': [ext],
1500 })
1501 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1502 query=compat_urllib_parse_urlencode(caption_qs, True)))
1503 sub_formats.append({
1504 'url': sub_url,
1505 'ext': ext,
1506 })
1507 captions[sub_lang] = sub_formats
1508 return captions
1509
1510 # New captions format as of 22.06.2017
1511 player_response = args.get('player_response')
1512 if player_response and isinstance(player_response, compat_str):
1513 player_response = self._parse_json(
1514 player_response, video_id, fatal=False)
1515 if player_response:
1516 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1517 base_url = renderer['captionTracks'][0]['baseUrl']
1518 sub_lang_list = []
1519 for lang in renderer['translationLanguages']:
1520 lang_code = lang.get('languageCode')
1521 if lang_code:
1522 sub_lang_list.append(lang_code)
1523 return make_captions(base_url, sub_lang_list)
1524
1525 # Some videos don't provide ttsurl but rather caption_tracks and
1526 # caption_translation_languages (e.g. 20LmZk1hakA)
1527 # Does not used anymore as of 22.06.2017
1528 caption_tracks = args['caption_tracks']
1529 caption_translation_languages = args['caption_translation_languages']
1530 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1531 sub_lang_list = []
1532 for lang in caption_translation_languages.split(','):
1533 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1534 sub_lang = lang_qs.get('lc', [None])[0]
1535 if sub_lang:
1536 sub_lang_list.append(sub_lang)
1537 return make_captions(caption_url, sub_lang_list)
1538 # An extractor error can be raise by the download process if there are
1539 # no automatic captions but there are subtitles
1540 except (KeyError, IndexError, ExtractorError):
1541 self._downloader.report_warning(err_msg)
1542 return {}
1543
1544 def _mark_watched(self, video_id, video_info, player_response):
1545 playback_url = url_or_none(try_get(
1546 player_response,
1547 lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1548 video_info, lambda x: x['videostats_playback_base_url'][0]))
1549 if not playback_url:
1550 return
1551 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1552 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1553
1554 # cpn generation algorithm is reverse engineered from base.js.
1555 # In fact it works even with dummy cpn.
1556 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1557 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1558
1559 qs.update({
1560 'ver': ['2'],
1561 'cpn': [cpn],
1562 })
1563 playback_url = compat_urlparse.urlunparse(
1564 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1565
1566 self._download_webpage(
1567 playback_url, video_id, 'Marking watched',
1568 'Unable to mark watched', fatal=False)
1569
1570 @staticmethod
1571 def _extract_urls(webpage):
1572 # Embedded YouTube player
1573 entries = [
1574 unescapeHTML(mobj.group('url'))
1575 for mobj in re.finditer(r'''(?x)
1576 (?:
1577 <iframe[^>]+?src=|
1578 data-video-url=|
1579 <embed[^>]+?src=|
1580 embedSWF\(?:\s*|
1581 <object[^>]+data=|
1582 new\s+SWFObject\(
1583 )
1584 (["\'])
1585 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1586 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1587 \1''', webpage)]
1588
1589 # lazyYT YouTube embed
1590 entries.extend(list(map(
1591 unescapeHTML,
1592 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1593
1594 # Wordpress "YouTube Video Importer" plugin
1595 matches = re.findall(r'''(?x)<div[^>]+
1596 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1597 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1598 entries.extend(m[-1] for m in matches)
1599
1600 return entries
1601
1602 @staticmethod
1603 def _extract_url(webpage):
1604 urls = YoutubeIE._extract_urls(webpage)
1605 return urls[0] if urls else None
1606
1607 @classmethod
1608 def extract_id(cls, url):
1609 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1610 if mobj is None:
1611 raise ExtractorError('Invalid URL: %s' % url)
1612 video_id = mobj.group(2)
1613 return video_id
1614
1615 @staticmethod
1616 def _extract_chapters(description, duration):
1617 if not description:
1618 return None
1619 chapter_lines = re.findall(
1620 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1621 description)
1622 if not chapter_lines:
1623 return None
1624 chapters = []
1625 for next_num, (chapter_line, time_point) in enumerate(
1626 chapter_lines, start=1):
1627 start_time = parse_duration(time_point)
1628 if start_time is None:
1629 continue
1630 if start_time > duration:
1631 break
1632 end_time = (duration if next_num == len(chapter_lines)
1633 else parse_duration(chapter_lines[next_num][1]))
1634 if end_time is None:
1635 continue
1636 if end_time > duration:
1637 end_time = duration
1638 if start_time > end_time:
1639 break
1640 chapter_title = re.sub(
1641 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1642 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1643 chapters.append({
1644 'start_time': start_time,
1645 'end_time': end_time,
1646 'title': chapter_title,
1647 })
1648 return chapters
1649
1650 def _real_extract(self, url):
1651 url, smuggled_data = unsmuggle_url(url, {})
1652
1653 proto = (
1654 'http' if self._downloader.params.get('prefer_insecure', False)
1655 else 'https')
1656
1657 start_time = None
1658 end_time = None
1659 parsed_url = compat_urllib_parse_urlparse(url)
1660 for component in [parsed_url.fragment, parsed_url.query]:
1661 query = compat_parse_qs(component)
1662 if start_time is None and 't' in query:
1663 start_time = parse_duration(query['t'][0])
1664 if start_time is None and 'start' in query:
1665 start_time = parse_duration(query['start'][0])
1666 if end_time is None and 'end' in query:
1667 end_time = parse_duration(query['end'][0])
1668
1669 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1670 mobj = re.search(self._NEXT_URL_RE, url)
1671 if mobj:
1672 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1673 video_id = self.extract_id(url)
1674
1675 # Get video webpage
1676 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1677 video_webpage = self._download_webpage(url, video_id)
1678
1679 # Attempt to extract SWF player URL
1680 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1681 if mobj is not None:
1682 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1683 else:
1684 player_url = None
1685
1686 dash_mpds = []
1687
1688 def add_dash_mpd(video_info):
1689 dash_mpd = video_info.get('dashmpd')
1690 if dash_mpd and dash_mpd[0] not in dash_mpds:
1691 dash_mpds.append(dash_mpd[0])
1692
1693 def add_dash_mpd_pr(pl_response):
1694 dash_mpd = url_or_none(try_get(
1695 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1696 compat_str))
1697 if dash_mpd and dash_mpd not in dash_mpds:
1698 dash_mpds.append(dash_mpd)
1699
1700 is_live = None
1701 view_count = None
1702
1703 def extract_view_count(v_info):
1704 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1705
1706 def extract_token(v_info):
1707 return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1708
1709 def extract_player_response(player_response, video_id):
1710 pl_response = str_or_none(player_response)
1711 if not pl_response:
1712 return
1713 pl_response = self._parse_json(pl_response, video_id, fatal=False)
1714 if isinstance(pl_response, dict):
1715 add_dash_mpd_pr(pl_response)
1716 return pl_response
1717
1718 player_response = {}
1719
1720 # Get video info
1721 embed_webpage = None
1722 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1723 age_gate = True
1724 # We simulate the access to the video from www.youtube.com/v/{video_id}
1725 # this can be viewed without login into Youtube
1726 url = proto + '://www.youtube.com/embed/%s' % video_id
1727 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1728 data = compat_urllib_parse_urlencode({
1729 'video_id': video_id,
1730 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1731 'sts': self._search_regex(
1732 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1733 })
1734 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1735 video_info_webpage = self._download_webpage(
1736 video_info_url, video_id,
1737 note='Refetching age-gated info webpage',
1738 errnote='unable to download video info webpage')
1739 video_info = compat_parse_qs(video_info_webpage)
1740 pl_response = video_info.get('player_response', [None])[0]
1741 player_response = extract_player_response(pl_response, video_id)
1742 add_dash_mpd(video_info)
1743 view_count = extract_view_count(video_info)
1744 else:
1745 age_gate = False
1746 video_info = None
1747 sts = None
1748 # Try looking directly into the video webpage
1749 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1750 if ytplayer_config:
1751 args = ytplayer_config['args']
1752 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1753 # Convert to the same format returned by compat_parse_qs
1754 video_info = dict((k, [v]) for k, v in args.items())
1755 add_dash_mpd(video_info)
1756 # Rental video is not rented but preview is available (e.g.
1757 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1758 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1759 if not video_info and args.get('ypc_vid'):
1760 return self.url_result(
1761 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1762 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1763 is_live = True
1764 sts = ytplayer_config.get('sts')
1765 if not player_response:
1766 player_response = extract_player_response(args.get('player_response'), video_id)
1767 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1768 add_dash_mpd_pr(player_response)
1769 # We also try looking in get_video_info since it may contain different dashmpd
1770 # URL that points to a DASH manifest with possibly different itag set (some itags
1771 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1772 # manifest pointed by get_video_info's dashmpd).
1773 # The general idea is to take a union of itags of both DASH manifests (for example
1774 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1775 self.report_video_info_webpage_download(video_id)
1776 for el in ('embedded', 'detailpage', 'vevo', ''):
1777 query = {
1778 'video_id': video_id,
1779 'ps': 'default',
1780 'eurl': '',
1781 'gl': 'US',
1782 'hl': 'en',
1783 }
1784 if el:
1785 query['el'] = el
1786 if sts:
1787 query['sts'] = sts
1788 video_info_webpage = self._download_webpage(
1789 '%s://www.youtube.com/get_video_info' % proto,
1790 video_id, note=False,
1791 errnote='unable to download video info webpage',
1792 fatal=False, query=query)
1793 if not video_info_webpage:
1794 continue
1795 get_video_info = compat_parse_qs(video_info_webpage)
1796 if not player_response:
1797 pl_response = get_video_info.get('player_response', [None])[0]
1798 player_response = extract_player_response(pl_response, video_id)
1799 add_dash_mpd(get_video_info)
1800 if view_count is None:
1801 view_count = extract_view_count(get_video_info)
1802 if not video_info:
1803 video_info = get_video_info
1804 get_token = extract_token(get_video_info)
1805 if get_token:
1806 # Different get_video_info requests may report different results, e.g.
1807 # some may report video unavailability, but some may serve it without
1808 # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1809 # the original webpage as well as el=info and el=embedded get_video_info
1810 # requests report video unavailability due to geo restriction while
1811 # el=detailpage succeeds and returns valid data). This is probably
1812 # due to YouTube measures against IP ranges of hosting providers.
1813 # Working around by preferring the first succeeded video_info containing
1814 # the token if no such video_info yet was found.
1815 token = extract_token(video_info)
1816 if not token:
1817 video_info = get_video_info
1818 break
1819
1820 def extract_unavailable_message():
1821 messages = []
1822 for tag, kind in (('h1', 'message'), ('div', 'submessage')):
1823 msg = self._html_search_regex(
1824 r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind),
1825 video_webpage, 'unavailable %s' % kind, default=None)
1826 if msg:
1827 messages.append(msg)
1828 if messages:
1829 return '\n'.join(messages)
1830
1831 if not video_info:
1832 unavailable_message = extract_unavailable_message()
1833 if not unavailable_message:
1834 unavailable_message = 'Unable to extract video data'
1835 raise ExtractorError(
1836 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1837
1838 video_details = try_get(
1839 player_response, lambda x: x['videoDetails'], dict) or {}
1840
1841 video_title = video_info.get('title', [None])[0] or video_details.get('title')
1842 if not video_title:
1843 self._downloader.report_warning('Unable to extract video title')
1844 video_title = '_'
1845
1846 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1847 if video_description:
1848
1849 def replace_url(m):
1850 redir_url = compat_urlparse.urljoin(url, m.group(1))
1851 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1852 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1853 qs = compat_parse_qs(parsed_redir_url.query)
1854 q = qs.get('q')
1855 if q and q[0]:
1856 return q[0]
1857 return redir_url
1858
1859 description_original = video_description = re.sub(r'''(?x)
1860 <a\s+
1861 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1862 (?:title|href)="([^"]+)"\s+
1863 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1864 class="[^"]*"[^>]*>
1865 [^<]+\.{3}\s*
1866 </a>
1867 ''', replace_url, video_description)
1868 video_description = clean_html(video_description)
1869 else:
1870 video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
1871
1872 if not smuggled_data.get('force_singlefeed', False):
1873 if not self._downloader.params.get('noplaylist'):
1874 multifeed_metadata_list = try_get(
1875 player_response,
1876 lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1877 compat_str) or try_get(
1878 video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1879 if multifeed_metadata_list:
1880 entries = []
1881 feed_ids = []
1882 for feed in multifeed_metadata_list.split(','):
1883 # Unquote should take place before split on comma (,) since textual
1884 # fields may contain comma as well (see
1885 # https://github.com/ytdl-org/youtube-dl/issues/8536)
1886 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1887 entries.append({
1888 '_type': 'url_transparent',
1889 'ie_key': 'Youtube',
1890 'url': smuggle_url(
1891 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1892 {'force_singlefeed': True}),
1893 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1894 })
1895 feed_ids.append(feed_data['id'][0])
1896 self.to_screen(
1897 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1898 % (', '.join(feed_ids), video_id))
1899 return self.playlist_result(entries, video_id, video_title, video_description)
1900 else:
1901 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1902
1903 if view_count is None:
1904 view_count = extract_view_count(video_info)
1905 if view_count is None and video_details:
1906 view_count = int_or_none(video_details.get('viewCount'))
1907
1908 if is_live is None:
1909 is_live = bool_or_none(video_details.get('isLive'))
1910
1911 # Check for "rental" videos
1912 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1913 raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1914
1915 def _extract_filesize(media_url):
1916 return int_or_none(self._search_regex(
1917 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1918
1919 streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or []
1920 streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or [])
1921
1922 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1923 self.report_rtmp_download()
1924 formats = [{
1925 'format_id': '_rtmp',
1926 'protocol': 'rtmp',
1927 'url': video_info['conn'][0],
1928 'player_url': player_url,
1929 }]
1930 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):
1931 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1932 if 'rtmpe%3Dyes' in encoded_url_map:
1933 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1934 formats = []
1935 formats_spec = {}
1936 fmt_list = video_info.get('fmt_list', [''])[0]
1937 if fmt_list:
1938 for fmt in fmt_list.split(','):
1939 spec = fmt.split('/')
1940 if len(spec) > 1:
1941 width_height = spec[1].split('x')
1942 if len(width_height) == 2:
1943 formats_spec[spec[0]] = {
1944 'resolution': spec[1],
1945 'width': int_or_none(width_height[0]),
1946 'height': int_or_none(width_height[1]),
1947 }
1948 for fmt in streaming_formats:
1949 itag = str_or_none(fmt.get('itag'))
1950 if not itag:
1951 continue
1952 quality = fmt.get('quality')
1953 quality_label = fmt.get('qualityLabel') or quality
1954 formats_spec[itag] = {
1955 'asr': int_or_none(fmt.get('audioSampleRate')),
1956 'filesize': int_or_none(fmt.get('contentLength')),
1957 'format_note': quality_label,
1958 'fps': int_or_none(fmt.get('fps')),
1959 'height': int_or_none(fmt.get('height')),
1960 # bitrate for itag 43 is always 2147483647
1961 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1962 'width': int_or_none(fmt.get('width')),
1963 }
1964
1965 for fmt in streaming_formats:
1966 if fmt.get('drm_families'):
1967 continue
1968 url = url_or_none(fmt.get('url'))
1969
1970 if not url:
1971 cipher = fmt.get('cipher')
1972 if not cipher:
1973 continue
1974 url_data = compat_parse_qs(cipher)
1975 url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str))
1976 if not url:
1977 continue
1978 else:
1979 cipher = None
1980 url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
1981
1982 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1983 # Unsupported FORMAT_STREAM_TYPE_OTF
1984 if stream_type == 3:
1985 continue
1986
1987 format_id = fmt.get('itag') or url_data['itag'][0]
1988 if not format_id:
1989 continue
1990 format_id = compat_str(format_id)
1991
1992 if cipher:
1993 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1994 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1995 jsplayer_url_json = self._search_regex(
1996 ASSETS_RE,
1997 embed_webpage if age_gate else video_webpage,
1998 'JS player URL (1)', default=None)
1999 if not jsplayer_url_json and not age_gate:
2000 # We need the embed website after all
2001 if embed_webpage is None:
2002 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
2003 embed_webpage = self._download_webpage(
2004 embed_url, video_id, 'Downloading embed webpage')
2005 jsplayer_url_json = self._search_regex(
2006 ASSETS_RE, embed_webpage, 'JS player URL')
2007
2008 player_url = json.loads(jsplayer_url_json)
2009 if player_url is None:
2010 player_url_json = self._search_regex(
2011 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
2012 video_webpage, 'age gate player URL')
2013 player_url = json.loads(player_url_json)
2014
2015 if 'sig' in url_data:
2016 url += '&signature=' + url_data['sig'][0]
2017 elif 's' in url_data:
2018 encrypted_sig = url_data['s'][0]
2019
2020 if self._downloader.params.get('verbose'):
2021 if player_url is None:
2022 player_version = 'unknown'
2023 player_desc = 'unknown'
2024 else:
2025 if player_url.endswith('swf'):
2026 player_version = self._search_regex(
2027 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
2028 'flash player', fatal=False)
2029 player_desc = 'flash player %s' % player_version
2030 else:
2031 player_version = self._search_regex(
2032 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
2033 r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
2034 player_url,
2035 'html5 player', fatal=False)
2036 player_desc = 'html5 player %s' % player_version
2037
2038 parts_sizes = self._signature_cache_id(encrypted_sig)
2039 self.to_screen('{%s} signature length %s, %s' %
2040 (format_id, parts_sizes, player_desc))
2041
2042 signature = self._decrypt_signature(
2043 encrypted_sig, video_id, player_url, age_gate)
2044 sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2045 url += '&%s=%s' % (sp, signature)
2046 if 'ratebypass' not in url:
2047 url += '&ratebypass=yes'
2048
2049 dct = {
2050 'format_id': format_id,
2051 'url': url,
2052 'player_url': player_url,
2053 }
2054 if format_id in self._formats:
2055 dct.update(self._formats[format_id])
2056 if format_id in formats_spec:
2057 dct.update(formats_spec[format_id])
2058
2059 # Some itags are not included in DASH manifest thus corresponding formats will
2060 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2061 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2062 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2063 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2064
2065 if width is None:
2066 width = int_or_none(fmt.get('width'))
2067 if height is None:
2068 height = int_or_none(fmt.get('height'))
2069
2070 filesize = int_or_none(url_data.get(
2071 'clen', [None])[0]) or _extract_filesize(url)
2072
2073 quality = url_data.get('quality', [None])[0] or fmt.get('quality')
2074 quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel')
2075
2076 tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000)
2077 or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None
2078 fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps'))
2079
2080 more_fields = {
2081 'filesize': filesize,
2082 'tbr': tbr,
2083 'width': width,
2084 'height': height,
2085 'fps': fps,
2086 'format_note': quality_label or quality,
2087 }
2088 for key, value in more_fields.items():
2089 if value:
2090 dct[key] = value
2091 type_ = url_data.get('type', [None])[0] or fmt.get('mimeType')
2092 if type_:
2093 type_split = type_.split(';')
2094 kind_ext = type_split[0].split('/')
2095 if len(kind_ext) == 2:
2096 kind, _ = kind_ext
2097 dct['ext'] = mimetype2ext(type_split[0])
2098 if kind in ('audio', 'video'):
2099 codecs = None
2100 for mobj in re.finditer(
2101 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2102 if mobj.group('key') == 'codecs':
2103 codecs = mobj.group('val')
2104 break
2105 if codecs:
2106 dct.update(parse_codecs(codecs))
2107 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2108 dct['downloader_options'] = {
2109 # Youtube throttles chunks >~10M
2110 'http_chunk_size': 10485760,
2111 }
2112 formats.append(dct)
2113 else:
2114 manifest_url = (
2115 url_or_none(try_get(
2116 player_response,
2117 lambda x: x['streamingData']['hlsManifestUrl'],
2118 compat_str))
2119 or url_or_none(try_get(
2120 video_info, lambda x: x['hlsvp'][0], compat_str)))
2121 if manifest_url:
2122 formats = []
2123 m3u8_formats = self._extract_m3u8_formats(
2124 manifest_url, video_id, 'mp4', fatal=False)
2125 for a_format in m3u8_formats:
2126 itag = self._search_regex(
2127 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2128 if itag:
2129 a_format['format_id'] = itag
2130 if itag in self._formats:
2131 dct = self._formats[itag].copy()
2132 dct.update(a_format)
2133 a_format = dct
2134 a_format['player_url'] = player_url
2135 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2136 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2137 formats.append(a_format)
2138 else:
2139 error_message = extract_unavailable_message()
2140 if not error_message:
2141 error_message = clean_html(try_get(
2142 player_response, lambda x: x['playabilityStatus']['reason'],
2143 compat_str))
2144 if not error_message:
2145 error_message = clean_html(
2146 try_get(video_info, lambda x: x['reason'][0], compat_str))
2147 if error_message:
2148 raise ExtractorError(error_message, expected=True)
2149 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2150
2151 # uploader
2152 video_uploader = try_get(
2153 video_info, lambda x: x['author'][0],
2154 compat_str) or str_or_none(video_details.get('author'))
2155 if video_uploader:
2156 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2157 else:
2158 self._downloader.report_warning('unable to extract uploader name')
2159
2160 # uploader_id
2161 video_uploader_id = None
2162 video_uploader_url = None
2163 mobj = re.search(
2164 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2165 video_webpage)
2166 if mobj is not None:
2167 video_uploader_id = mobj.group('uploader_id')
2168 video_uploader_url = mobj.group('uploader_url')
2169 else:
2170 self._downloader.report_warning('unable to extract uploader nickname')
2171
2172 channel_id = (
2173 str_or_none(video_details.get('channelId'))
2174 or self._html_search_meta(
2175 'channelId', video_webpage, 'channel id', default=None)
2176 or self._search_regex(
2177 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2178 video_webpage, 'channel id', default=None, group='id'))
2179 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2180
2181 # thumbnail image
2182 # We try first to get a high quality image:
2183 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2184 video_webpage, re.DOTALL)
2185 if m_thumb is not None:
2186 video_thumbnail = m_thumb.group(1)
2187 elif 'thumbnail_url' not in video_info:
2188 self._downloader.report_warning('unable to extract video thumbnail')
2189 video_thumbnail = None
2190 else: # don't panic if we can't find it
2191 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2192
2193 # upload date
2194 upload_date = self._html_search_meta(
2195 'datePublished', video_webpage, 'upload date', default=None)
2196 if not upload_date:
2197 upload_date = self._search_regex(
2198 [r'(?s)id="eow-date.*?>(.*?)</span>',
2199 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2200 video_webpage, 'upload date', default=None)
2201 upload_date = unified_strdate(upload_date)
2202
2203 video_license = self._html_search_regex(
2204 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2205 video_webpage, 'license', default=None)
2206
2207 m_music = re.search(
2208 r'''(?x)
2209 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2210 <ul[^>]*>\s*
2211 <li>(?P<title>.+?)
2212 by (?P<creator>.+?)
2213 (?:
2214 \(.+?\)|
2215 <a[^>]*
2216 (?:
2217 \bhref=["\']/red[^>]*>| # drop possible
2218 >\s*Listen ad-free with YouTube Red # YouTube Red ad
2219 )
2220 .*?
2221 )?</li
2222 ''',
2223 video_webpage)
2224 if m_music:
2225 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2226 video_creator = clean_html(m_music.group('creator'))
2227 else:
2228 video_alt_title = video_creator = None
2229
2230 def extract_meta(field):
2231 return self._html_search_regex(
2232 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2233 video_webpage, field, default=None)
2234
2235 track = extract_meta('Song')
2236 artist = extract_meta('Artist')
2237 album = extract_meta('Album')
2238
2239 # Youtube Music Auto-generated description
2240 release_date = release_year = None
2241 if video_description:
2242 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)
2243 if mobj:
2244 if not track:
2245 track = mobj.group('track').strip()
2246 if not artist:
2247 artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('Ā·'))
2248 if not album:
2249 album = mobj.group('album'.strip())
2250 release_year = mobj.group('release_year')
2251 release_date = mobj.group('release_date')
2252 if release_date:
2253 release_date = release_date.replace('-', '')
2254 if not release_year:
2255 release_year = int(release_date[:4])
2256 if release_year:
2257 release_year = int(release_year)
2258
2259 m_episode = re.search(
2260 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
2261 video_webpage)
2262 if m_episode:
2263 series = unescapeHTML(m_episode.group('series'))
2264 season_number = int(m_episode.group('season'))
2265 episode_number = int(m_episode.group('episode'))
2266 else:
2267 series = season_number = episode_number = None
2268
2269 m_cat_container = self._search_regex(
2270 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2271 video_webpage, 'categories', default=None)
2272 if m_cat_container:
2273 category = self._html_search_regex(
2274 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2275 default=None)
2276 video_categories = None if category is None else [category]
2277 else:
2278 video_categories = None
2279
2280 video_tags = [
2281 unescapeHTML(m.group('content'))
2282 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2283
2284 def _extract_count(count_name):
2285 return str_to_int(self._search_regex(
2286 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2287 % re.escape(count_name),
2288 video_webpage, count_name, default=None))
2289
2290 like_count = _extract_count('like')
2291 dislike_count = _extract_count('dislike')
2292
2293 if view_count is None:
2294 view_count = str_to_int(self._search_regex(
2295 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2296 'view count', default=None))
2297
2298 average_rating = (
2299 float_or_none(video_details.get('averageRating'))
2300 or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2301
2302 # subtitles
2303 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2304 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2305
2306 video_duration = try_get(
2307 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2308 if not video_duration:
2309 video_duration = int_or_none(video_details.get('lengthSeconds'))
2310 if not video_duration:
2311 video_duration = parse_duration(self._html_search_meta(
2312 'duration', video_webpage, 'video duration'))
2313
2314 # annotations
2315 video_annotations = None
2316 if self._downloader.params.get('writeannotations', False):
2317 xsrf_token = self._search_regex(
2318 r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>[A-Za-z0-9+/=]+)\2',
2319 video_webpage, 'xsrf token', group='xsrf_token', fatal=False)
2320 invideo_url = try_get(
2321 player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str)
2322 if xsrf_token and invideo_url:
2323 xsrf_field_name = self._search_regex(
2324 r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2',
2325 video_webpage, 'xsrf field name',
2326 group='xsrf_field_name', default='session_token')
2327 video_annotations = self._download_webpage(
2328 self._proto_relative_url(invideo_url),
2329 video_id, note='Downloading annotations',
2330 errnote='Unable to download video annotations', fatal=False,
2331 data=urlencode_postdata({xsrf_field_name: xsrf_token}))
2332
2333 chapters = self._extract_chapters(description_original, video_duration)
2334
2335 # Look for the DASH manifest
2336 if self._downloader.params.get('youtube_include_dash_manifest', True):
2337 dash_mpd_fatal = True
2338 for mpd_url in dash_mpds:
2339 dash_formats = {}
2340 try:
2341 def decrypt_sig(mobj):
2342 s = mobj.group(1)
2343 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2344 return '/signature/%s' % dec_s
2345
2346 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2347
2348 for df in self._extract_mpd_formats(
2349 mpd_url, video_id, fatal=dash_mpd_fatal,
2350 formats_dict=self._formats):
2351 if not df.get('filesize'):
2352 df['filesize'] = _extract_filesize(df['url'])
2353 # Do not overwrite DASH format found in some previous DASH manifest
2354 if df['format_id'] not in dash_formats:
2355 dash_formats[df['format_id']] = df
2356 # Additional DASH manifests may end up in HTTP Error 403 therefore
2357 # allow them to fail without bug report message if we already have
2358 # some DASH manifest succeeded. This is temporary workaround to reduce
2359 # burst of bug reports until we figure out the reason and whether it
2360 # can be fixed at all.
2361 dash_mpd_fatal = False
2362 except (ExtractorError, KeyError) as e:
2363 self.report_warning(
2364 'Skipping DASH manifest: %r' % e, video_id)
2365 if dash_formats:
2366 # Remove the formats we found through non-DASH, they
2367 # contain less info and it can be wrong, because we use
2368 # fixed values (for example the resolution). See
2369 # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2370 # example.
2371 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2372 formats.extend(dash_formats.values())
2373
2374 # Check for malformed aspect ratio
2375 stretched_m = re.search(
2376 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2377 video_webpage)
2378 if stretched_m:
2379 w = float(stretched_m.group('w'))
2380 h = float(stretched_m.group('h'))
2381 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2382 # We will only process correct ratios.
2383 if w > 0 and h > 0:
2384 ratio = w / h
2385 for f in formats:
2386 if f.get('vcodec') != 'none':
2387 f['stretched_ratio'] = ratio
2388
2389 if not formats:
2390 token = extract_token(video_info)
2391 if not token:
2392 if 'reason' in video_info:
2393 if 'The uploader has not made this video available in your country.' in video_info['reason']:
2394 regions_allowed = self._html_search_meta(
2395 'regionsAllowed', video_webpage, default=None)
2396 countries = regions_allowed.split(',') if regions_allowed else None
2397 self.raise_geo_restricted(
2398 msg=video_info['reason'][0], countries=countries)
2399 reason = video_info['reason'][0]
2400 if 'Invalid parameters' in reason:
2401 unavailable_message = extract_unavailable_message()
2402 if unavailable_message:
2403 reason = unavailable_message
2404 raise ExtractorError(
2405 'YouTube said: %s' % reason,
2406 expected=True, video_id=video_id)
2407 else:
2408 raise ExtractorError(
2409 '"token" parameter not in video info for unknown reason',
2410 video_id=video_id)
2411
2412 if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2413 raise ExtractorError('This video is DRM protected.', expected=True)
2414
2415 self._sort_formats(formats)
2416
2417 self.mark_watched(video_id, video_info, player_response)
2418
2419 return {
2420 'id': video_id,
2421 'uploader': video_uploader,
2422 'uploader_id': video_uploader_id,
2423 'uploader_url': video_uploader_url,
2424 'channel_id': channel_id,
2425 'channel_url': channel_url,
2426 'upload_date': upload_date,
2427 'license': video_license,
2428 'creator': video_creator or artist,
2429 'title': video_title,
2430 'alt_title': video_alt_title or track,
2431 'thumbnail': video_thumbnail,
2432 'description': video_description,
2433 'categories': video_categories,
2434 'tags': video_tags,
2435 'subtitles': video_subtitles,
2436 'automatic_captions': automatic_captions,
2437 'duration': video_duration,
2438 'age_limit': 18 if age_gate else 0,
2439 'annotations': video_annotations,
2440 'chapters': chapters,
2441 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2442 'view_count': view_count,
2443 'like_count': like_count,
2444 'dislike_count': dislike_count,
2445 'average_rating': average_rating,
2446 'formats': formats,
2447 'is_live': is_live,
2448 'start_time': start_time,
2449 'end_time': end_time,
2450 'series': series,
2451 'season_number': season_number,
2452 'episode_number': episode_number,
2453 'track': track,
2454 'artist': artist,
2455 'album': album,
2456 'release_date': release_date,
2457 'release_year': release_year,
2458 }
2459
2460
2461 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2462 IE_DESC = 'YouTube.com playlists'
2463 _VALID_URL = r"""(?x)(?:
2464 (?:https?://)?
2465 (?:\w+\.)?
2466 (?:
2467 (?:
2468 youtube\.com|
2469 invidio\.us
2470 )
2471 /
2472 (?:
2473 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2474 \? (?:.*?[&;])*? (?:p|a|list)=
2475 | p/
2476 )|
2477 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2478 )
2479 (
2480 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2481 # Top tracks, they can also include dots
2482 |(?:MC)[\w\.]*
2483 )
2484 .*
2485 |
2486 (%(playlist_id)s)
2487 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2488 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2489 _VIDEO_RE_TPL = r'href="\s*/watch\?v=%s(?:&amp;(?:[^"]*?index=(?P<index>\d+))?(?:[^>]+>(?P<title>[^<]+))?)?'
2490 _VIDEO_RE = _VIDEO_RE_TPL % r'(?P<id>[0-9A-Za-z_-]{11})'
2491 IE_NAME = 'youtube:playlist'
2492 _TESTS = [{
2493 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2494 'info_dict': {
2495 'title': 'ytdl test PL',
2496 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2497 },
2498 'playlist_count': 3,
2499 }, {
2500 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2501 'info_dict': {
2502 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2503 'title': 'YDL_Empty_List',
2504 },
2505 'playlist_count': 0,
2506 'skip': 'This playlist is private',
2507 }, {
2508 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2509 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2510 'info_dict': {
2511 'title': '29C3: Not my department',
2512 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2513 'uploader': 'Christiaan008',
2514 'uploader_id': 'ChRiStIaAn008',
2515 },
2516 'playlist_count': 95,
2517 }, {
2518 'note': 'issue #673',
2519 'url': 'PLBB231211A4F62143',
2520 'info_dict': {
2521 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2522 'id': 'PLBB231211A4F62143',
2523 'uploader': 'Wickydoo',
2524 'uploader_id': 'Wickydoo',
2525 },
2526 'playlist_mincount': 26,
2527 }, {
2528 'note': 'Large playlist',
2529 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2530 'info_dict': {
2531 'title': 'Uploads from Cauchemar',
2532 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2533 'uploader': 'Cauchemar',
2534 'uploader_id': 'Cauchemar89',
2535 },
2536 'playlist_mincount': 799,
2537 }, {
2538 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2539 'info_dict': {
2540 'title': 'YDL_safe_search',
2541 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2542 },
2543 'playlist_count': 2,
2544 'skip': 'This playlist is private',
2545 }, {
2546 'note': 'embedded',
2547 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2548 'playlist_count': 4,
2549 'info_dict': {
2550 'title': 'JODA15',
2551 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2552 'uploader': 'milan',
2553 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw',
2554 }
2555 }, {
2556 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2557 'playlist_mincount': 485,
2558 'info_dict': {
2559 'title': '2018 Chinese New Singles (11/6 updated)',
2560 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2561 'uploader': 'LBK',
2562 'uploader_id': 'sdragonfang',
2563 }
2564 }, {
2565 'note': 'Embedded SWF player',
2566 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2567 'playlist_count': 4,
2568 'info_dict': {
2569 'title': 'JODA7',
2570 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2571 },
2572 'skip': 'This playlist does not exist',
2573 }, {
2574 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2575 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2576 'info_dict': {
2577 'title': 'Uploads from Interstellar Movie',
2578 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2579 'uploader': 'Interstellar Movie',
2580 'uploader_id': 'InterstellarMovie1',
2581 },
2582 'playlist_mincount': 21,
2583 }, {
2584 # Playlist URL that does not actually serve a playlist
2585 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2586 'info_dict': {
2587 'id': 'FqZTN594JQw',
2588 'ext': 'webm',
2589 'title': "Smiley's People 01 detective, Adventure Series, Action",
2590 'uploader': 'STREEM',
2591 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2592 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2593 'upload_date': '20150526',
2594 'license': 'Standard YouTube License',
2595 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2596 'categories': ['People & Blogs'],
2597 'tags': list,
2598 'view_count': int,
2599 'like_count': int,
2600 'dislike_count': int,
2601 },
2602 'params': {
2603 'skip_download': True,
2604 },
2605 'skip': 'This video is not available.',
2606 'add_ie': [YoutubeIE.ie_key()],
2607 }, {
2608 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2609 'info_dict': {
2610 'id': 'yeWKywCrFtk',
2611 'ext': 'mp4',
2612 'title': 'Small Scale Baler and Braiding Rugs',
2613 'uploader': 'Backus-Page House Museum',
2614 'uploader_id': 'backuspagemuseum',
2615 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2616 'upload_date': '20161008',
2617 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2618 'categories': ['Nonprofits & Activism'],
2619 'tags': list,
2620 'like_count': int,
2621 'dislike_count': int,
2622 },
2623 'params': {
2624 'noplaylist': True,
2625 'skip_download': True,
2626 },
2627 }, {
2628 # https://github.com/ytdl-org/youtube-dl/issues/21844
2629 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2630 'info_dict': {
2631 'title': 'Data Analysis with Dr Mike Pound',
2632 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba',
2633 'uploader_id': 'Computerphile',
2634 'uploader': 'Computerphile',
2635 },
2636 'playlist_mincount': 11,
2637 }, {
2638 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2639 'only_matching': True,
2640 }, {
2641 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2642 'only_matching': True,
2643 }, {
2644 # music album playlist
2645 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2646 'only_matching': True,
2647 }, {
2648 'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2649 'only_matching': True,
2650 }]
2651
2652 def _real_initialize(self):
2653 self._login()
2654
2655 def extract_videos_from_page(self, page):
2656 ids_in_page = []
2657 titles_in_page = []
2658
2659 for item in re.findall(
2660 r'(<[^>]*\bdata-video-id\s*=\s*["\'][0-9A-Za-z_-]{11}[^>]+>)', page):
2661 attrs = extract_attributes(item)
2662 video_id = attrs['data-video-id']
2663 video_title = unescapeHTML(attrs.get('data-title'))
2664 if video_title:
2665 video_title = video_title.strip()
2666 ids_in_page.append(video_id)
2667 titles_in_page.append(video_title)
2668
2669 # Fallback with old _VIDEO_RE
2670 self.extract_videos_from_page_impl(
2671 self._VIDEO_RE, page, ids_in_page, titles_in_page)
2672
2673 # Relaxed fallbacks
2674 self.extract_videos_from_page_impl(
2675 r'href="\s*/watch\?v\s*=\s*(?P<id>[0-9A-Za-z_-]{11})', page,
2676 ids_in_page, titles_in_page)
2677 self.extract_videos_from_page_impl(
2678 r'data-video-ids\s*=\s*["\'](?P<id>[0-9A-Za-z_-]{11})', page,
2679 ids_in_page, titles_in_page)
2680
2681 return zip(ids_in_page, titles_in_page)
2682
2683 def _extract_mix(self, playlist_id):
2684 # The mixes are generated from a single video
2685 # the id of the playlist is just 'RD' + video_id
2686 ids = []
2687 last_id = playlist_id[-11:]
2688 for n in itertools.count(1):
2689 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2690 webpage = self._download_webpage(
2691 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2692 new_ids = orderedSet(re.findall(
2693 r'''(?xs)data-video-username=".*?".*?
2694 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2695 webpage))
2696 # Fetch new pages until all the videos are repeated, it seems that
2697 # there are always 51 unique videos.
2698 new_ids = [_id for _id in new_ids if _id not in ids]
2699 if not new_ids:
2700 break
2701 ids.extend(new_ids)
2702 last_id = ids[-1]
2703
2704 url_results = self._ids_to_results(ids)
2705
2706 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2707 title_span = (
2708 search_title('playlist-title')
2709 or search_title('title long-title')
2710 or search_title('title'))
2711 title = clean_html(title_span)
2712
2713 return self.playlist_result(url_results, playlist_id, title)
2714
2715 def _extract_playlist(self, playlist_id):
2716 url = self._TEMPLATE_URL % playlist_id
2717 page = self._download_webpage(url, playlist_id)
2718
2719 # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2720 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2721 match = match.strip()
2722 # Check if the playlist exists or is private
2723 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2724 if mobj:
2725 reason = mobj.group('reason')
2726 message = 'This playlist %s' % reason
2727 if 'private' in reason:
2728 message += ', use --username or --netrc to access it'
2729 message += '.'
2730 raise ExtractorError(message, expected=True)
2731 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2732 raise ExtractorError(
2733 'Invalid parameters. Maybe URL is incorrect.',
2734 expected=True)
2735 elif re.match(r'[^<]*Choose your language[^<]*', match):
2736 continue
2737 else:
2738 self.report_warning('Youtube gives an alert message: ' + match)
2739
2740 playlist_title = self._html_search_regex(
2741 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2742 page, 'title', default=None)
2743
2744 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2745 uploader = self._html_search_regex(
2746 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2747 page, 'uploader', default=None)
2748 mobj = re.search(
2749 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2750 page)
2751 if mobj:
2752 uploader_id = mobj.group('uploader_id')
2753 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2754 else:
2755 uploader_id = uploader_url = None
2756
2757 has_videos = True
2758
2759 if not playlist_title:
2760 try:
2761 # Some playlist URLs don't actually serve a playlist (e.g.
2762 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2763 next(self._entries(page, playlist_id))
2764 except StopIteration:
2765 has_videos = False
2766
2767 playlist = self.playlist_result(
2768 self._entries(page, playlist_id), playlist_id, playlist_title)
2769 playlist.update({
2770 'uploader': uploader,
2771 'uploader_id': uploader_id,
2772 'uploader_url': uploader_url,
2773 })
2774
2775 return has_videos, playlist
2776
2777 def _check_download_just_video(self, url, playlist_id):
2778 # Check if it's a video-specific URL
2779 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2780 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2781 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2782 'video id', default=None)
2783 if video_id:
2784 if self._downloader.params.get('noplaylist'):
2785 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2786 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2787 else:
2788 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2789 return video_id, None
2790 return None, None
2791
2792 def _real_extract(self, url):
2793 # Extract playlist id
2794 mobj = re.match(self._VALID_URL, url)
2795 if mobj is None:
2796 raise ExtractorError('Invalid URL: %s' % url)
2797 playlist_id = mobj.group(1) or mobj.group(2)
2798
2799 video_id, video = self._check_download_just_video(url, playlist_id)
2800 if video:
2801 return video
2802
2803 if playlist_id.startswith(('RD', 'UL', 'PU')):
2804 # Mixes require a custom extraction process
2805 return self._extract_mix(playlist_id)
2806
2807 has_videos, playlist = self._extract_playlist(playlist_id)
2808 if has_videos or not video_id:
2809 return playlist
2810
2811 # Some playlist URLs don't actually serve a playlist (see
2812 # https://github.com/ytdl-org/youtube-dl/issues/10537).
2813 # Fallback to plain video extraction if there is a video id
2814 # along with playlist id.
2815 return self.url_result(video_id, 'Youtube', video_id=video_id)
2816
2817
2818 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2819 IE_DESC = 'YouTube.com channels'
2820 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2821 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2822 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2823 IE_NAME = 'youtube:channel'
2824 _TESTS = [{
2825 'note': 'paginated channel',
2826 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2827 'playlist_mincount': 91,
2828 'info_dict': {
2829 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2830 'title': 'Uploads from lex will',
2831 'uploader': 'lex will',
2832 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
2833 }
2834 }, {
2835 'note': 'Age restricted channel',
2836 # from https://www.youtube.com/user/DeusExOfficial
2837 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2838 'playlist_mincount': 64,
2839 'info_dict': {
2840 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2841 'title': 'Uploads from Deus Ex',
2842 'uploader': 'Deus Ex',
2843 'uploader_id': 'DeusExOfficial',
2844 },
2845 }, {
2846 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2847 'only_matching': True,
2848 }]
2849
2850 @classmethod
2851 def suitable(cls, url):
2852 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2853 else super(YoutubeChannelIE, cls).suitable(url))
2854
2855 def _build_template_url(self, url, channel_id):
2856 return self._TEMPLATE_URL % channel_id
2857
2858 def _real_extract(self, url):
2859 channel_id = self._match_id(url)
2860
2861 url = self._build_template_url(url, channel_id)
2862
2863 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2864 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2865 # otherwise fallback on channel by page extraction
2866 channel_page = self._download_webpage(
2867 url + '?view=57', channel_id,
2868 'Downloading channel page', fatal=False)
2869 if channel_page is False:
2870 channel_playlist_id = False
2871 else:
2872 channel_playlist_id = self._html_search_meta(
2873 'channelId', channel_page, 'channel id', default=None)
2874 if not channel_playlist_id:
2875 channel_url = self._html_search_meta(
2876 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2877 channel_page, 'channel url', default=None)
2878 if channel_url:
2879 channel_playlist_id = self._search_regex(
2880 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2881 channel_url, 'channel id', default=None)
2882 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2883 playlist_id = 'UU' + channel_playlist_id[2:]
2884 return self.url_result(
2885 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2886
2887 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2888 autogenerated = re.search(r'''(?x)
2889 class="[^"]*?(?:
2890 channel-header-autogenerated-label|
2891 yt-channel-title-autogenerated
2892 )[^"]*"''', channel_page) is not None
2893
2894 if autogenerated:
2895 # The videos are contained in a single page
2896 # the ajax pages can't be used, they are empty
2897 entries = [
2898 self.url_result(
2899 video_id, 'Youtube', video_id=video_id,
2900 video_title=video_title)
2901 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2902 return self.playlist_result(entries, channel_id)
2903
2904 try:
2905 next(self._entries(channel_page, channel_id))
2906 except StopIteration:
2907 alert_message = self._html_search_regex(
2908 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2909 channel_page, 'alert', default=None, group='alert')
2910 if alert_message:
2911 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2912
2913 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2914
2915
2916 class YoutubeUserIE(YoutubeChannelIE):
2917 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2918 _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_-]+)'
2919 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2920 IE_NAME = 'youtube:user'
2921
2922 _TESTS = [{
2923 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2924 'playlist_mincount': 320,
2925 'info_dict': {
2926 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2927 'title': 'Uploads from The Linux Foundation',
2928 'uploader': 'The Linux Foundation',
2929 'uploader_id': 'TheLinuxFoundation',
2930 }
2931 }, {
2932 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2933 # but not https://www.youtube.com/user/12minuteathlete/videos
2934 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2935 'playlist_mincount': 249,
2936 'info_dict': {
2937 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2938 'title': 'Uploads from 12 Minute Athlete',
2939 'uploader': '12 Minute Athlete',
2940 'uploader_id': 'the12minuteathlete',
2941 }
2942 }, {
2943 'url': 'ytuser:phihag',
2944 'only_matching': True,
2945 }, {
2946 'url': 'https://www.youtube.com/c/gametrailers',
2947 'only_matching': True,
2948 }, {
2949 'url': 'https://www.youtube.com/gametrailers',
2950 'only_matching': True,
2951 }, {
2952 # This channel is not available, geo restricted to JP
2953 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2954 'only_matching': True,
2955 }]
2956
2957 @classmethod
2958 def suitable(cls, url):
2959 # Don't return True if the url can be extracted with other youtube
2960 # extractor, the regex would is too permissive and it would match.
2961 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2962 if any(ie.suitable(url) for ie in other_yt_ies):
2963 return False
2964 else:
2965 return super(YoutubeUserIE, cls).suitable(url)
2966
2967 def _build_template_url(self, url, channel_id):
2968 mobj = re.match(self._VALID_URL, url)
2969 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2970
2971
2972 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2973 IE_DESC = 'YouTube.com live streams'
2974 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2975 IE_NAME = 'youtube:live'
2976
2977 _TESTS = [{
2978 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2979 'info_dict': {
2980 'id': 'a48o2S1cPoo',
2981 'ext': 'mp4',
2982 'title': 'The Young Turks - Live Main Show',
2983 'uploader': 'The Young Turks',
2984 'uploader_id': 'TheYoungTurks',
2985 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2986 'upload_date': '20150715',
2987 'license': 'Standard YouTube License',
2988 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2989 'categories': ['News & Politics'],
2990 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2991 'like_count': int,
2992 'dislike_count': int,
2993 },
2994 'params': {
2995 'skip_download': True,
2996 },
2997 }, {
2998 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2999 'only_matching': True,
3000 }, {
3001 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
3002 'only_matching': True,
3003 }, {
3004 'url': 'https://www.youtube.com/TheYoungTurks/live',
3005 'only_matching': True,
3006 }]
3007
3008 def _real_extract(self, url):
3009 mobj = re.match(self._VALID_URL, url)
3010 channel_id = mobj.group('id')
3011 base_url = mobj.group('base_url')
3012 webpage = self._download_webpage(url, channel_id, fatal=False)
3013 if webpage:
3014 page_type = self._og_search_property(
3015 'type', webpage, 'page type', default='')
3016 video_id = self._html_search_meta(
3017 'videoId', webpage, 'video id', default=None)
3018 if page_type.startswith('video') and video_id and re.match(
3019 r'^[0-9A-Za-z_-]{11}$', video_id):
3020 return self.url_result(video_id, YoutubeIE.ie_key())
3021 return self.url_result(base_url)
3022
3023
3024 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
3025 IE_DESC = 'YouTube.com user/channel playlists'
3026 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
3027 IE_NAME = 'youtube:playlists'
3028
3029 _TESTS = [{
3030 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
3031 'playlist_mincount': 4,
3032 'info_dict': {
3033 'id': 'ThirstForScience',
3034 'title': 'ThirstForScience',
3035 },
3036 }, {
3037 # with "Load more" button
3038 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
3039 'playlist_mincount': 70,
3040 'info_dict': {
3041 'id': 'igorkle1',
3042 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
3043 },
3044 }, {
3045 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
3046 'playlist_mincount': 17,
3047 'info_dict': {
3048 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
3049 'title': 'Chem Player',
3050 },
3051 'skip': 'Blocked',
3052 }]
3053
3054
3055 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
3056 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
3057
3058
3059 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
3060 IE_DESC = 'YouTube.com searches'
3061 # there doesn't appear to be a real limit, for example if you search for
3062 # 'python' you get more than 8.000.000 results
3063 _MAX_RESULTS = float('inf')
3064 IE_NAME = 'youtube:search'
3065 _SEARCH_KEY = 'ytsearch'
3066 _EXTRA_QUERY_ARGS = {}
3067 _TESTS = []
3068
3069 def _get_n_results(self, query, n):
3070 """Get a specified number of results for a query"""
3071
3072 videos = []
3073 limit = n
3074
3075 url_query = {
3076 'search_query': query.encode('utf-8'),
3077 }
3078 url_query.update(self._EXTRA_QUERY_ARGS)
3079 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
3080
3081 for pagenum in itertools.count(1):
3082 data = self._download_json(
3083 result_url, video_id='query "%s"' % query,
3084 note='Downloading page %s' % pagenum,
3085 errnote='Unable to download API page',
3086 query={'spf': 'navigate'})
3087 html_content = data[1]['body']['content']
3088
3089 if 'class="search-message' in html_content:
3090 raise ExtractorError(
3091 '[youtube] No video results', expected=True)
3092
3093 new_videos = list(self._process_page(html_content))
3094 videos += new_videos
3095 if not new_videos or len(videos) > limit:
3096 break
3097 next_link = self._html_search_regex(
3098 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
3099 html_content, 'next link', default=None)
3100 if next_link is None:
3101 break
3102 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
3103
3104 if len(videos) > n:
3105 videos = videos[:n]
3106 return self.playlist_result(videos, query)
3107
3108
3109 class YoutubeSearchDateIE(YoutubeSearchIE):
3110 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
3111 _SEARCH_KEY = 'ytsearchdate'
3112 IE_DESC = 'YouTube.com searches, newest videos first'
3113 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
3114
3115
3116 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
3117 IE_DESC = 'YouTube.com search URLs'
3118 IE_NAME = 'youtube:search_url'
3119 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
3120 _TESTS = [{
3121 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
3122 'playlist_mincount': 5,
3123 'info_dict': {
3124 'title': 'youtube-dl test video',
3125 }
3126 }, {
3127 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3128 'only_matching': True,
3129 }]
3130
3131 def _real_extract(self, url):
3132 mobj = re.match(self._VALID_URL, url)
3133 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3134 webpage = self._download_webpage(url, query)
3135 return self.playlist_result(self._process_page(webpage), playlist_title=query)
3136
3137
3138 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3139 IE_DESC = 'YouTube.com (multi-season) shows'
3140 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3141 IE_NAME = 'youtube:show'
3142 _TESTS = [{
3143 'url': 'https://www.youtube.com/show/airdisasters',
3144 'playlist_mincount': 5,
3145 'info_dict': {
3146 'id': 'airdisasters',
3147 'title': 'Air Disasters',
3148 }
3149 }]
3150
3151 def _real_extract(self, url):
3152 playlist_id = self._match_id(url)
3153 return super(YoutubeShowIE, self)._real_extract(
3154 'https://www.youtube.com/show/%s/playlists' % playlist_id)
3155
3156
3157 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3158 """
3159 Base class for feed extractors
3160 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3161 """
3162 _LOGIN_REQUIRED = True
3163
3164 @property
3165 def IE_NAME(self):
3166 return 'youtube:%s' % self._FEED_NAME
3167
3168 def _real_initialize(self):
3169 self._login()
3170
3171 def _entries(self, page):
3172 # The extraction process is the same as for playlists, but the regex
3173 # for the video ids doesn't contain an index
3174 ids = []
3175 more_widget_html = content_html = page
3176 for page_num in itertools.count(1):
3177 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3178
3179 # 'recommended' feed has infinite 'load more' and each new portion spins
3180 # the same videos in (sometimes) slightly different order, so we'll check
3181 # for unicity and break when portion has no new videos
3182 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3183 if not new_ids:
3184 break
3185
3186 ids.extend(new_ids)
3187
3188 for entry in self._ids_to_results(new_ids):
3189 yield entry
3190
3191 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3192 if not mobj:
3193 break
3194
3195 more = self._download_json(
3196 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3197 'Downloading page #%s' % page_num,
3198 transform_source=uppercase_escape)
3199 content_html = more['content_html']
3200 more_widget_html = more['load_more_widget_html']
3201
3202 def _real_extract(self, url):
3203 page = self._download_webpage(
3204 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3205 self._PLAYLIST_TITLE)
3206 return self.playlist_result(
3207 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3208
3209
3210 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3211 IE_NAME = 'youtube:watchlater'
3212 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3213 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3214
3215 _TESTS = [{
3216 'url': 'https://www.youtube.com/playlist?list=WL',
3217 'only_matching': True,
3218 }, {
3219 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3220 'only_matching': True,
3221 }]
3222
3223 def _real_extract(self, url):
3224 _, video = self._check_download_just_video(url, 'WL')
3225 if video:
3226 return video
3227 _, playlist = self._extract_playlist('WL')
3228 return playlist
3229
3230
3231 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3232 IE_NAME = 'youtube:favorites'
3233 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3234 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3235 _LOGIN_REQUIRED = True
3236
3237 def _real_extract(self, url):
3238 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3239 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3240 return self.url_result(playlist_id, 'YoutubePlaylist')
3241
3242
3243 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3244 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3245 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3246 _FEED_NAME = 'recommended'
3247 _PLAYLIST_TITLE = 'Youtube Recommended videos'
3248
3249
3250 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3251 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3252 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3253 _FEED_NAME = 'subscriptions'
3254 _PLAYLIST_TITLE = 'Youtube Subscriptions'
3255
3256
3257 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3258 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3259 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3260 _FEED_NAME = 'history'
3261 _PLAYLIST_TITLE = 'Youtube History'
3262
3263
3264 class YoutubeTruncatedURLIE(InfoExtractor):
3265 IE_NAME = 'youtube:truncated_url'
3266 IE_DESC = False # Do not list
3267 _VALID_URL = r'''(?x)
3268 (?:https?://)?
3269 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3270 (?:watch\?(?:
3271 feature=[a-z_]+|
3272 annotation_id=annotation_[^&]+|
3273 x-yt-cl=[0-9]+|
3274 hl=[^&]*|
3275 t=[0-9]+
3276 )?
3277 |
3278 attribution_link\?a=[^&]+
3279 )
3280 $
3281 '''
3282
3283 _TESTS = [{
3284 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3285 'only_matching': True,
3286 }, {
3287 'url': 'https://www.youtube.com/watch?',
3288 'only_matching': True,
3289 }, {
3290 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3291 'only_matching': True,
3292 }, {
3293 'url': 'https://www.youtube.com/watch?feature=foo',
3294 'only_matching': True,
3295 }, {
3296 'url': 'https://www.youtube.com/watch?hl=en-GB',
3297 'only_matching': True,
3298 }, {
3299 'url': 'https://www.youtube.com/watch?t=2372',
3300 'only_matching': True,
3301 }]
3302
3303 def _real_extract(self, url):
3304 raise ExtractorError(
3305 'Did you forget to quote the URL? Remember that & is a meta '
3306 'character in most shells, so you want to put the URL in quotes, '
3307 'like youtube-dl '
3308 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3309 ' or simply youtube-dl BaW_jenozKc .',
3310 expected=True)
3311
3312
3313 class YoutubeTruncatedIDIE(InfoExtractor):
3314 IE_NAME = 'youtube:truncated_id'
3315 IE_DESC = False # Do not list
3316 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3317
3318 _TESTS = [{
3319 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3320 'only_matching': True,
3321 }]
3322
3323 def _real_extract(self, url):
3324 video_id = self._match_id(url)
3325 raise ExtractorError(
3326 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3327 expected=True)