]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
release 2018.11.03
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_kwargs,
20 compat_parse_qs,
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlencode,
24 compat_urllib_parse_urlparse,
25 compat_urlparse,
26 compat_str,
27 )
28 from ..utils import (
29 clean_html,
30 error_to_compat_str,
31 ExtractorError,
32 float_or_none,
33 get_element_by_attribute,
34 get_element_by_id,
35 int_or_none,
36 mimetype2ext,
37 orderedSet,
38 parse_codecs,
39 parse_duration,
40 qualities,
41 remove_quotes,
42 remove_start,
43 smuggle_url,
44 str_to_int,
45 try_get,
46 unescapeHTML,
47 unified_strdate,
48 unsmuggle_url,
49 uppercase_escape,
50 urlencode_postdata,
51 )
52
53
54 class YoutubeBaseInfoExtractor(InfoExtractor):
55 """Provide base functions for Youtube extractors"""
56 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
57 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
58
59 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
60 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
61 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
62
63 _NETRC_MACHINE = 'youtube'
64 # If True it will raise an error if no login info is provided
65 _LOGIN_REQUIRED = False
66
67 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
68
69 def _set_language(self):
70 self._set_cookie(
71 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
72 # YouTube sets the expire time to about two months
73 expire_time=time.time() + 2 * 30 * 24 * 3600)
74
75 def _ids_to_results(self, ids):
76 return [
77 self.url_result(vid_id, 'Youtube', video_id=vid_id)
78 for vid_id in ids]
79
80 def _login(self):
81 """
82 Attempt to log in to YouTube.
83 True is returned if successful or skipped.
84 False is returned if login failed.
85
86 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
87 """
88 username, password = self._get_login_info()
89 # No authentication to be performed
90 if username is None:
91 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
92 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
93 return True
94
95 login_page = self._download_webpage(
96 self._LOGIN_URL, None,
97 note='Downloading login page',
98 errnote='unable to fetch login page', fatal=False)
99 if login_page is False:
100 return
101
102 login_form = self._hidden_inputs(login_page)
103
104 def req(url, f_req, note, errnote):
105 data = login_form.copy()
106 data.update({
107 'pstMsg': 1,
108 'checkConnection': 'youtube',
109 'checkedDomains': 'youtube',
110 'hl': 'en',
111 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
112 'f.req': json.dumps(f_req),
113 'flowName': 'GlifWebSignIn',
114 'flowEntry': 'ServiceLogin',
115 })
116 return self._download_json(
117 url, None, note=note, errnote=errnote,
118 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
119 fatal=False,
120 data=urlencode_postdata(data), headers={
121 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
122 'Google-Accounts-XSRF': 1,
123 })
124
125 def warn(message):
126 self._downloader.report_warning(message)
127
128 lookup_req = [
129 username,
130 None, [], None, 'US', None, None, 2, False, True,
131 [
132 None, None,
133 [2, 1, None, 1,
134 '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',
135 None, [], 4],
136 1, [None, None, []], None, None, None, True
137 ],
138 username,
139 ]
140
141 lookup_results = req(
142 self._LOOKUP_URL, lookup_req,
143 'Looking up account info', 'Unable to look up account info')
144
145 if lookup_results is False:
146 return False
147
148 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
149 if not user_hash:
150 warn('Unable to extract user hash')
151 return False
152
153 challenge_req = [
154 user_hash,
155 None, 1, None, [1, None, None, None, [password, None, True]],
156 [
157 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],
158 1, [None, None, []], None, None, None, True
159 ]]
160
161 challenge_results = req(
162 self._CHALLENGE_URL, challenge_req,
163 'Logging in', 'Unable to log in')
164
165 if challenge_results is False:
166 return
167
168 login_res = try_get(challenge_results, lambda x: x[0][5], list)
169 if login_res:
170 login_msg = try_get(login_res, lambda x: x[5], compat_str)
171 warn(
172 'Unable to login: %s' % 'Invalid password'
173 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
174 return False
175
176 res = try_get(challenge_results, lambda x: x[0][-1], list)
177 if not res:
178 warn('Unable to extract result entry')
179 return False
180
181 login_challenge = try_get(res, lambda x: x[0][0], list)
182 if login_challenge:
183 challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
184 if challenge_str == 'TWO_STEP_VERIFICATION':
185 # SEND_SUCCESS - TFA code has been successfully sent to phone
186 # QUOTA_EXCEEDED - reached the limit of TFA codes
187 status = try_get(login_challenge, lambda x: x[5], compat_str)
188 if status == 'QUOTA_EXCEEDED':
189 warn('Exceeded the limit of TFA codes, try later')
190 return False
191
192 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
193 if not tl:
194 warn('Unable to extract TL')
195 return False
196
197 tfa_code = self._get_tfa_info('2-step verification code')
198
199 if not tfa_code:
200 warn(
201 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
202 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
203 return False
204
205 tfa_code = remove_start(tfa_code, 'G-')
206
207 tfa_req = [
208 user_hash, None, 2, None,
209 [
210 9, None, None, None, None, None, None, None,
211 [None, tfa_code, True, 2]
212 ]]
213
214 tfa_results = req(
215 self._TFA_URL.format(tl), tfa_req,
216 'Submitting TFA code', 'Unable to submit TFA code')
217
218 if tfa_results is False:
219 return False
220
221 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
222 if tfa_res:
223 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
224 warn(
225 'Unable to finish TFA: %s' % 'Invalid TFA code'
226 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
227 return False
228
229 check_cookie_url = try_get(
230 tfa_results, lambda x: x[0][-1][2], compat_str)
231 else:
232 CHALLENGES = {
233 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
234 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
235 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
236 }
237 challenge = CHALLENGES.get(
238 challenge_str,
239 '%s returned error %s.' % (self.IE_NAME, challenge_str))
240 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
241 return False
242 else:
243 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
244
245 if not check_cookie_url:
246 warn('Unable to extract CheckCookie URL')
247 return False
248
249 check_cookie_results = self._download_webpage(
250 check_cookie_url, None, 'Checking cookie', fatal=False)
251
252 if check_cookie_results is False:
253 return False
254
255 if 'https://myaccount.google.com/' not in check_cookie_results:
256 warn('Unable to log in')
257 return False
258
259 return True
260
261 def _download_webpage_handle(self, *args, **kwargs):
262 query = kwargs.get('query', {}).copy()
263 query['disable_polymer'] = 'true'
264 kwargs['query'] = query
265 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
266 *args, **compat_kwargs(kwargs))
267
268 def _real_initialize(self):
269 if self._downloader is None:
270 return
271 self._set_language()
272 if not self._login():
273 return
274
275
276 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
277 # Extract entries from page with "Load more" button
278 def _entries(self, page, playlist_id):
279 more_widget_html = content_html = page
280 for page_num in itertools.count(1):
281 for entry in self._process_page(content_html):
282 yield entry
283
284 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
285 if not mobj:
286 break
287
288 more = self._download_json(
289 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
290 'Downloading page #%s' % page_num,
291 transform_source=uppercase_escape)
292 content_html = more['content_html']
293 if not content_html.strip():
294 # Some webpages show a "Load more" button but they don't
295 # have more videos
296 break
297 more_widget_html = more['load_more_widget_html']
298
299
300 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
301 def _process_page(self, content):
302 for video_id, video_title in self.extract_videos_from_page(content):
303 yield self.url_result(video_id, 'Youtube', video_id, video_title)
304
305 def extract_videos_from_page(self, page):
306 ids_in_page = []
307 titles_in_page = []
308 for mobj in re.finditer(self._VIDEO_RE, page):
309 # The link with index 0 is not the first video of the playlist (not sure if still actual)
310 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
311 continue
312 video_id = mobj.group('id')
313 video_title = unescapeHTML(mobj.group('title'))
314 if video_title:
315 video_title = video_title.strip()
316 try:
317 idx = ids_in_page.index(video_id)
318 if video_title and not titles_in_page[idx]:
319 titles_in_page[idx] = video_title
320 except ValueError:
321 ids_in_page.append(video_id)
322 titles_in_page.append(video_title)
323 return zip(ids_in_page, titles_in_page)
324
325
326 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
327 def _process_page(self, content):
328 for playlist_id in orderedSet(re.findall(
329 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
330 content)):
331 yield self.url_result(
332 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
333
334 def _real_extract(self, url):
335 playlist_id = self._match_id(url)
336 webpage = self._download_webpage(url, playlist_id)
337 title = self._og_search_title(webpage, fatal=False)
338 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
339
340
341 class YoutubeIE(YoutubeBaseInfoExtractor):
342 IE_DESC = 'YouTube.com'
343 _VALID_URL = r"""(?x)^
344 (
345 (?:https?://|//) # http(s):// or protocol-independent URL
346 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
347 (?:www\.)?deturl\.com/www\.youtube\.com/|
348 (?:www\.)?pwnyoutube\.com/|
349 (?:www\.)?hooktube\.com/|
350 (?:www\.)?yourepeat\.com/|
351 tube\.majestyc\.net/|
352 (?:www\.)?invidio\.us/|
353 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
354 (?:.*?\#/)? # handle anchor (#/) redirect urls
355 (?: # the various things that can precede the ID:
356 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
357 |(?: # or the v= param in all its forms
358 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
359 (?:\?|\#!?) # the params delimiter ? or # or #!
360 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
361 v=
362 )
363 ))
364 |(?:
365 youtu\.be| # just youtu.be/xxxx
366 vid\.plus| # or vid.plus/xxxx
367 zwearz\.com/watch| # or zwearz.com/watch/xxxx
368 )/
369 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
370 )
371 )? # all until now is optional -> you can pass the naked ID
372 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
373 (?!.*?\blist=
374 (?:
375 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
376 WL # WL are handled by the watch later IE
377 )
378 )
379 (?(1).+)? # if we found the ID, everything can follow
380 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
381 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
382 _formats = {
383 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
384 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
385 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
386 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
387 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
388 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
389 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
390 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
391 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
392 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
393 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
394 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
395 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
396 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
397 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
398 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
399 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
400 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
401
402
403 # 3D videos
404 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
405 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
406 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
407 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
408 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
409 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
410 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
411
412 # Apple HTTP Live Streaming
413 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
414 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
415 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
416 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
417 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
418 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
419 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
420 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
421
422 # DASH mp4 video
423 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
424 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
425 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
426 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
427 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
428 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
429 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
430 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
431 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
432 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
433 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
434 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
435
436 # Dash mp4 audio
437 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
438 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
439 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
440 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
441 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
442 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
443 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
444
445 # Dash webm
446 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
447 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
448 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
449 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
450 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
451 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
452 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
453 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
454 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
455 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
456 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
457 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
458 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
459 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
460 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
461 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
462 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
463 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
464 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
465 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
466 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
467 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
468
469 # Dash webm audio
470 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
471 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
472
473 # Dash webm audio with opus inside
474 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
475 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
476 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
477
478 # RTMP (unnamed)
479 '_rtmp': {'protocol': 'rtmp'},
480 }
481 _SUBTITLE_FORMATS = ('ttml', 'vtt')
482
483 _GEO_BYPASS = False
484
485 IE_NAME = 'youtube'
486 _TESTS = [
487 {
488 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
489 'info_dict': {
490 'id': 'BaW_jenozKc',
491 'ext': 'mp4',
492 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
493 'uploader': 'Philipp Hagemeister',
494 'uploader_id': 'phihag',
495 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
496 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
497 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
498 'upload_date': '20121002',
499 'license': 'Standard YouTube License',
500 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
501 'categories': ['Science & Technology'],
502 'tags': ['youtube-dl'],
503 'duration': 10,
504 'like_count': int,
505 'dislike_count': int,
506 'start_time': 1,
507 'end_time': 9,
508 }
509 },
510 {
511 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
512 'note': 'Test generic use_cipher_signature video (#897)',
513 'info_dict': {
514 'id': 'UxxajLWwzqY',
515 'ext': 'mp4',
516 'upload_date': '20120506',
517 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
518 'alt_title': 'I Love It (feat. Charli XCX)',
519 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
520 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
521 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
522 'iconic ep', 'iconic', 'love', 'it'],
523 'duration': 180,
524 'uploader': 'Icona Pop',
525 'uploader_id': 'IconaPop',
526 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
527 'license': 'Standard YouTube License',
528 'creator': 'Icona Pop',
529 'track': 'I Love It (feat. Charli XCX)',
530 'artist': 'Icona Pop',
531 }
532 },
533 {
534 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
535 'note': 'Test VEVO video with age protection (#956)',
536 'info_dict': {
537 'id': '07FYdnEawAQ',
538 'ext': 'mp4',
539 'upload_date': '20130703',
540 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
541 'alt_title': 'Tunnel Vision',
542 'description': 'md5:64249768eec3bc4276236606ea996373',
543 'duration': 419,
544 'uploader': 'justintimberlakeVEVO',
545 'uploader_id': 'justintimberlakeVEVO',
546 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
547 'license': 'Standard YouTube License',
548 'creator': 'Justin Timberlake',
549 'track': 'Tunnel Vision',
550 'artist': 'Justin Timberlake',
551 'age_limit': 18,
552 }
553 },
554 {
555 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
556 'note': 'Embed-only video (#1746)',
557 'info_dict': {
558 'id': 'yZIXLfi8CZQ',
559 'ext': 'mp4',
560 'upload_date': '20120608',
561 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
562 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
563 'uploader': 'SET India',
564 'uploader_id': 'setindia',
565 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
566 'license': 'Standard YouTube License',
567 'age_limit': 18,
568 }
569 },
570 {
571 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
572 'note': 'Use the first video ID in the URL',
573 'info_dict': {
574 'id': 'BaW_jenozKc',
575 'ext': 'mp4',
576 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
577 'uploader': 'Philipp Hagemeister',
578 'uploader_id': 'phihag',
579 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
580 'upload_date': '20121002',
581 'license': 'Standard YouTube License',
582 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
583 'categories': ['Science & Technology'],
584 'tags': ['youtube-dl'],
585 'duration': 10,
586 'like_count': int,
587 'dislike_count': int,
588 },
589 'params': {
590 'skip_download': True,
591 },
592 },
593 {
594 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
595 'note': '256k DASH audio (format 141) via DASH manifest',
596 'info_dict': {
597 'id': 'a9LDPn-MO4I',
598 'ext': 'm4a',
599 'upload_date': '20121002',
600 'uploader_id': '8KVIDEO',
601 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
602 'description': '',
603 'uploader': '8KVIDEO',
604 'license': 'Standard YouTube License',
605 'title': 'UHDTV TEST 8K VIDEO.mp4'
606 },
607 'params': {
608 'youtube_include_dash_manifest': True,
609 'format': '141',
610 },
611 'skip': 'format 141 not served anymore',
612 },
613 # DASH manifest with encrypted signature
614 {
615 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
616 'info_dict': {
617 'id': 'IB3lcPjvWLA',
618 'ext': 'm4a',
619 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
620 'description': 'md5:1900ed86ee514927b9e00fbead6969a5',
621 'duration': 244,
622 'uploader': 'AfrojackVEVO',
623 'uploader_id': 'AfrojackVEVO',
624 'upload_date': '20131011',
625 'license': 'Standard YouTube License',
626 },
627 'params': {
628 'youtube_include_dash_manifest': True,
629 'format': '141/bestaudio[ext=m4a]',
630 },
631 },
632 # JS player signature function name containing $
633 {
634 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
635 'info_dict': {
636 'id': 'nfWlot6h_JM',
637 'ext': 'm4a',
638 'title': 'Taylor Swift - Shake It Off',
639 'alt_title': 'Shake It Off',
640 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
641 'duration': 242,
642 'uploader': 'TaylorSwiftVEVO',
643 'uploader_id': 'TaylorSwiftVEVO',
644 'upload_date': '20140818',
645 'license': 'Standard YouTube License',
646 'creator': 'Taylor Swift',
647 },
648 'params': {
649 'youtube_include_dash_manifest': True,
650 'format': '141/bestaudio[ext=m4a]',
651 },
652 },
653 # Controversy video
654 {
655 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
656 'info_dict': {
657 'id': 'T4XJQO3qol8',
658 'ext': 'mp4',
659 'duration': 219,
660 'upload_date': '20100909',
661 'uploader': 'TJ Kirk',
662 'uploader_id': 'TheAmazingAtheist',
663 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
664 'license': 'Standard YouTube License',
665 'title': 'Burning Everyone\'s Koran',
666 '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',
667 }
668 },
669 # Normal age-gate video (No vevo, embed allowed)
670 {
671 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
672 'info_dict': {
673 'id': 'HtVdAasjOgU',
674 'ext': 'mp4',
675 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
676 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
677 'duration': 142,
678 'uploader': 'The Witcher',
679 'uploader_id': 'WitcherGame',
680 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
681 'upload_date': '20140605',
682 'license': 'Standard YouTube License',
683 'age_limit': 18,
684 },
685 },
686 # Age-gate video with encrypted signature
687 {
688 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
689 'info_dict': {
690 'id': '6kLq3WMV1nU',
691 'ext': 'webm',
692 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
693 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
694 'duration': 246,
695 'uploader': 'LloydVEVO',
696 'uploader_id': 'LloydVEVO',
697 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
698 'upload_date': '20110629',
699 'license': 'Standard YouTube License',
700 'age_limit': 18,
701 },
702 },
703 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
704 # YouTube Red ad is not captured for creator
705 {
706 'url': '__2ABJjxzNo',
707 'info_dict': {
708 'id': '__2ABJjxzNo',
709 'ext': 'mp4',
710 'duration': 266,
711 'upload_date': '20100430',
712 'uploader_id': 'deadmau5',
713 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
714 'creator': 'deadmau5',
715 'description': 'md5:12c56784b8032162bb936a5f76d55360',
716 'uploader': 'deadmau5',
717 'license': 'Standard YouTube License',
718 'title': 'Deadmau5 - Some Chords (HD)',
719 'alt_title': 'Some Chords',
720 },
721 'expected_warnings': [
722 'DASH manifest missing',
723 ]
724 },
725 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
726 {
727 'url': 'lqQg6PlCWgI',
728 'info_dict': {
729 'id': 'lqQg6PlCWgI',
730 'ext': 'mp4',
731 'duration': 6085,
732 'upload_date': '20150827',
733 'uploader_id': 'olympic',
734 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
735 'license': 'Standard YouTube License',
736 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
737 'uploader': 'Olympic',
738 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
739 },
740 'params': {
741 'skip_download': 'requires avconv',
742 }
743 },
744 # Non-square pixels
745 {
746 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
747 'info_dict': {
748 'id': '_b-2C3KPAM0',
749 'ext': 'mp4',
750 'stretched_ratio': 16 / 9.,
751 'duration': 85,
752 'upload_date': '20110310',
753 'uploader_id': 'AllenMeow',
754 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
755 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
756 'uploader': '孫ᄋᄅ',
757 'license': 'Standard YouTube License',
758 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
759 },
760 },
761 # url_encoded_fmt_stream_map is empty string
762 {
763 'url': 'qEJwOuvDf7I',
764 'info_dict': {
765 'id': 'qEJwOuvDf7I',
766 'ext': 'webm',
767 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
768 'description': '',
769 'upload_date': '20150404',
770 'uploader_id': 'spbelect',
771 'uploader': 'Наблюдатели Петербурга',
772 },
773 'params': {
774 'skip_download': 'requires avconv',
775 },
776 'skip': 'This live event has ended.',
777 },
778 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
779 {
780 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
781 'info_dict': {
782 'id': 'FIl7x6_3R5Y',
783 'ext': 'webm',
784 'title': 'md5:7b81415841e02ecd4313668cde88737a',
785 'description': 'md5:116377fd2963b81ec4ce64b542173306',
786 'duration': 220,
787 'upload_date': '20150625',
788 'uploader_id': 'dorappi2000',
789 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
790 'uploader': 'dorappi2000',
791 'license': 'Standard YouTube License',
792 'formats': 'mincount:31',
793 },
794 'skip': 'not actual anymore',
795 },
796 # DASH manifest with segment_list
797 {
798 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
799 'md5': '8ce563a1d667b599d21064e982ab9e31',
800 'info_dict': {
801 'id': 'CsmdDsKjzN8',
802 'ext': 'mp4',
803 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
804 'uploader': 'Airtek',
805 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
806 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
807 'license': 'Standard YouTube License',
808 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
809 },
810 'params': {
811 'youtube_include_dash_manifest': True,
812 'format': '135', # bestvideo
813 },
814 'skip': 'This live event has ended.',
815 },
816 {
817 # Multifeed videos (multiple cameras), URL is for Main Camera
818 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
819 'info_dict': {
820 'id': 'jqWvoWXjCVs',
821 'title': 'teamPGP: Rocket League Noob Stream',
822 'description': 'md5:dc7872fb300e143831327f1bae3af010',
823 },
824 'playlist': [{
825 'info_dict': {
826 'id': 'jqWvoWXjCVs',
827 'ext': 'mp4',
828 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
829 'description': 'md5:dc7872fb300e143831327f1bae3af010',
830 'duration': 7335,
831 'upload_date': '20150721',
832 'uploader': 'Beer Games Beer',
833 'uploader_id': 'beergamesbeer',
834 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
835 'license': 'Standard YouTube License',
836 },
837 }, {
838 'info_dict': {
839 'id': '6h8e8xoXJzg',
840 'ext': 'mp4',
841 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
842 'description': 'md5:dc7872fb300e143831327f1bae3af010',
843 'duration': 7337,
844 'upload_date': '20150721',
845 'uploader': 'Beer Games Beer',
846 'uploader_id': 'beergamesbeer',
847 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
848 'license': 'Standard YouTube License',
849 },
850 }, {
851 'info_dict': {
852 'id': 'PUOgX5z9xZw',
853 'ext': 'mp4',
854 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
855 'description': 'md5:dc7872fb300e143831327f1bae3af010',
856 'duration': 7337,
857 'upload_date': '20150721',
858 'uploader': 'Beer Games Beer',
859 'uploader_id': 'beergamesbeer',
860 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
861 'license': 'Standard YouTube License',
862 },
863 }, {
864 'info_dict': {
865 'id': 'teuwxikvS5k',
866 'ext': 'mp4',
867 'title': 'teamPGP: Rocket League Noob Stream (zim)',
868 'description': 'md5:dc7872fb300e143831327f1bae3af010',
869 'duration': 7334,
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 'params': {
878 'skip_download': True,
879 },
880 },
881 {
882 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
883 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
884 'info_dict': {
885 'id': 'gVfLd0zydlo',
886 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
887 },
888 'playlist_count': 2,
889 'skip': 'Not multifeed anymore',
890 },
891 {
892 'url': 'https://vid.plus/FlRa-iH7PGw',
893 'only_matching': True,
894 },
895 {
896 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
897 'only_matching': True,
898 },
899 {
900 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
901 # Also tests cut-off URL expansion in video description (see
902 # https://github.com/rg3/youtube-dl/issues/1892,
903 # https://github.com/rg3/youtube-dl/issues/8164)
904 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
905 'info_dict': {
906 'id': 'lsguqyKfVQg',
907 'ext': 'mp4',
908 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
909 'alt_title': 'Dark Walk - Position Music',
910 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
911 'duration': 133,
912 'upload_date': '20151119',
913 'uploader_id': 'IronSoulElf',
914 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
915 'uploader': 'IronSoulElf',
916 'license': 'Standard YouTube License',
917 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
918 'track': 'Dark Walk - Position Music',
919 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
920 },
921 'params': {
922 'skip_download': True,
923 },
924 },
925 {
926 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
927 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
928 'only_matching': True,
929 },
930 {
931 # Video with yt:stretch=17:0
932 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
933 'info_dict': {
934 'id': 'Q39EVAstoRM',
935 'ext': 'mp4',
936 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
937 'description': 'md5:ee18a25c350637c8faff806845bddee9',
938 'upload_date': '20151107',
939 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
940 'uploader': 'CH GAMER DROID',
941 },
942 'params': {
943 'skip_download': True,
944 },
945 'skip': 'This video does not exist.',
946 },
947 {
948 # Video licensed under Creative Commons
949 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
950 'info_dict': {
951 'id': 'M4gD1WSo5mA',
952 'ext': 'mp4',
953 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
954 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
955 'duration': 721,
956 'upload_date': '20150127',
957 'uploader_id': 'BerkmanCenter',
958 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
959 'uploader': 'The Berkman Klein Center for Internet & Society',
960 'license': 'Creative Commons Attribution license (reuse allowed)',
961 },
962 'params': {
963 'skip_download': True,
964 },
965 },
966 {
967 # Channel-like uploader_url
968 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
969 'info_dict': {
970 'id': 'eQcmzGIKrzg',
971 'ext': 'mp4',
972 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
973 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
974 'duration': 4060,
975 'upload_date': '20151119',
976 'uploader': 'Bernie Sanders',
977 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
978 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
979 'license': 'Creative Commons Attribution license (reuse allowed)',
980 },
981 'params': {
982 'skip_download': True,
983 },
984 },
985 {
986 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
987 'only_matching': True,
988 },
989 {
990 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
991 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
992 'only_matching': True,
993 },
994 {
995 # Rental video preview
996 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
997 'info_dict': {
998 'id': 'uGpuVWrhIzE',
999 'ext': 'mp4',
1000 'title': 'Piku - Trailer',
1001 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1002 'upload_date': '20150811',
1003 'uploader': 'FlixMatrix',
1004 'uploader_id': 'FlixMatrixKaravan',
1005 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1006 'license': 'Standard YouTube License',
1007 },
1008 'params': {
1009 'skip_download': True,
1010 },
1011 'skip': 'This video is not available.',
1012 },
1013 {
1014 # YouTube Red video with episode data
1015 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1016 'info_dict': {
1017 'id': 'iqKdEhx-dD4',
1018 'ext': 'mp4',
1019 'title': 'Isolation - Mind Field (Ep 1)',
1020 'description': 'md5:25b78d2f64ae81719f5c96319889b736',
1021 'duration': 2085,
1022 'upload_date': '20170118',
1023 'uploader': 'Vsauce',
1024 'uploader_id': 'Vsauce',
1025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1026 'license': 'Standard YouTube License',
1027 'series': 'Mind Field',
1028 'season_number': 1,
1029 'episode_number': 1,
1030 },
1031 'params': {
1032 'skip_download': True,
1033 },
1034 'expected_warnings': [
1035 'Skipping DASH manifest',
1036 ],
1037 },
1038 {
1039 # The following content has been identified by the YouTube community
1040 # as inappropriate or offensive to some audiences.
1041 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1042 'info_dict': {
1043 'id': '6SJNVb0GnPI',
1044 'ext': 'mp4',
1045 'title': 'Race Differences in Intelligence',
1046 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1047 'duration': 965,
1048 'upload_date': '20140124',
1049 'uploader': 'New Century Foundation',
1050 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1051 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1052 'license': 'Standard YouTube License',
1053 },
1054 'params': {
1055 'skip_download': True,
1056 },
1057 },
1058 {
1059 # itag 212
1060 'url': '1t24XAntNCY',
1061 'only_matching': True,
1062 },
1063 {
1064 # geo restricted to JP
1065 'url': 'sJL6WA-aGkQ',
1066 'only_matching': True,
1067 },
1068 {
1069 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1070 'only_matching': True,
1071 },
1072 {
1073 'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1074 'only_matching': True,
1075 },
1076 ]
1077
1078 def __init__(self, *args, **kwargs):
1079 super(YoutubeIE, self).__init__(*args, **kwargs)
1080 self._player_cache = {}
1081
1082 def report_video_info_webpage_download(self, video_id):
1083 """Report attempt to download video info webpage."""
1084 self.to_screen('%s: Downloading video info webpage' % video_id)
1085
1086 def report_information_extraction(self, video_id):
1087 """Report attempt to extract video information."""
1088 self.to_screen('%s: Extracting video information' % video_id)
1089
1090 def report_unavailable_format(self, video_id, format):
1091 """Report extracted video URL."""
1092 self.to_screen('%s: Format %s not available' % (video_id, format))
1093
1094 def report_rtmp_download(self):
1095 """Indicate the download will use the RTMP protocol."""
1096 self.to_screen('RTMP download detected')
1097
1098 def _signature_cache_id(self, example_sig):
1099 """ Return a string representation of a signature """
1100 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1101
1102 def _extract_signature_function(self, video_id, player_url, example_sig):
1103 id_m = re.match(
1104 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1105 player_url)
1106 if not id_m:
1107 raise ExtractorError('Cannot identify player %r' % player_url)
1108 player_type = id_m.group('ext')
1109 player_id = id_m.group('id')
1110
1111 # Read from filesystem cache
1112 func_id = '%s_%s_%s' % (
1113 player_type, player_id, self._signature_cache_id(example_sig))
1114 assert os.path.basename(func_id) == func_id
1115
1116 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1117 if cache_spec is not None:
1118 return lambda s: ''.join(s[i] for i in cache_spec)
1119
1120 download_note = (
1121 'Downloading player %s' % player_url
1122 if self._downloader.params.get('verbose') else
1123 'Downloading %s player %s' % (player_type, player_id)
1124 )
1125 if player_type == 'js':
1126 code = self._download_webpage(
1127 player_url, video_id,
1128 note=download_note,
1129 errnote='Download of %s failed' % player_url)
1130 res = self._parse_sig_js(code)
1131 elif player_type == 'swf':
1132 urlh = self._request_webpage(
1133 player_url, video_id,
1134 note=download_note,
1135 errnote='Download of %s failed' % player_url)
1136 code = urlh.read()
1137 res = self._parse_sig_swf(code)
1138 else:
1139 assert False, 'Invalid player type %r' % player_type
1140
1141 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1142 cache_res = res(test_string)
1143 cache_spec = [ord(c) for c in cache_res]
1144
1145 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1146 return res
1147
1148 def _print_sig_code(self, func, example_sig):
1149 def gen_sig_code(idxs):
1150 def _genslice(start, end, step):
1151 starts = '' if start == 0 else str(start)
1152 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1153 steps = '' if step == 1 else (':%d' % step)
1154 return 's[%s%s%s]' % (starts, ends, steps)
1155
1156 step = None
1157 # Quelch pyflakes warnings - start will be set when step is set
1158 start = '(Never used)'
1159 for i, prev in zip(idxs[1:], idxs[:-1]):
1160 if step is not None:
1161 if i - prev == step:
1162 continue
1163 yield _genslice(start, prev, step)
1164 step = None
1165 continue
1166 if i - prev in [-1, 1]:
1167 step = i - prev
1168 start = prev
1169 continue
1170 else:
1171 yield 's[%d]' % prev
1172 if step is None:
1173 yield 's[%d]' % i
1174 else:
1175 yield _genslice(start, i, step)
1176
1177 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1178 cache_res = func(test_string)
1179 cache_spec = [ord(c) for c in cache_res]
1180 expr_code = ' + '.join(gen_sig_code(cache_spec))
1181 signature_id_tuple = '(%s)' % (
1182 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1183 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1184 ' return %s\n') % (signature_id_tuple, expr_code)
1185 self.to_screen('Extracted signature function:\n' + code)
1186
1187 def _parse_sig_js(self, jscode):
1188 funcname = self._search_regex(
1189 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1190 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1191 r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1192 r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1193 jscode, 'Initial JS player signature function name', group='sig')
1194
1195 jsi = JSInterpreter(jscode)
1196 initial_function = jsi.extract_function(funcname)
1197 return lambda s: initial_function([s])
1198
1199 def _parse_sig_swf(self, file_contents):
1200 swfi = SWFInterpreter(file_contents)
1201 TARGET_CLASSNAME = 'SignatureDecipher'
1202 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1203 initial_function = swfi.extract_function(searched_class, 'decipher')
1204 return lambda s: initial_function([s])
1205
1206 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1207 """Turn the encrypted s field into a working signature"""
1208
1209 if player_url is None:
1210 raise ExtractorError('Cannot decrypt signature without player_url')
1211
1212 if player_url.startswith('//'):
1213 player_url = 'https:' + player_url
1214 elif not re.match(r'https?://', player_url):
1215 player_url = compat_urlparse.urljoin(
1216 'https://www.youtube.com', player_url)
1217 try:
1218 player_id = (player_url, self._signature_cache_id(s))
1219 if player_id not in self._player_cache:
1220 func = self._extract_signature_function(
1221 video_id, player_url, s
1222 )
1223 self._player_cache[player_id] = func
1224 func = self._player_cache[player_id]
1225 if self._downloader.params.get('youtube_print_sig_code'):
1226 self._print_sig_code(func, s)
1227 return func(s)
1228 except Exception as e:
1229 tb = traceback.format_exc()
1230 raise ExtractorError(
1231 'Signature extraction failed: ' + tb, cause=e)
1232
1233 def _get_subtitles(self, video_id, webpage):
1234 try:
1235 subs_doc = self._download_xml(
1236 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1237 video_id, note=False)
1238 except ExtractorError as err:
1239 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1240 return {}
1241
1242 sub_lang_list = {}
1243 for track in subs_doc.findall('track'):
1244 lang = track.attrib['lang_code']
1245 if lang in sub_lang_list:
1246 continue
1247 sub_formats = []
1248 for ext in self._SUBTITLE_FORMATS:
1249 params = compat_urllib_parse_urlencode({
1250 'lang': lang,
1251 'v': video_id,
1252 'fmt': ext,
1253 'name': track.attrib['name'].encode('utf-8'),
1254 })
1255 sub_formats.append({
1256 'url': 'https://www.youtube.com/api/timedtext?' + params,
1257 'ext': ext,
1258 })
1259 sub_lang_list[lang] = sub_formats
1260 if not sub_lang_list:
1261 self._downloader.report_warning('video doesn\'t have subtitles')
1262 return {}
1263 return sub_lang_list
1264
1265 def _get_ytplayer_config(self, video_id, webpage):
1266 patterns = (
1267 # User data may contain arbitrary character sequences that may affect
1268 # JSON extraction with regex, e.g. when '};' is contained the second
1269 # regex won't capture the whole JSON. Yet working around by trying more
1270 # concrete regex first keeping in mind proper quoted string handling
1271 # to be implemented in future that will replace this workaround (see
1272 # https://github.com/rg3/youtube-dl/issues/7468,
1273 # https://github.com/rg3/youtube-dl/pull/7599)
1274 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1275 r';ytplayer\.config\s*=\s*({.+?});',
1276 )
1277 config = self._search_regex(
1278 patterns, webpage, 'ytplayer.config', default=None)
1279 if config:
1280 return self._parse_json(
1281 uppercase_escape(config), video_id, fatal=False)
1282
1283 def _get_automatic_captions(self, video_id, webpage):
1284 """We need the webpage for getting the captions url, pass it as an
1285 argument to speed up the process."""
1286 self.to_screen('%s: Looking for automatic captions' % video_id)
1287 player_config = self._get_ytplayer_config(video_id, webpage)
1288 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1289 if not player_config:
1290 self._downloader.report_warning(err_msg)
1291 return {}
1292 try:
1293 args = player_config['args']
1294 caption_url = args.get('ttsurl')
1295 if caption_url:
1296 timestamp = args['timestamp']
1297 # We get the available subtitles
1298 list_params = compat_urllib_parse_urlencode({
1299 'type': 'list',
1300 'tlangs': 1,
1301 'asrs': 1,
1302 })
1303 list_url = caption_url + '&' + list_params
1304 caption_list = self._download_xml(list_url, video_id)
1305 original_lang_node = caption_list.find('track')
1306 if original_lang_node is None:
1307 self._downloader.report_warning('Video doesn\'t have automatic captions')
1308 return {}
1309 original_lang = original_lang_node.attrib['lang_code']
1310 caption_kind = original_lang_node.attrib.get('kind', '')
1311
1312 sub_lang_list = {}
1313 for lang_node in caption_list.findall('target'):
1314 sub_lang = lang_node.attrib['lang_code']
1315 sub_formats = []
1316 for ext in self._SUBTITLE_FORMATS:
1317 params = compat_urllib_parse_urlencode({
1318 'lang': original_lang,
1319 'tlang': sub_lang,
1320 'fmt': ext,
1321 'ts': timestamp,
1322 'kind': caption_kind,
1323 })
1324 sub_formats.append({
1325 'url': caption_url + '&' + params,
1326 'ext': ext,
1327 })
1328 sub_lang_list[sub_lang] = sub_formats
1329 return sub_lang_list
1330
1331 def make_captions(sub_url, sub_langs):
1332 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1333 caption_qs = compat_parse_qs(parsed_sub_url.query)
1334 captions = {}
1335 for sub_lang in sub_langs:
1336 sub_formats = []
1337 for ext in self._SUBTITLE_FORMATS:
1338 caption_qs.update({
1339 'tlang': [sub_lang],
1340 'fmt': [ext],
1341 })
1342 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1343 query=compat_urllib_parse_urlencode(caption_qs, True)))
1344 sub_formats.append({
1345 'url': sub_url,
1346 'ext': ext,
1347 })
1348 captions[sub_lang] = sub_formats
1349 return captions
1350
1351 # New captions format as of 22.06.2017
1352 player_response = args.get('player_response')
1353 if player_response and isinstance(player_response, compat_str):
1354 player_response = self._parse_json(
1355 player_response, video_id, fatal=False)
1356 if player_response:
1357 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1358 base_url = renderer['captionTracks'][0]['baseUrl']
1359 sub_lang_list = []
1360 for lang in renderer['translationLanguages']:
1361 lang_code = lang.get('languageCode')
1362 if lang_code:
1363 sub_lang_list.append(lang_code)
1364 return make_captions(base_url, sub_lang_list)
1365
1366 # Some videos don't provide ttsurl but rather caption_tracks and
1367 # caption_translation_languages (e.g. 20LmZk1hakA)
1368 # Does not used anymore as of 22.06.2017
1369 caption_tracks = args['caption_tracks']
1370 caption_translation_languages = args['caption_translation_languages']
1371 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1372 sub_lang_list = []
1373 for lang in caption_translation_languages.split(','):
1374 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1375 sub_lang = lang_qs.get('lc', [None])[0]
1376 if sub_lang:
1377 sub_lang_list.append(sub_lang)
1378 return make_captions(caption_url, sub_lang_list)
1379 # An extractor error can be raise by the download process if there are
1380 # no automatic captions but there are subtitles
1381 except (KeyError, IndexError, ExtractorError):
1382 self._downloader.report_warning(err_msg)
1383 return {}
1384
1385 def _mark_watched(self, video_id, video_info):
1386 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1387 if not playback_url:
1388 return
1389 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1390 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1391
1392 # cpn generation algorithm is reverse engineered from base.js.
1393 # In fact it works even with dummy cpn.
1394 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1395 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1396
1397 qs.update({
1398 'ver': ['2'],
1399 'cpn': [cpn],
1400 })
1401 playback_url = compat_urlparse.urlunparse(
1402 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1403
1404 self._download_webpage(
1405 playback_url, video_id, 'Marking watched',
1406 'Unable to mark watched', fatal=False)
1407
1408 @staticmethod
1409 def _extract_urls(webpage):
1410 # Embedded YouTube player
1411 entries = [
1412 unescapeHTML(mobj.group('url'))
1413 for mobj in re.finditer(r'''(?x)
1414 (?:
1415 <iframe[^>]+?src=|
1416 data-video-url=|
1417 <embed[^>]+?src=|
1418 embedSWF\(?:\s*|
1419 <object[^>]+data=|
1420 new\s+SWFObject\(
1421 )
1422 (["\'])
1423 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1424 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1425 \1''', webpage)]
1426
1427 # lazyYT YouTube embed
1428 entries.extend(list(map(
1429 unescapeHTML,
1430 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1431
1432 # Wordpress "YouTube Video Importer" plugin
1433 matches = re.findall(r'''(?x)<div[^>]+
1434 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1435 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1436 entries.extend(m[-1] for m in matches)
1437
1438 return entries
1439
1440 @staticmethod
1441 def _extract_url(webpage):
1442 urls = YoutubeIE._extract_urls(webpage)
1443 return urls[0] if urls else None
1444
1445 @classmethod
1446 def extract_id(cls, url):
1447 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1448 if mobj is None:
1449 raise ExtractorError('Invalid URL: %s' % url)
1450 video_id = mobj.group(2)
1451 return video_id
1452
1453 def _extract_annotations(self, video_id):
1454 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1455 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1456
1457 @staticmethod
1458 def _extract_chapters(description, duration):
1459 if not description:
1460 return None
1461 chapter_lines = re.findall(
1462 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1463 description)
1464 if not chapter_lines:
1465 return None
1466 chapters = []
1467 for next_num, (chapter_line, time_point) in enumerate(
1468 chapter_lines, start=1):
1469 start_time = parse_duration(time_point)
1470 if start_time is None:
1471 continue
1472 if start_time > duration:
1473 break
1474 end_time = (duration if next_num == len(chapter_lines)
1475 else parse_duration(chapter_lines[next_num][1]))
1476 if end_time is None:
1477 continue
1478 if end_time > duration:
1479 end_time = duration
1480 if start_time > end_time:
1481 break
1482 chapter_title = re.sub(
1483 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1484 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1485 chapters.append({
1486 'start_time': start_time,
1487 'end_time': end_time,
1488 'title': chapter_title,
1489 })
1490 return chapters
1491
1492 def _real_extract(self, url):
1493 url, smuggled_data = unsmuggle_url(url, {})
1494
1495 proto = (
1496 'http' if self._downloader.params.get('prefer_insecure', False)
1497 else 'https')
1498
1499 start_time = None
1500 end_time = None
1501 parsed_url = compat_urllib_parse_urlparse(url)
1502 for component in [parsed_url.fragment, parsed_url.query]:
1503 query = compat_parse_qs(component)
1504 if start_time is None and 't' in query:
1505 start_time = parse_duration(query['t'][0])
1506 if start_time is None and 'start' in query:
1507 start_time = parse_duration(query['start'][0])
1508 if end_time is None and 'end' in query:
1509 end_time = parse_duration(query['end'][0])
1510
1511 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1512 mobj = re.search(self._NEXT_URL_RE, url)
1513 if mobj:
1514 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1515 video_id = self.extract_id(url)
1516
1517 # Get video webpage
1518 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1519 video_webpage = self._download_webpage(url, video_id)
1520
1521 # Attempt to extract SWF player URL
1522 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1523 if mobj is not None:
1524 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1525 else:
1526 player_url = None
1527
1528 dash_mpds = []
1529
1530 def add_dash_mpd(video_info):
1531 dash_mpd = video_info.get('dashmpd')
1532 if dash_mpd and dash_mpd[0] not in dash_mpds:
1533 dash_mpds.append(dash_mpd[0])
1534
1535 is_live = None
1536 view_count = None
1537
1538 def extract_view_count(v_info):
1539 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1540
1541 # Get video info
1542 embed_webpage = None
1543 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1544 age_gate = True
1545 # We simulate the access to the video from www.youtube.com/v/{video_id}
1546 # this can be viewed without login into Youtube
1547 url = proto + '://www.youtube.com/embed/%s' % video_id
1548 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1549 data = compat_urllib_parse_urlencode({
1550 'video_id': video_id,
1551 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1552 'sts': self._search_regex(
1553 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1554 })
1555 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1556 video_info_webpage = self._download_webpage(
1557 video_info_url, video_id,
1558 note='Refetching age-gated info webpage',
1559 errnote='unable to download video info webpage')
1560 video_info = compat_parse_qs(video_info_webpage)
1561 add_dash_mpd(video_info)
1562 else:
1563 age_gate = False
1564 video_info = None
1565 sts = None
1566 # Try looking directly into the video webpage
1567 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1568 if ytplayer_config:
1569 args = ytplayer_config['args']
1570 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1571 # Convert to the same format returned by compat_parse_qs
1572 video_info = dict((k, [v]) for k, v in args.items())
1573 add_dash_mpd(video_info)
1574 # Rental video is not rented but preview is available (e.g.
1575 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1576 # https://github.com/rg3/youtube-dl/issues/10532)
1577 if not video_info and args.get('ypc_vid'):
1578 return self.url_result(
1579 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1580 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1581 is_live = True
1582 sts = ytplayer_config.get('sts')
1583 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1584 # We also try looking in get_video_info since it may contain different dashmpd
1585 # URL that points to a DASH manifest with possibly different itag set (some itags
1586 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1587 # manifest pointed by get_video_info's dashmpd).
1588 # The general idea is to take a union of itags of both DASH manifests (for example
1589 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1590 self.report_video_info_webpage_download(video_id)
1591 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1592 query = {
1593 'video_id': video_id,
1594 'ps': 'default',
1595 'eurl': '',
1596 'gl': 'US',
1597 'hl': 'en',
1598 }
1599 if el:
1600 query['el'] = el
1601 if sts:
1602 query['sts'] = sts
1603 video_info_webpage = self._download_webpage(
1604 '%s://www.youtube.com/get_video_info' % proto,
1605 video_id, note=False,
1606 errnote='unable to download video info webpage',
1607 fatal=False, query=query)
1608 if not video_info_webpage:
1609 continue
1610 get_video_info = compat_parse_qs(video_info_webpage)
1611 add_dash_mpd(get_video_info)
1612 if view_count is None:
1613 view_count = extract_view_count(get_video_info)
1614 if not video_info:
1615 video_info = get_video_info
1616 if 'token' in get_video_info:
1617 # Different get_video_info requests may report different results, e.g.
1618 # some may report video unavailability, but some may serve it without
1619 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1620 # the original webpage as well as el=info and el=embedded get_video_info
1621 # requests report video unavailability due to geo restriction while
1622 # el=detailpage succeeds and returns valid data). This is probably
1623 # due to YouTube measures against IP ranges of hosting providers.
1624 # Working around by preferring the first succeeded video_info containing
1625 # the token if no such video_info yet was found.
1626 if 'token' not in video_info:
1627 video_info = get_video_info
1628 break
1629
1630 def extract_unavailable_message():
1631 return self._html_search_regex(
1632 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1633 video_webpage, 'unavailable message', default=None)
1634
1635 if 'token' not in video_info:
1636 if 'reason' in video_info:
1637 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1638 regions_allowed = self._html_search_meta(
1639 'regionsAllowed', video_webpage, default=None)
1640 countries = regions_allowed.split(',') if regions_allowed else None
1641 self.raise_geo_restricted(
1642 msg=video_info['reason'][0], countries=countries)
1643 reason = video_info['reason'][0]
1644 if 'Invalid parameters' in reason:
1645 unavailable_message = extract_unavailable_message()
1646 if unavailable_message:
1647 reason = unavailable_message
1648 raise ExtractorError(
1649 'YouTube said: %s' % reason,
1650 expected=True, video_id=video_id)
1651 else:
1652 raise ExtractorError(
1653 '"token" parameter not in video info for unknown reason',
1654 video_id=video_id)
1655
1656 # title
1657 if 'title' in video_info:
1658 video_title = video_info['title'][0]
1659 else:
1660 self._downloader.report_warning('Unable to extract video title')
1661 video_title = '_'
1662
1663 # description
1664 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1665 if video_description:
1666
1667 def replace_url(m):
1668 redir_url = compat_urlparse.urljoin(url, m.group(1))
1669 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1670 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1671 qs = compat_parse_qs(parsed_redir_url.query)
1672 q = qs.get('q')
1673 if q and q[0]:
1674 return q[0]
1675 return redir_url
1676
1677 description_original = video_description = re.sub(r'''(?x)
1678 <a\s+
1679 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1680 (?:title|href)="([^"]+)"\s+
1681 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1682 class="[^"]*"[^>]*>
1683 [^<]+\.{3}\s*
1684 </a>
1685 ''', replace_url, video_description)
1686 video_description = clean_html(video_description)
1687 else:
1688 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1689 if fd_mobj:
1690 video_description = unescapeHTML(fd_mobj.group(1))
1691 else:
1692 video_description = ''
1693
1694 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1695 if not self._downloader.params.get('noplaylist'):
1696 entries = []
1697 feed_ids = []
1698 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1699 for feed in multifeed_metadata_list.split(','):
1700 # Unquote should take place before split on comma (,) since textual
1701 # fields may contain comma as well (see
1702 # https://github.com/rg3/youtube-dl/issues/8536)
1703 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1704 entries.append({
1705 '_type': 'url_transparent',
1706 'ie_key': 'Youtube',
1707 'url': smuggle_url(
1708 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1709 {'force_singlefeed': True}),
1710 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1711 })
1712 feed_ids.append(feed_data['id'][0])
1713 self.to_screen(
1714 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1715 % (', '.join(feed_ids), video_id))
1716 return self.playlist_result(entries, video_id, video_title, video_description)
1717 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1718
1719 if view_count is None:
1720 view_count = extract_view_count(video_info)
1721
1722 # Check for "rental" videos
1723 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1724 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1725
1726 def _extract_filesize(media_url):
1727 return int_or_none(self._search_regex(
1728 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1729
1730 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1731 self.report_rtmp_download()
1732 formats = [{
1733 'format_id': '_rtmp',
1734 'protocol': 'rtmp',
1735 'url': video_info['conn'][0],
1736 'player_url': player_url,
1737 }]
1738 elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1739 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1740 if 'rtmpe%3Dyes' in encoded_url_map:
1741 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1742 formats_spec = {}
1743 fmt_list = video_info.get('fmt_list', [''])[0]
1744 if fmt_list:
1745 for fmt in fmt_list.split(','):
1746 spec = fmt.split('/')
1747 if len(spec) > 1:
1748 width_height = spec[1].split('x')
1749 if len(width_height) == 2:
1750 formats_spec[spec[0]] = {
1751 'resolution': spec[1],
1752 'width': int_or_none(width_height[0]),
1753 'height': int_or_none(width_height[1]),
1754 }
1755 q = qualities(['small', 'medium', 'hd720'])
1756 formats = []
1757 for url_data_str in encoded_url_map.split(','):
1758 url_data = compat_parse_qs(url_data_str)
1759 if 'itag' not in url_data or 'url' not in url_data:
1760 continue
1761 format_id = url_data['itag'][0]
1762 url = url_data['url'][0]
1763
1764 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1765 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1766 jsplayer_url_json = self._search_regex(
1767 ASSETS_RE,
1768 embed_webpage if age_gate else video_webpage,
1769 'JS player URL (1)', default=None)
1770 if not jsplayer_url_json and not age_gate:
1771 # We need the embed website after all
1772 if embed_webpage is None:
1773 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1774 embed_webpage = self._download_webpage(
1775 embed_url, video_id, 'Downloading embed webpage')
1776 jsplayer_url_json = self._search_regex(
1777 ASSETS_RE, embed_webpage, 'JS player URL')
1778
1779 player_url = json.loads(jsplayer_url_json)
1780 if player_url is None:
1781 player_url_json = self._search_regex(
1782 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1783 video_webpage, 'age gate player URL')
1784 player_url = json.loads(player_url_json)
1785
1786 if 'sig' in url_data:
1787 url += '&signature=' + url_data['sig'][0]
1788 elif 's' in url_data:
1789 encrypted_sig = url_data['s'][0]
1790
1791 if self._downloader.params.get('verbose'):
1792 if player_url is None:
1793 player_version = 'unknown'
1794 player_desc = 'unknown'
1795 else:
1796 if player_url.endswith('swf'):
1797 player_version = self._search_regex(
1798 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1799 'flash player', fatal=False)
1800 player_desc = 'flash player %s' % player_version
1801 else:
1802 player_version = self._search_regex(
1803 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1804 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1805 player_url,
1806 'html5 player', fatal=False)
1807 player_desc = 'html5 player %s' % player_version
1808
1809 parts_sizes = self._signature_cache_id(encrypted_sig)
1810 self.to_screen('{%s} signature length %s, %s' %
1811 (format_id, parts_sizes, player_desc))
1812
1813 signature = self._decrypt_signature(
1814 encrypted_sig, video_id, player_url, age_gate)
1815 url += '&signature=' + signature
1816 if 'ratebypass' not in url:
1817 url += '&ratebypass=yes'
1818
1819 dct = {
1820 'format_id': format_id,
1821 'url': url,
1822 'player_url': player_url,
1823 }
1824 if format_id in self._formats:
1825 dct.update(self._formats[format_id])
1826 if format_id in formats_spec:
1827 dct.update(formats_spec[format_id])
1828
1829 # Some itags are not included in DASH manifest thus corresponding formats will
1830 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1831 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1832 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1833 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1834
1835 filesize = int_or_none(url_data.get(
1836 'clen', [None])[0]) or _extract_filesize(url)
1837
1838 quality = url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0]
1839
1840 more_fields = {
1841 'filesize': filesize,
1842 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1843 'width': width,
1844 'height': height,
1845 'fps': int_or_none(url_data.get('fps', [None])[0]),
1846 'format_note': quality,
1847 'quality': q(quality),
1848 }
1849 for key, value in more_fields.items():
1850 if value:
1851 dct[key] = value
1852 type_ = url_data.get('type', [None])[0]
1853 if type_:
1854 type_split = type_.split(';')
1855 kind_ext = type_split[0].split('/')
1856 if len(kind_ext) == 2:
1857 kind, _ = kind_ext
1858 dct['ext'] = mimetype2ext(type_split[0])
1859 if kind in ('audio', 'video'):
1860 codecs = None
1861 for mobj in re.finditer(
1862 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1863 if mobj.group('key') == 'codecs':
1864 codecs = mobj.group('val')
1865 break
1866 if codecs:
1867 dct.update(parse_codecs(codecs))
1868 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1869 dct['downloader_options'] = {
1870 # Youtube throttles chunks >~10M
1871 'http_chunk_size': 10485760,
1872 }
1873 formats.append(dct)
1874 elif video_info.get('hlsvp'):
1875 manifest_url = video_info['hlsvp'][0]
1876 formats = []
1877 m3u8_formats = self._extract_m3u8_formats(
1878 manifest_url, video_id, 'mp4', fatal=False)
1879 for a_format in m3u8_formats:
1880 itag = self._search_regex(
1881 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1882 if itag:
1883 a_format['format_id'] = itag
1884 if itag in self._formats:
1885 dct = self._formats[itag].copy()
1886 dct.update(a_format)
1887 a_format = dct
1888 a_format['player_url'] = player_url
1889 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1890 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1891 formats.append(a_format)
1892 else:
1893 error_message = clean_html(video_info.get('reason', [None])[0])
1894 if not error_message:
1895 error_message = extract_unavailable_message()
1896 if error_message:
1897 raise ExtractorError(error_message, expected=True)
1898 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1899
1900 # uploader
1901 video_uploader = try_get(video_info, lambda x: x['author'][0], compat_str)
1902 if video_uploader:
1903 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1904 else:
1905 self._downloader.report_warning('unable to extract uploader name')
1906
1907 # uploader_id
1908 video_uploader_id = None
1909 video_uploader_url = None
1910 mobj = re.search(
1911 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1912 video_webpage)
1913 if mobj is not None:
1914 video_uploader_id = mobj.group('uploader_id')
1915 video_uploader_url = mobj.group('uploader_url')
1916 else:
1917 self._downloader.report_warning('unable to extract uploader nickname')
1918
1919 channel_id = self._html_search_meta(
1920 'channelId', video_webpage, 'channel id')
1921 channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
1922
1923 # thumbnail image
1924 # We try first to get a high quality image:
1925 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1926 video_webpage, re.DOTALL)
1927 if m_thumb is not None:
1928 video_thumbnail = m_thumb.group(1)
1929 elif 'thumbnail_url' not in video_info:
1930 self._downloader.report_warning('unable to extract video thumbnail')
1931 video_thumbnail = None
1932 else: # don't panic if we can't find it
1933 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1934
1935 # upload date
1936 upload_date = self._html_search_meta(
1937 'datePublished', video_webpage, 'upload date', default=None)
1938 if not upload_date:
1939 upload_date = self._search_regex(
1940 [r'(?s)id="eow-date.*?>(.*?)</span>',
1941 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1942 video_webpage, 'upload date', default=None)
1943 upload_date = unified_strdate(upload_date)
1944
1945 video_license = self._html_search_regex(
1946 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1947 video_webpage, 'license', default=None)
1948
1949 m_music = re.search(
1950 r'''(?x)
1951 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1952 <ul[^>]*>\s*
1953 <li>(?P<title>.+?)
1954 by (?P<creator>.+?)
1955 (?:
1956 \(.+?\)|
1957 <a[^>]*
1958 (?:
1959 \bhref=["\']/red[^>]*>| # drop possible
1960 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1961 )
1962 .*?
1963 )?</li
1964 ''',
1965 video_webpage)
1966 if m_music:
1967 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1968 video_creator = clean_html(m_music.group('creator'))
1969 else:
1970 video_alt_title = video_creator = None
1971
1972 def extract_meta(field):
1973 return self._html_search_regex(
1974 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
1975 video_webpage, field, default=None)
1976
1977 track = extract_meta('Song')
1978 artist = extract_meta('Artist')
1979
1980 m_episode = re.search(
1981 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1982 video_webpage)
1983 if m_episode:
1984 series = m_episode.group('series')
1985 season_number = int(m_episode.group('season'))
1986 episode_number = int(m_episode.group('episode'))
1987 else:
1988 series = season_number = episode_number = None
1989
1990 m_cat_container = self._search_regex(
1991 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1992 video_webpage, 'categories', default=None)
1993 if m_cat_container:
1994 category = self._html_search_regex(
1995 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1996 default=None)
1997 video_categories = None if category is None else [category]
1998 else:
1999 video_categories = None
2000
2001 video_tags = [
2002 unescapeHTML(m.group('content'))
2003 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2004
2005 def _extract_count(count_name):
2006 return str_to_int(self._search_regex(
2007 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2008 % re.escape(count_name),
2009 video_webpage, count_name, default=None))
2010
2011 like_count = _extract_count('like')
2012 dislike_count = _extract_count('dislike')
2013
2014 # subtitles
2015 video_subtitles = self.extract_subtitles(video_id, video_webpage)
2016 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2017
2018 video_duration = try_get(
2019 video_info, lambda x: int_or_none(x['length_seconds'][0]))
2020 if not video_duration:
2021 video_duration = parse_duration(self._html_search_meta(
2022 'duration', video_webpage, 'video duration'))
2023
2024 # annotations
2025 video_annotations = None
2026 if self._downloader.params.get('writeannotations', False):
2027 video_annotations = self._extract_annotations(video_id)
2028
2029 chapters = self._extract_chapters(description_original, video_duration)
2030
2031 # Look for the DASH manifest
2032 if self._downloader.params.get('youtube_include_dash_manifest', True):
2033 dash_mpd_fatal = True
2034 for mpd_url in dash_mpds:
2035 dash_formats = {}
2036 try:
2037 def decrypt_sig(mobj):
2038 s = mobj.group(1)
2039 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2040 return '/signature/%s' % dec_s
2041
2042 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2043
2044 for df in self._extract_mpd_formats(
2045 mpd_url, video_id, fatal=dash_mpd_fatal,
2046 formats_dict=self._formats):
2047 if not df.get('filesize'):
2048 df['filesize'] = _extract_filesize(df['url'])
2049 # Do not overwrite DASH format found in some previous DASH manifest
2050 if df['format_id'] not in dash_formats:
2051 dash_formats[df['format_id']] = df
2052 # Additional DASH manifests may end up in HTTP Error 403 therefore
2053 # allow them to fail without bug report message if we already have
2054 # some DASH manifest succeeded. This is temporary workaround to reduce
2055 # burst of bug reports until we figure out the reason and whether it
2056 # can be fixed at all.
2057 dash_mpd_fatal = False
2058 except (ExtractorError, KeyError) as e:
2059 self.report_warning(
2060 'Skipping DASH manifest: %r' % e, video_id)
2061 if dash_formats:
2062 # Remove the formats we found through non-DASH, they
2063 # contain less info and it can be wrong, because we use
2064 # fixed values (for example the resolution). See
2065 # https://github.com/rg3/youtube-dl/issues/5774 for an
2066 # example.
2067 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2068 formats.extend(dash_formats.values())
2069
2070 # Check for malformed aspect ratio
2071 stretched_m = re.search(
2072 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2073 video_webpage)
2074 if stretched_m:
2075 w = float(stretched_m.group('w'))
2076 h = float(stretched_m.group('h'))
2077 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2078 # We will only process correct ratios.
2079 if w > 0 and h > 0:
2080 ratio = w / h
2081 for f in formats:
2082 if f.get('vcodec') != 'none':
2083 f['stretched_ratio'] = ratio
2084
2085 self._sort_formats(formats)
2086
2087 self.mark_watched(video_id, video_info)
2088
2089 return {
2090 'id': video_id,
2091 'uploader': video_uploader,
2092 'uploader_id': video_uploader_id,
2093 'uploader_url': video_uploader_url,
2094 'channel_id': channel_id,
2095 'channel_url': channel_url,
2096 'upload_date': upload_date,
2097 'license': video_license,
2098 'creator': video_creator or artist,
2099 'title': video_title,
2100 'alt_title': video_alt_title or track,
2101 'thumbnail': video_thumbnail,
2102 'description': video_description,
2103 'categories': video_categories,
2104 'tags': video_tags,
2105 'subtitles': video_subtitles,
2106 'automatic_captions': automatic_captions,
2107 'duration': video_duration,
2108 'age_limit': 18 if age_gate else 0,
2109 'annotations': video_annotations,
2110 'chapters': chapters,
2111 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2112 'view_count': view_count,
2113 'like_count': like_count,
2114 'dislike_count': dislike_count,
2115 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2116 'formats': formats,
2117 'is_live': is_live,
2118 'start_time': start_time,
2119 'end_time': end_time,
2120 'series': series,
2121 'season_number': season_number,
2122 'episode_number': episode_number,
2123 'track': track,
2124 'artist': artist,
2125 }
2126
2127
2128 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2129 IE_DESC = 'YouTube.com playlists'
2130 _VALID_URL = r"""(?x)(?:
2131 (?:https?://)?
2132 (?:\w+\.)?
2133 (?:
2134 youtube\.com/
2135 (?:
2136 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2137 \? (?:.*?[&;])*? (?:p|a|list)=
2138 | p/
2139 )|
2140 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2141 )
2142 (
2143 (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2144 # Top tracks, they can also include dots
2145 |(?:MC)[\w\.]*
2146 )
2147 .*
2148 |
2149 (%(playlist_id)s)
2150 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2151 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2152 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2153 IE_NAME = 'youtube:playlist'
2154 _TESTS = [{
2155 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2156 'info_dict': {
2157 'title': 'ytdl test PL',
2158 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2159 },
2160 'playlist_count': 3,
2161 }, {
2162 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2163 'info_dict': {
2164 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2165 'title': 'YDL_Empty_List',
2166 },
2167 'playlist_count': 0,
2168 'skip': 'This playlist is private',
2169 }, {
2170 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2171 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2172 'info_dict': {
2173 'title': '29C3: Not my department',
2174 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2175 },
2176 'playlist_count': 95,
2177 }, {
2178 'note': 'issue #673',
2179 'url': 'PLBB231211A4F62143',
2180 'info_dict': {
2181 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2182 'id': 'PLBB231211A4F62143',
2183 },
2184 'playlist_mincount': 26,
2185 }, {
2186 'note': 'Large playlist',
2187 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2188 'info_dict': {
2189 'title': 'Uploads from Cauchemar',
2190 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2191 },
2192 'playlist_mincount': 799,
2193 }, {
2194 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2195 'info_dict': {
2196 'title': 'YDL_safe_search',
2197 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2198 },
2199 'playlist_count': 2,
2200 'skip': 'This playlist is private',
2201 }, {
2202 'note': 'embedded',
2203 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2204 'playlist_count': 4,
2205 'info_dict': {
2206 'title': 'JODA15',
2207 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2208 }
2209 }, {
2210 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2211 'playlist_mincount': 485,
2212 'info_dict': {
2213 'title': '2017 華語最新單曲 (2/24更新)',
2214 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2215 }
2216 }, {
2217 'note': 'Embedded SWF player',
2218 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2219 'playlist_count': 4,
2220 'info_dict': {
2221 'title': 'JODA7',
2222 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2223 }
2224 }, {
2225 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2226 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2227 'info_dict': {
2228 'title': 'Uploads from Interstellar Movie',
2229 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2230 },
2231 'playlist_mincount': 21,
2232 }, {
2233 # Playlist URL that does not actually serve a playlist
2234 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2235 'info_dict': {
2236 'id': 'FqZTN594JQw',
2237 'ext': 'webm',
2238 'title': "Smiley's People 01 detective, Adventure Series, Action",
2239 'uploader': 'STREEM',
2240 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2241 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2242 'upload_date': '20150526',
2243 'license': 'Standard YouTube License',
2244 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2245 'categories': ['People & Blogs'],
2246 'tags': list,
2247 'like_count': int,
2248 'dislike_count': int,
2249 },
2250 'params': {
2251 'skip_download': True,
2252 },
2253 'add_ie': [YoutubeIE.ie_key()],
2254 }, {
2255 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2256 'info_dict': {
2257 'id': 'yeWKywCrFtk',
2258 'ext': 'mp4',
2259 'title': 'Small Scale Baler and Braiding Rugs',
2260 'uploader': 'Backus-Page House Museum',
2261 'uploader_id': 'backuspagemuseum',
2262 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2263 'upload_date': '20161008',
2264 'license': 'Standard YouTube License',
2265 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2266 'categories': ['Nonprofits & Activism'],
2267 'tags': list,
2268 'like_count': int,
2269 'dislike_count': int,
2270 },
2271 'params': {
2272 'noplaylist': True,
2273 'skip_download': True,
2274 },
2275 }, {
2276 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2277 'only_matching': True,
2278 }, {
2279 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2280 'only_matching': True,
2281 }, {
2282 # music album playlist
2283 'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2284 'only_matching': True,
2285 }]
2286
2287 def _real_initialize(self):
2288 self._login()
2289
2290 def _extract_mix(self, playlist_id):
2291 # The mixes are generated from a single video
2292 # the id of the playlist is just 'RD' + video_id
2293 ids = []
2294 last_id = playlist_id[-11:]
2295 for n in itertools.count(1):
2296 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2297 webpage = self._download_webpage(
2298 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2299 new_ids = orderedSet(re.findall(
2300 r'''(?xs)data-video-username=".*?".*?
2301 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2302 webpage))
2303 # Fetch new pages until all the videos are repeated, it seems that
2304 # there are always 51 unique videos.
2305 new_ids = [_id for _id in new_ids if _id not in ids]
2306 if not new_ids:
2307 break
2308 ids.extend(new_ids)
2309 last_id = ids[-1]
2310
2311 url_results = self._ids_to_results(ids)
2312
2313 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2314 title_span = (
2315 search_title('playlist-title') or
2316 search_title('title long-title') or
2317 search_title('title'))
2318 title = clean_html(title_span)
2319
2320 return self.playlist_result(url_results, playlist_id, title)
2321
2322 def _extract_playlist(self, playlist_id):
2323 url = self._TEMPLATE_URL % playlist_id
2324 page = self._download_webpage(url, playlist_id)
2325
2326 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2327 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2328 match = match.strip()
2329 # Check if the playlist exists or is private
2330 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2331 if mobj:
2332 reason = mobj.group('reason')
2333 message = 'This playlist %s' % reason
2334 if 'private' in reason:
2335 message += ', use --username or --netrc to access it'
2336 message += '.'
2337 raise ExtractorError(message, expected=True)
2338 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2339 raise ExtractorError(
2340 'Invalid parameters. Maybe URL is incorrect.',
2341 expected=True)
2342 elif re.match(r'[^<]*Choose your language[^<]*', match):
2343 continue
2344 else:
2345 self.report_warning('Youtube gives an alert message: ' + match)
2346
2347 playlist_title = self._html_search_regex(
2348 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2349 page, 'title', default=None)
2350
2351 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2352 uploader = self._search_regex(
2353 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2354 page, 'uploader', default=None)
2355 mobj = re.search(
2356 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2357 page)
2358 if mobj:
2359 uploader_id = mobj.group('uploader_id')
2360 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2361 else:
2362 uploader_id = uploader_url = None
2363
2364 has_videos = True
2365
2366 if not playlist_title:
2367 try:
2368 # Some playlist URLs don't actually serve a playlist (e.g.
2369 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2370 next(self._entries(page, playlist_id))
2371 except StopIteration:
2372 has_videos = False
2373
2374 playlist = self.playlist_result(
2375 self._entries(page, playlist_id), playlist_id, playlist_title)
2376 playlist.update({
2377 'uploader': uploader,
2378 'uploader_id': uploader_id,
2379 'uploader_url': uploader_url,
2380 })
2381
2382 return has_videos, playlist
2383
2384 def _check_download_just_video(self, url, playlist_id):
2385 # Check if it's a video-specific URL
2386 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2387 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2388 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2389 'video id', default=None)
2390 if video_id:
2391 if self._downloader.params.get('noplaylist'):
2392 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2393 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2394 else:
2395 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2396 return video_id, None
2397 return None, None
2398
2399 def _real_extract(self, url):
2400 # Extract playlist id
2401 mobj = re.match(self._VALID_URL, url)
2402 if mobj is None:
2403 raise ExtractorError('Invalid URL: %s' % url)
2404 playlist_id = mobj.group(1) or mobj.group(2)
2405
2406 video_id, video = self._check_download_just_video(url, playlist_id)
2407 if video:
2408 return video
2409
2410 if playlist_id.startswith(('RD', 'UL', 'PU')):
2411 # Mixes require a custom extraction process
2412 return self._extract_mix(playlist_id)
2413
2414 has_videos, playlist = self._extract_playlist(playlist_id)
2415 if has_videos or not video_id:
2416 return playlist
2417
2418 # Some playlist URLs don't actually serve a playlist (see
2419 # https://github.com/rg3/youtube-dl/issues/10537).
2420 # Fallback to plain video extraction if there is a video id
2421 # along with playlist id.
2422 return self.url_result(video_id, 'Youtube', video_id=video_id)
2423
2424
2425 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2426 IE_DESC = 'YouTube.com channels'
2427 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2428 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2429 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2430 IE_NAME = 'youtube:channel'
2431 _TESTS = [{
2432 'note': 'paginated channel',
2433 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2434 'playlist_mincount': 91,
2435 'info_dict': {
2436 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2437 'title': 'Uploads from lex will',
2438 }
2439 }, {
2440 'note': 'Age restricted channel',
2441 # from https://www.youtube.com/user/DeusExOfficial
2442 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2443 'playlist_mincount': 64,
2444 'info_dict': {
2445 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2446 'title': 'Uploads from Deus Ex',
2447 },
2448 }, {
2449 'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2450 'only_matching': True,
2451 }]
2452
2453 @classmethod
2454 def suitable(cls, url):
2455 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2456 else super(YoutubeChannelIE, cls).suitable(url))
2457
2458 def _build_template_url(self, url, channel_id):
2459 return self._TEMPLATE_URL % channel_id
2460
2461 def _real_extract(self, url):
2462 channel_id = self._match_id(url)
2463
2464 url = self._build_template_url(url, channel_id)
2465
2466 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2467 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2468 # otherwise fallback on channel by page extraction
2469 channel_page = self._download_webpage(
2470 url + '?view=57', channel_id,
2471 'Downloading channel page', fatal=False)
2472 if channel_page is False:
2473 channel_playlist_id = False
2474 else:
2475 channel_playlist_id = self._html_search_meta(
2476 'channelId', channel_page, 'channel id', default=None)
2477 if not channel_playlist_id:
2478 channel_url = self._html_search_meta(
2479 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2480 channel_page, 'channel url', default=None)
2481 if channel_url:
2482 channel_playlist_id = self._search_regex(
2483 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2484 channel_url, 'channel id', default=None)
2485 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2486 playlist_id = 'UU' + channel_playlist_id[2:]
2487 return self.url_result(
2488 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2489
2490 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2491 autogenerated = re.search(r'''(?x)
2492 class="[^"]*?(?:
2493 channel-header-autogenerated-label|
2494 yt-channel-title-autogenerated
2495 )[^"]*"''', channel_page) is not None
2496
2497 if autogenerated:
2498 # The videos are contained in a single page
2499 # the ajax pages can't be used, they are empty
2500 entries = [
2501 self.url_result(
2502 video_id, 'Youtube', video_id=video_id,
2503 video_title=video_title)
2504 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2505 return self.playlist_result(entries, channel_id)
2506
2507 try:
2508 next(self._entries(channel_page, channel_id))
2509 except StopIteration:
2510 alert_message = self._html_search_regex(
2511 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2512 channel_page, 'alert', default=None, group='alert')
2513 if alert_message:
2514 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2515
2516 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2517
2518
2519 class YoutubeUserIE(YoutubeChannelIE):
2520 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2521 _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_-]+)'
2522 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2523 IE_NAME = 'youtube:user'
2524
2525 _TESTS = [{
2526 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2527 'playlist_mincount': 320,
2528 'info_dict': {
2529 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2530 'title': 'Uploads from The Linux Foundation',
2531 }
2532 }, {
2533 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2534 # but not https://www.youtube.com/user/12minuteathlete/videos
2535 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2536 'playlist_mincount': 249,
2537 'info_dict': {
2538 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2539 'title': 'Uploads from 12 Minute Athlete',
2540 }
2541 }, {
2542 'url': 'ytuser:phihag',
2543 'only_matching': True,
2544 }, {
2545 'url': 'https://www.youtube.com/c/gametrailers',
2546 'only_matching': True,
2547 }, {
2548 'url': 'https://www.youtube.com/gametrailers',
2549 'only_matching': True,
2550 }, {
2551 # This channel is not available, geo restricted to JP
2552 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2553 'only_matching': True,
2554 }]
2555
2556 @classmethod
2557 def suitable(cls, url):
2558 # Don't return True if the url can be extracted with other youtube
2559 # extractor, the regex would is too permissive and it would match.
2560 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2561 if any(ie.suitable(url) for ie in other_yt_ies):
2562 return False
2563 else:
2564 return super(YoutubeUserIE, cls).suitable(url)
2565
2566 def _build_template_url(self, url, channel_id):
2567 mobj = re.match(self._VALID_URL, url)
2568 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2569
2570
2571 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2572 IE_DESC = 'YouTube.com live streams'
2573 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2574 IE_NAME = 'youtube:live'
2575
2576 _TESTS = [{
2577 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2578 'info_dict': {
2579 'id': 'a48o2S1cPoo',
2580 'ext': 'mp4',
2581 'title': 'The Young Turks - Live Main Show',
2582 'uploader': 'The Young Turks',
2583 'uploader_id': 'TheYoungTurks',
2584 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2585 'upload_date': '20150715',
2586 'license': 'Standard YouTube License',
2587 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2588 'categories': ['News & Politics'],
2589 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2590 'like_count': int,
2591 'dislike_count': int,
2592 },
2593 'params': {
2594 'skip_download': True,
2595 },
2596 }, {
2597 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2598 'only_matching': True,
2599 }, {
2600 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2601 'only_matching': True,
2602 }, {
2603 'url': 'https://www.youtube.com/TheYoungTurks/live',
2604 'only_matching': True,
2605 }]
2606
2607 def _real_extract(self, url):
2608 mobj = re.match(self._VALID_URL, url)
2609 channel_id = mobj.group('id')
2610 base_url = mobj.group('base_url')
2611 webpage = self._download_webpage(url, channel_id, fatal=False)
2612 if webpage:
2613 page_type = self._og_search_property(
2614 'type', webpage, 'page type', default='')
2615 video_id = self._html_search_meta(
2616 'videoId', webpage, 'video id', default=None)
2617 if page_type.startswith('video') and video_id and re.match(
2618 r'^[0-9A-Za-z_-]{11}$', video_id):
2619 return self.url_result(video_id, YoutubeIE.ie_key())
2620 return self.url_result(base_url)
2621
2622
2623 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2624 IE_DESC = 'YouTube.com user/channel playlists'
2625 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2626 IE_NAME = 'youtube:playlists'
2627
2628 _TESTS = [{
2629 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2630 'playlist_mincount': 4,
2631 'info_dict': {
2632 'id': 'ThirstForScience',
2633 'title': 'Thirst for Science',
2634 },
2635 }, {
2636 # with "Load more" button
2637 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2638 'playlist_mincount': 70,
2639 'info_dict': {
2640 'id': 'igorkle1',
2641 'title': 'Игорь Клейнер',
2642 },
2643 }, {
2644 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2645 'playlist_mincount': 17,
2646 'info_dict': {
2647 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2648 'title': 'Chem Player',
2649 },
2650 }]
2651
2652
2653 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2654 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2655
2656
2657 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2658 IE_DESC = 'YouTube.com searches'
2659 # there doesn't appear to be a real limit, for example if you search for
2660 # 'python' you get more than 8.000.000 results
2661 _MAX_RESULTS = float('inf')
2662 IE_NAME = 'youtube:search'
2663 _SEARCH_KEY = 'ytsearch'
2664 _EXTRA_QUERY_ARGS = {}
2665 _TESTS = []
2666
2667 def _get_n_results(self, query, n):
2668 """Get a specified number of results for a query"""
2669
2670 videos = []
2671 limit = n
2672
2673 url_query = {
2674 'search_query': query.encode('utf-8'),
2675 }
2676 url_query.update(self._EXTRA_QUERY_ARGS)
2677 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2678
2679 for pagenum in itertools.count(1):
2680 data = self._download_json(
2681 result_url, video_id='query "%s"' % query,
2682 note='Downloading page %s' % pagenum,
2683 errnote='Unable to download API page',
2684 query={'spf': 'navigate'})
2685 html_content = data[1]['body']['content']
2686
2687 if 'class="search-message' in html_content:
2688 raise ExtractorError(
2689 '[youtube] No video results', expected=True)
2690
2691 new_videos = list(self._process_page(html_content))
2692 videos += new_videos
2693 if not new_videos or len(videos) > limit:
2694 break
2695 next_link = self._html_search_regex(
2696 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2697 html_content, 'next link', default=None)
2698 if next_link is None:
2699 break
2700 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2701
2702 if len(videos) > n:
2703 videos = videos[:n]
2704 return self.playlist_result(videos, query)
2705
2706
2707 class YoutubeSearchDateIE(YoutubeSearchIE):
2708 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2709 _SEARCH_KEY = 'ytsearchdate'
2710 IE_DESC = 'YouTube.com searches, newest videos first'
2711 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2712
2713
2714 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2715 IE_DESC = 'YouTube.com search URLs'
2716 IE_NAME = 'youtube:search_url'
2717 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2718 _TESTS = [{
2719 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2720 'playlist_mincount': 5,
2721 'info_dict': {
2722 'title': 'youtube-dl test video',
2723 }
2724 }, {
2725 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2726 'only_matching': True,
2727 }]
2728
2729 def _real_extract(self, url):
2730 mobj = re.match(self._VALID_URL, url)
2731 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2732 webpage = self._download_webpage(url, query)
2733 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2734
2735
2736 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2737 IE_DESC = 'YouTube.com (multi-season) shows'
2738 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2739 IE_NAME = 'youtube:show'
2740 _TESTS = [{
2741 'url': 'https://www.youtube.com/show/airdisasters',
2742 'playlist_mincount': 5,
2743 'info_dict': {
2744 'id': 'airdisasters',
2745 'title': 'Air Disasters',
2746 }
2747 }]
2748
2749 def _real_extract(self, url):
2750 playlist_id = self._match_id(url)
2751 return super(YoutubeShowIE, self)._real_extract(
2752 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2753
2754
2755 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2756 """
2757 Base class for feed extractors
2758 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2759 """
2760 _LOGIN_REQUIRED = True
2761
2762 @property
2763 def IE_NAME(self):
2764 return 'youtube:%s' % self._FEED_NAME
2765
2766 def _real_initialize(self):
2767 self._login()
2768
2769 def _entries(self, page):
2770 # The extraction process is the same as for playlists, but the regex
2771 # for the video ids doesn't contain an index
2772 ids = []
2773 more_widget_html = content_html = page
2774 for page_num in itertools.count(1):
2775 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2776
2777 # 'recommended' feed has infinite 'load more' and each new portion spins
2778 # the same videos in (sometimes) slightly different order, so we'll check
2779 # for unicity and break when portion has no new videos
2780 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2781 if not new_ids:
2782 break
2783
2784 ids.extend(new_ids)
2785
2786 for entry in self._ids_to_results(new_ids):
2787 yield entry
2788
2789 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2790 if not mobj:
2791 break
2792
2793 more = self._download_json(
2794 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2795 'Downloading page #%s' % page_num,
2796 transform_source=uppercase_escape)
2797 content_html = more['content_html']
2798 more_widget_html = more['load_more_widget_html']
2799
2800 def _real_extract(self, url):
2801 page = self._download_webpage(
2802 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2803 self._PLAYLIST_TITLE)
2804 return self.playlist_result(
2805 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2806
2807
2808 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2809 IE_NAME = 'youtube:watchlater'
2810 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2811 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2812
2813 _TESTS = [{
2814 'url': 'https://www.youtube.com/playlist?list=WL',
2815 'only_matching': True,
2816 }, {
2817 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2818 'only_matching': True,
2819 }]
2820
2821 def _real_extract(self, url):
2822 _, video = self._check_download_just_video(url, 'WL')
2823 if video:
2824 return video
2825 _, playlist = self._extract_playlist('WL')
2826 return playlist
2827
2828
2829 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2830 IE_NAME = 'youtube:favorites'
2831 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2832 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2833 _LOGIN_REQUIRED = True
2834
2835 def _real_extract(self, url):
2836 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2837 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2838 return self.url_result(playlist_id, 'YoutubePlaylist')
2839
2840
2841 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2842 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2843 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2844 _FEED_NAME = 'recommended'
2845 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2846
2847
2848 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2849 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2850 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2851 _FEED_NAME = 'subscriptions'
2852 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2853
2854
2855 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2856 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2857 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2858 _FEED_NAME = 'history'
2859 _PLAYLIST_TITLE = 'Youtube History'
2860
2861
2862 class YoutubeTruncatedURLIE(InfoExtractor):
2863 IE_NAME = 'youtube:truncated_url'
2864 IE_DESC = False # Do not list
2865 _VALID_URL = r'''(?x)
2866 (?:https?://)?
2867 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2868 (?:watch\?(?:
2869 feature=[a-z_]+|
2870 annotation_id=annotation_[^&]+|
2871 x-yt-cl=[0-9]+|
2872 hl=[^&]*|
2873 t=[0-9]+
2874 )?
2875 |
2876 attribution_link\?a=[^&]+
2877 )
2878 $
2879 '''
2880
2881 _TESTS = [{
2882 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2883 'only_matching': True,
2884 }, {
2885 'url': 'https://www.youtube.com/watch?',
2886 'only_matching': True,
2887 }, {
2888 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2889 'only_matching': True,
2890 }, {
2891 'url': 'https://www.youtube.com/watch?feature=foo',
2892 'only_matching': True,
2893 }, {
2894 'url': 'https://www.youtube.com/watch?hl=en-GB',
2895 'only_matching': True,
2896 }, {
2897 'url': 'https://www.youtube.com/watch?t=2372',
2898 'only_matching': True,
2899 }]
2900
2901 def _real_extract(self, url):
2902 raise ExtractorError(
2903 'Did you forget to quote the URL? Remember that & is a meta '
2904 'character in most shells, so you want to put the URL in quotes, '
2905 'like youtube-dl '
2906 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2907 ' or simply youtube-dl BaW_jenozKc .',
2908 expected=True)
2909
2910
2911 class YoutubeTruncatedIDIE(InfoExtractor):
2912 IE_NAME = 'youtube:truncated_id'
2913 IE_DESC = False # Do not list
2914 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2915
2916 _TESTS = [{
2917 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2918 'only_matching': True,
2919 }]
2920
2921 def _real_extract(self, url):
2922 video_id = self._match_id(url)
2923 raise ExtractorError(
2924 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2925 expected=True)