]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
Rapsys Git / youtubedl / blob
? search:
summary | shortlog | log | commit | commitdiff | tree
history | raw | HEAD
Updated version 2017.09.24 from 'upstream/2017.09.24'
[youtubedl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_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 remove_quotes,
41 remove_start,
42 smuggle_url,
43 str_to_int,
44 try_get,
45 unescapeHTML,
46 unified_strdate,
47 unsmuggle_url,
48 uppercase_escape,
49 urlencode_postdata,
50 )
51
52
53 class YoutubeBaseInfoExtractor(InfoExtractor):
54 """Provide base functions for Youtube extractors"""
55 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
56 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
57
58 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
59 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
60 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
61
62 _NETRC_MACHINE = 'youtube'
63 # If True it will raise an error if no login info is provided
64 _LOGIN_REQUIRED = False
65
66 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
67
68 def _set_language(self):
69 self._set_cookie(
70 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
71 # YouTube sets the expire time to about two months
72 expire_time=time.time() + 2 * 30 * 24 * 3600)
73
74 def _ids_to_results(self, ids):
75 return [
76 self.url_result(vid_id, 'Youtube', video_id=vid_id)
77 for vid_id in ids]
78
79 def _login(self):
80 """
81 Attempt to log in to YouTube.
82 True is returned if successful or skipped.
83 False is returned if login failed.
84
85 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
86 """
87 (username, password) = self._get_login_info()
88 # No authentication to be performed
89 if username is None:
90 if self._LOGIN_REQUIRED:
91 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
92 return True
93
94 login_page = self._download_webpage(
95 self._LOGIN_URL, None,
96 note='Downloading login page',
97 errnote='unable to fetch login page', fatal=False)
98 if login_page is False:
99 return
100
101 login_form = self._hidden_inputs(login_page)
102
103 def req(url, f_req, note, errnote):
104 data = login_form.copy()
105 data.update({
106 'pstMsg': 1,
107 'checkConnection': 'youtube',
108 'checkedDomains': 'youtube',
109 'hl': 'en',
110 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
111 'f.req': json.dumps(f_req),
112 'flowName': 'GlifWebSignIn',
113 'flowEntry': 'ServiceLogin',
114 })
115 return self._download_json(
116 url, None, note=note, errnote=errnote,
117 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
118 fatal=False,
119 data=urlencode_postdata(data), headers={
120 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
121 'Google-Accounts-XSRF': 1,
122 })
123
124 def warn(message):
125 self._downloader.report_warning(message)
126
127 lookup_req = [
128 username,
129 None, [], None, 'US', None, None, 2, False, True,
130 [
131 None, None,
132 [2, 1, None, 1,
133 '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',
134 None, [], 4],
135 1, [None, None, []], None, None, None, True
136 ],
137 username,
138 ]
139
140 lookup_results = req(
141 self._LOOKUP_URL, lookup_req,
142 'Looking up account info', 'Unable to look up account info')
143
144 if lookup_results is False:
145 return False
146
147 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
148 if not user_hash:
149 warn('Unable to extract user hash')
150 return False
151
152 challenge_req = [
153 user_hash,
154 None, 1, None, [1, None, None, None, [password, None, True]],
155 [
156 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],
157 1, [None, None, []], None, None, None, True
158 ]]
159
160 challenge_results = req(
161 self._CHALLENGE_URL, challenge_req,
162 'Logging in', 'Unable to log in')
163
164 if challenge_results is False:
165 return
166
167 login_res = try_get(challenge_results, lambda x: x[0][5], list)
168 if login_res:
169 login_msg = try_get(login_res, lambda x: x[5], compat_str)
170 warn(
171 'Unable to login: %s' % 'Invalid password'
172 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
173 return False
174
175 res = try_get(challenge_results, lambda x: x[0][-1], list)
176 if not res:
177 warn('Unable to extract result entry')
178 return False
179
180 tfa = try_get(res, lambda x: x[0][0], list)
181 if tfa:
182 tfa_str = try_get(tfa, lambda x: x[2], compat_str)
183 if tfa_str == 'TWO_STEP_VERIFICATION':
184 # SEND_SUCCESS - TFA code has been successfully sent to phone
185 # QUOTA_EXCEEDED - reached the limit of TFA codes
186 status = try_get(tfa, lambda x: x[5], compat_str)
187 if status == 'QUOTA_EXCEEDED':
188 warn('Exceeded the limit of TFA codes, try later')
189 return False
190
191 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
192 if not tl:
193 warn('Unable to extract TL')
194 return False
195
196 tfa_code = self._get_tfa_info('2-step verification code')
197
198 if not tfa_code:
199 warn(
200 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
201 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
202 return False
203
204 tfa_code = remove_start(tfa_code, 'G-')
205
206 tfa_req = [
207 user_hash, None, 2, None,
208 [
209 9, None, None, None, None, None, None, None,
210 [None, tfa_code, True, 2]
211 ]]
212
213 tfa_results = req(
214 self._TFA_URL.format(tl), tfa_req,
215 'Submitting TFA code', 'Unable to submit TFA code')
216
217 if tfa_results is False:
218 return False
219
220 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
221 if tfa_res:
222 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
223 warn(
224 'Unable to finish TFA: %s' % 'Invalid TFA code'
225 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
226 return False
227
228 check_cookie_url = try_get(
229 tfa_results, lambda x: x[0][-1][2], compat_str)
230 else:
231 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
232
233 if not check_cookie_url:
234 warn('Unable to extract CheckCookie URL')
235 return False
236
237 check_cookie_results = self._download_webpage(
238 check_cookie_url, None, 'Checking cookie', fatal=False)
239
240 if check_cookie_results is False:
241 return False
242
243 if 'https://myaccount.google.com/' not in check_cookie_results:
244 warn('Unable to log in')
245 return False
246
247 return True
248
249 def _download_webpage(self, *args, **kwargs):
250 kwargs.setdefault('query', {})['disable_polymer'] = 'true'
251 return super(YoutubeBaseInfoExtractor, self)._download_webpage(
252 *args, **compat_kwargs(kwargs))
253
254 def _real_initialize(self):
255 if self._downloader is None:
256 return
257 self._set_language()
258 if not self._login():
259 return
260
261
262 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
263 # Extract entries from page with "Load more" button
264 def _entries(self, page, playlist_id):
265 more_widget_html = content_html = page
266 for page_num in itertools.count(1):
267 for entry in self._process_page(content_html):
268 yield entry
269
270 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
271 if not mobj:
272 break
273
274 more = self._download_json(
275 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
276 'Downloading page #%s' % page_num,
277 transform_source=uppercase_escape)
278 content_html = more['content_html']
279 if not content_html.strip():
280 # Some webpages show a "Load more" button but they don't
281 # have more videos
282 break
283 more_widget_html = more['load_more_widget_html']
284
285
286 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
287 def _process_page(self, content):
288 for video_id, video_title in self.extract_videos_from_page(content):
289 yield self.url_result(video_id, 'Youtube', video_id, video_title)
290
291 def extract_videos_from_page(self, page):
292 ids_in_page = []
293 titles_in_page = []
294 for mobj in re.finditer(self._VIDEO_RE, page):
295 # The link with index 0 is not the first video of the playlist (not sure if still actual)
296 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
297 continue
298 video_id = mobj.group('id')
299 video_title = unescapeHTML(mobj.group('title'))
300 if video_title:
301 video_title = video_title.strip()
302 try:
303 idx = ids_in_page.index(video_id)
304 if video_title and not titles_in_page[idx]:
305 titles_in_page[idx] = video_title
306 except ValueError:
307 ids_in_page.append(video_id)
308 titles_in_page.append(video_title)
309 return zip(ids_in_page, titles_in_page)
310
311
312 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
313 def _process_page(self, content):
314 for playlist_id in orderedSet(re.findall(
315 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
316 content)):
317 yield self.url_result(
318 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
319
320 def _real_extract(self, url):
321 playlist_id = self._match_id(url)
322 webpage = self._download_webpage(url, playlist_id)
323 title = self._og_search_title(webpage, fatal=False)
324 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
325
326
327 class YoutubeIE(YoutubeBaseInfoExtractor):
328 IE_DESC = 'YouTube.com'
329 _VALID_URL = r"""(?x)^
330 (
331 (?:https?://|//) # http(s):// or protocol-independent URL
332 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
333 (?:www\.)?deturl\.com/www\.youtube\.com/|
334 (?:www\.)?pwnyoutube\.com/|
335 (?:www\.)?yourepeat\.com/|
336 tube\.majestyc\.net/|
337 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
338 (?:.*?\#/)? # handle anchor (#/) redirect urls
339 (?: # the various things that can precede the ID:
340 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
341 |(?: # or the v= param in all its forms
342 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
343 (?:\?|\#!?) # the params delimiter ? or # or #!
344 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
345 v=
346 )
347 ))
348 |(?:
349 youtu\.be| # just youtu.be/xxxx
350 vid\.plus| # or vid.plus/xxxx
351 zwearz\.com/watch| # or zwearz.com/watch/xxxx
352 )/
353 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
354 )
355 )? # all until now is optional -> you can pass the naked ID
356 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
357 (?!.*?\blist=
358 (?:
359 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
360 WL # WL are handled by the watch later IE
361 )
362 )
363 (?(1).+)? # if we found the ID, everything can follow
364 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
365 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
366 _formats = {
367 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
368 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
369 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
370 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
371 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
372 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
373 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
374 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
375 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
376 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
377 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
378 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
379 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
380 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
381 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
382 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
383 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
384 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
385
386
387 # 3D videos
388 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
389 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
390 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
391 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
392 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
393 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
394 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
395
396 # Apple HTTP Live Streaming
397 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
398 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
399 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
400 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
401 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
402 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
403 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
404 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
405
406 # DASH mp4 video
407 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
408 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
409 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
410 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
411 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
412 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
413 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
414 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
415 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
416 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
417 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
418 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
419
420 # Dash mp4 audio
421 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
422 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
423 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
424 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
425 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
426 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
427 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
428
429 # Dash webm
430 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
431 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
432 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
433 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
434 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
435 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
436 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
437 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
438 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
439 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
440 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
441 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
442 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
443 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
444 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
445 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
446 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
447 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
448 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
449 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
450 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
451 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
452
453 # Dash webm audio
454 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
455 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
456
457 # Dash webm audio with opus inside
458 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
459 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
460 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
461
462 # RTMP (unnamed)
463 '_rtmp': {'protocol': 'rtmp'},
464 }
465 _SUBTITLE_FORMATS = ('ttml', 'vtt')
466
467 _GEO_BYPASS = False
468
469 IE_NAME = 'youtube'
470 _TESTS = [
471 {
472 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
473 'info_dict': {
474 'id': 'BaW_jenozKc',
475 'ext': 'mp4',
476 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
477 'uploader': 'Philipp Hagemeister',
478 'uploader_id': 'phihag',
479 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
480 'upload_date': '20121002',
481 'license': 'Standard YouTube License',
482 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
483 'categories': ['Science & Technology'],
484 'tags': ['youtube-dl'],
485 'duration': 10,
486 'like_count': int,
487 'dislike_count': int,
488 'start_time': 1,
489 'end_time': 9,
490 }
491 },
492 {
493 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
494 'note': 'Test generic use_cipher_signature video (#897)',
495 'info_dict': {
496 'id': 'UxxajLWwzqY',
497 'ext': 'mp4',
498 'upload_date': '20120506',
499 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
500 'alt_title': 'I Love It (feat. Charli XCX)',
501 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
502 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
503 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
504 'iconic ep', 'iconic', 'love', 'it'],
505 'duration': 180,
506 'uploader': 'Icona Pop',
507 'uploader_id': 'IconaPop',
508 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
509 'license': 'Standard YouTube License',
510 'creator': 'Icona Pop',
511 }
512 },
513 {
514 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
515 'note': 'Test VEVO video with age protection (#956)',
516 'info_dict': {
517 'id': '07FYdnEawAQ',
518 'ext': 'mp4',
519 'upload_date': '20130703',
520 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
521 'alt_title': 'Tunnel Vision',
522 'description': 'md5:64249768eec3bc4276236606ea996373',
523 'duration': 419,
524 'uploader': 'justintimberlakeVEVO',
525 'uploader_id': 'justintimberlakeVEVO',
526 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
527 'license': 'Standard YouTube License',
528 'creator': 'Justin Timberlake',
529 'age_limit': 18,
530 }
531 },
532 {
533 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
534 'note': 'Embed-only video (#1746)',
535 'info_dict': {
536 'id': 'yZIXLfi8CZQ',
537 'ext': 'mp4',
538 'upload_date': '20120608',
539 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
540 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
541 'uploader': 'SET India',
542 'uploader_id': 'setindia',
543 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
544 'license': 'Standard YouTube License',
545 'age_limit': 18,
546 }
547 },
548 {
549 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
550 'note': 'Use the first video ID in the URL',
551 'info_dict': {
552 'id': 'BaW_jenozKc',
553 'ext': 'mp4',
554 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
555 'uploader': 'Philipp Hagemeister',
556 'uploader_id': 'phihag',
557 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
558 'upload_date': '20121002',
559 'license': 'Standard YouTube License',
560 'description': 'test chars: "\'/\\Ƥā†­š•\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
561 'categories': ['Science & Technology'],
562 'tags': ['youtube-dl'],
563 'duration': 10,
564 'like_count': int,
565 'dislike_count': int,
566 },
567 'params': {
568 'skip_download': True,
569 },
570 },
571 {
572 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
573 'note': '256k DASH audio (format 141) via DASH manifest',
574 'info_dict': {
575 'id': 'a9LDPn-MO4I',
576 'ext': 'm4a',
577 'upload_date': '20121002',
578 'uploader_id': '8KVIDEO',
579 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
580 'description': '',
581 'uploader': '8KVIDEO',
582 'license': 'Standard YouTube License',
583 'title': 'UHDTV TEST 8K VIDEO.mp4'
584 },
585 'params': {
586 'youtube_include_dash_manifest': True,
587 'format': '141',
588 },
589 'skip': 'format 141 not served anymore',
590 },
591 # DASH manifest with encrypted signature
592 {
593 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
594 'info_dict': {
595 'id': 'IB3lcPjvWLA',
596 'ext': 'm4a',
597 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
598 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
599 'duration': 244,
600 'uploader': 'AfrojackVEVO',
601 'uploader_id': 'AfrojackVEVO',
602 'upload_date': '20131011',
603 'license': 'Standard YouTube License',
604 },
605 'params': {
606 'youtube_include_dash_manifest': True,
607 'format': '141/bestaudio[ext=m4a]',
608 },
609 },
610 # JS player signature function name containing $
611 {
612 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
613 'info_dict': {
614 'id': 'nfWlot6h_JM',
615 'ext': 'm4a',
616 'title': 'Taylor Swift - Shake It Off',
617 'alt_title': 'Shake It Off',
618 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
619 'duration': 242,
620 'uploader': 'TaylorSwiftVEVO',
621 'uploader_id': 'TaylorSwiftVEVO',
622 'upload_date': '20140818',
623 'license': 'Standard YouTube License',
624 'creator': 'Taylor Swift',
625 },
626 'params': {
627 'youtube_include_dash_manifest': True,
628 'format': '141/bestaudio[ext=m4a]',
629 },
630 },
631 # Controversy video
632 {
633 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
634 'info_dict': {
635 'id': 'T4XJQO3qol8',
636 'ext': 'mp4',
637 'duration': 219,
638 'upload_date': '20100909',
639 'uploader': 'The Amazing Atheist',
640 'uploader_id': 'TheAmazingAtheist',
641 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
642 'license': 'Standard YouTube License',
643 'title': 'Burning Everyone\'s Koran',
644 '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',
645 }
646 },
647 # Normal age-gate video (No vevo, embed allowed)
648 {
649 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
650 'info_dict': {
651 'id': 'HtVdAasjOgU',
652 'ext': 'mp4',
653 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
654 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
655 'duration': 142,
656 'uploader': 'The Witcher',
657 'uploader_id': 'WitcherGame',
658 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
659 'upload_date': '20140605',
660 'license': 'Standard YouTube License',
661 'age_limit': 18,
662 },
663 },
664 # Age-gate video with encrypted signature
665 {
666 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
667 'info_dict': {
668 'id': '6kLq3WMV1nU',
669 'ext': 'mp4',
670 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
671 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
672 'duration': 247,
673 'uploader': 'LloydVEVO',
674 'uploader_id': 'LloydVEVO',
675 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
676 'upload_date': '20110629',
677 'license': 'Standard YouTube License',
678 'age_limit': 18,
679 },
680 },
681 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
682 # YouTube Red ad is not captured for creator
683 {
684 'url': '__2ABJjxzNo',
685 'info_dict': {
686 'id': '__2ABJjxzNo',
687 'ext': 'mp4',
688 'duration': 266,
689 'upload_date': '20100430',
690 'uploader_id': 'deadmau5',
691 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
692 'creator': 'deadmau5',
693 'description': 'md5:12c56784b8032162bb936a5f76d55360',
694 'uploader': 'deadmau5',
695 'license': 'Standard YouTube License',
696 'title': 'Deadmau5 - Some Chords (HD)',
697 'alt_title': 'Some Chords',
698 },
699 'expected_warnings': [
700 'DASH manifest missing',
701 ]
702 },
703 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
704 {
705 'url': 'lqQg6PlCWgI',
706 'info_dict': {
707 'id': 'lqQg6PlCWgI',
708 'ext': 'mp4',
709 'duration': 6085,
710 'upload_date': '20150827',
711 'uploader_id': 'olympic',
712 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
713 'license': 'Standard YouTube License',
714 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
715 'uploader': 'Olympic',
716 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
717 },
718 'params': {
719 'skip_download': 'requires avconv',
720 }
721 },
722 # Non-square pixels
723 {
724 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
725 'info_dict': {
726 'id': '_b-2C3KPAM0',
727 'ext': 'mp4',
728 'stretched_ratio': 16 / 9.,
729 'duration': 85,
730 'upload_date': '20110310',
731 'uploader_id': 'AllenMeow',
732 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
733 'description': 'made by Wacom from Korea | 字幕&åŠ ę²¹ę·»é†‹ by TY\'s Allen | ę„Ÿč¬heylisa00cavey1001同å­øē†±ęƒ…ęä¾›ę¢—åŠēæ»č­Æ',
734 'uploader': 'å­«č‰¾å€«',
735 'license': 'Standard YouTube License',
736 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
737 },
738 },
739 # url_encoded_fmt_stream_map is empty string
740 {
741 'url': 'qEJwOuvDf7I',
742 'info_dict': {
743 'id': 'qEJwOuvDf7I',
744 'ext': 'webm',
745 'title': 'ŠžŠ±ŃŃƒŠ¶Š“ŠµŠ½ŠøŠµ суŠ“ŠµŠ±Š½Š¾Š¹ ŠæрŠ°ŠŗтŠøŠŗŠø ŠæŠ¾ Š²Ń‹Š±Š¾Ń€Š°Š¼ 14 сŠµŠ½Ń‚яŠ±Ń€Ń 2014 Š³Š¾Š“Š° Š² Š”Š°Š½Šŗт-ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Šµ',
746 'description': '',
747 'upload_date': '20150404',
748 'uploader_id': 'spbelect',
749 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
750 },
751 'params': {
752 'skip_download': 'requires avconv',
753 },
754 'skip': 'This live event has ended.',
755 },
756 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
757 {
758 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
759 'info_dict': {
760 'id': 'FIl7x6_3R5Y',
761 'ext': 'mp4',
762 'title': 'md5:7b81415841e02ecd4313668cde88737a',
763 'description': 'md5:116377fd2963b81ec4ce64b542173306',
764 'duration': 220,
765 'upload_date': '20150625',
766 'uploader_id': 'dorappi2000',
767 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
768 'uploader': 'dorappi2000',
769 'license': 'Standard YouTube License',
770 'formats': 'mincount:32',
771 },
772 },
773 # DASH manifest with segment_list
774 {
775 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
776 'md5': '8ce563a1d667b599d21064e982ab9e31',
777 'info_dict': {
778 'id': 'CsmdDsKjzN8',
779 'ext': 'mp4',
780 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
781 'uploader': 'Airtek',
782 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
783 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
784 'license': 'Standard YouTube License',
785 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
786 },
787 'params': {
788 'youtube_include_dash_manifest': True,
789 'format': '135', # bestvideo
790 },
791 'skip': 'This live event has ended.',
792 },
793 {
794 # Multifeed videos (multiple cameras), URL is for Main Camera
795 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
796 'info_dict': {
797 'id': 'jqWvoWXjCVs',
798 'title': 'teamPGP: Rocket League Noob Stream',
799 'description': 'md5:dc7872fb300e143831327f1bae3af010',
800 },
801 'playlist': [{
802 'info_dict': {
803 'id': 'jqWvoWXjCVs',
804 'ext': 'mp4',
805 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
806 'description': 'md5:dc7872fb300e143831327f1bae3af010',
807 'duration': 7335,
808 'upload_date': '20150721',
809 'uploader': 'Beer Games Beer',
810 'uploader_id': 'beergamesbeer',
811 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
812 'license': 'Standard YouTube License',
813 },
814 }, {
815 'info_dict': {
816 'id': '6h8e8xoXJzg',
817 'ext': 'mp4',
818 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
819 'description': 'md5:dc7872fb300e143831327f1bae3af010',
820 'duration': 7337,
821 'upload_date': '20150721',
822 'uploader': 'Beer Games Beer',
823 'uploader_id': 'beergamesbeer',
824 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
825 'license': 'Standard YouTube License',
826 },
827 }, {
828 'info_dict': {
829 'id': 'PUOgX5z9xZw',
830 'ext': 'mp4',
831 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
832 'description': 'md5:dc7872fb300e143831327f1bae3af010',
833 'duration': 7337,
834 'upload_date': '20150721',
835 'uploader': 'Beer Games Beer',
836 'uploader_id': 'beergamesbeer',
837 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
838 'license': 'Standard YouTube License',
839 },
840 }, {
841 'info_dict': {
842 'id': 'teuwxikvS5k',
843 'ext': 'mp4',
844 'title': 'teamPGP: Rocket League Noob Stream (zim)',
845 'description': 'md5:dc7872fb300e143831327f1bae3af010',
846 'duration': 7334,
847 'upload_date': '20150721',
848 'uploader': 'Beer Games Beer',
849 'uploader_id': 'beergamesbeer',
850 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
851 'license': 'Standard YouTube License',
852 },
853 }],
854 'params': {
855 'skip_download': True,
856 },
857 },
858 {
859 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
860 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
861 'info_dict': {
862 'id': 'gVfLd0zydlo',
863 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
864 },
865 'playlist_count': 2,
866 'skip': 'Not multifeed anymore',
867 },
868 {
869 'url': 'https://vid.plus/FlRa-iH7PGw',
870 'only_matching': True,
871 },
872 {
873 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
874 'only_matching': True,
875 },
876 {
877 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
878 # Also tests cut-off URL expansion in video description (see
879 # https://github.com/rg3/youtube-dl/issues/1892,
880 # https://github.com/rg3/youtube-dl/issues/8164)
881 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
882 'info_dict': {
883 'id': 'lsguqyKfVQg',
884 'ext': 'mp4',
885 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
886 'alt_title': 'Dark Walk',
887 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
888 'duration': 133,
889 'upload_date': '20151119',
890 'uploader_id': 'IronSoulElf',
891 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
892 'uploader': 'IronSoulElf',
893 'license': 'Standard YouTube License',
894 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
895 },
896 'params': {
897 'skip_download': True,
898 },
899 },
900 {
901 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
902 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
903 'only_matching': True,
904 },
905 {
906 # Video with yt:stretch=17:0
907 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
908 'info_dict': {
909 'id': 'Q39EVAstoRM',
910 'ext': 'mp4',
911 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
912 'description': 'md5:ee18a25c350637c8faff806845bddee9',
913 'upload_date': '20151107',
914 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
915 'uploader': 'CH GAMER DROID',
916 },
917 'params': {
918 'skip_download': True,
919 },
920 'skip': 'This video does not exist.',
921 },
922 {
923 # Video licensed under Creative Commons
924 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
925 'info_dict': {
926 'id': 'M4gD1WSo5mA',
927 'ext': 'mp4',
928 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
929 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
930 'duration': 721,
931 'upload_date': '20150127',
932 'uploader_id': 'BerkmanCenter',
933 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
934 'uploader': 'The Berkman Klein Center for Internet & Society',
935 'license': 'Creative Commons Attribution license (reuse allowed)',
936 },
937 'params': {
938 'skip_download': True,
939 },
940 },
941 {
942 # Channel-like uploader_url
943 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
944 'info_dict': {
945 'id': 'eQcmzGIKrzg',
946 'ext': 'mp4',
947 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
948 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
949 'duration': 4060,
950 'upload_date': '20151119',
951 'uploader': 'Bernie 2016',
952 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
953 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
954 'license': 'Creative Commons Attribution license (reuse allowed)',
955 },
956 'params': {
957 'skip_download': True,
958 },
959 },
960 {
961 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
962 'only_matching': True,
963 },
964 {
965 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
966 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
967 'only_matching': True,
968 },
969 {
970 # Rental video preview
971 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
972 'info_dict': {
973 'id': 'uGpuVWrhIzE',
974 'ext': 'mp4',
975 'title': 'Piku - Trailer',
976 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
977 'upload_date': '20150811',
978 'uploader': 'FlixMatrix',
979 'uploader_id': 'FlixMatrixKaravan',
980 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
981 'license': 'Standard YouTube License',
982 },
983 'params': {
984 'skip_download': True,
985 },
986 },
987 {
988 # YouTube Red video with episode data
989 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
990 'info_dict': {
991 'id': 'iqKdEhx-dD4',
992 'ext': 'mp4',
993 'title': 'Isolation - Mind Field (Ep 1)',
994 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
995 'duration': 2085,
996 'upload_date': '20170118',
997 'uploader': 'Vsauce',
998 'uploader_id': 'Vsauce',
999 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1000 'license': 'Standard YouTube License',
1001 'series': 'Mind Field',
1002 'season_number': 1,
1003 'episode_number': 1,
1004 },
1005 'params': {
1006 'skip_download': True,
1007 },
1008 'expected_warnings': [
1009 'Skipping DASH manifest',
1010 ],
1011 },
1012 {
1013 # The following content has been identified by the YouTube community
1014 # as inappropriate or offensive to some audiences.
1015 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1016 'info_dict': {
1017 'id': '6SJNVb0GnPI',
1018 'ext': 'mp4',
1019 'title': 'Race Differences in Intelligence',
1020 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1021 'duration': 965,
1022 'upload_date': '20140124',
1023 'uploader': 'New Century Foundation',
1024 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1026 'license': 'Standard YouTube License',
1027 'view_count': int,
1028 },
1029 'params': {
1030 'skip_download': True,
1031 },
1032 },
1033 {
1034 # itag 212
1035 'url': '1t24XAntNCY',
1036 'only_matching': True,
1037 },
1038 {
1039 # geo restricted to JP
1040 'url': 'sJL6WA-aGkQ',
1041 'only_matching': True,
1042 },
1043 {
1044 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1045 'only_matching': True,
1046 },
1047 ]
1048
1049 def __init__(self, *args, **kwargs):
1050 super(YoutubeIE, self).__init__(*args, **kwargs)
1051 self._player_cache = {}
1052
1053 def report_video_info_webpage_download(self, video_id):
1054 """Report attempt to download video info webpage."""
1055 self.to_screen('%s: Downloading video info webpage' % video_id)
1056
1057 def report_information_extraction(self, video_id):
1058 """Report attempt to extract video information."""
1059 self.to_screen('%s: Extracting video information' % video_id)
1060
1061 def report_unavailable_format(self, video_id, format):
1062 """Report extracted video URL."""
1063 self.to_screen('%s: Format %s not available' % (video_id, format))
1064
1065 def report_rtmp_download(self):
1066 """Indicate the download will use the RTMP protocol."""
1067 self.to_screen('RTMP download detected')
1068
1069 def _signature_cache_id(self, example_sig):
1070 """ Return a string representation of a signature """
1071 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1072
1073 def _extract_signature_function(self, video_id, player_url, example_sig):
1074 id_m = re.match(
1075 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1076 player_url)
1077 if not id_m:
1078 raise ExtractorError('Cannot identify player %r' % player_url)
1079 player_type = id_m.group('ext')
1080 player_id = id_m.group('id')
1081
1082 # Read from filesystem cache
1083 func_id = '%s_%s_%s' % (
1084 player_type, player_id, self._signature_cache_id(example_sig))
1085 assert os.path.basename(func_id) == func_id
1086
1087 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1088 if cache_spec is not None:
1089 return lambda s: ''.join(s[i] for i in cache_spec)
1090
1091 download_note = (
1092 'Downloading player %s' % player_url
1093 if self._downloader.params.get('verbose') else
1094 'Downloading %s player %s' % (player_type, player_id)
1095 )
1096 if player_type == 'js':
1097 code = self._download_webpage(
1098 player_url, video_id,
1099 note=download_note,
1100 errnote='Download of %s failed' % player_url)
1101 res = self._parse_sig_js(code)
1102 elif player_type == 'swf':
1103 urlh = self._request_webpage(
1104 player_url, video_id,
1105 note=download_note,
1106 errnote='Download of %s failed' % player_url)
1107 code = urlh.read()
1108 res = self._parse_sig_swf(code)
1109 else:
1110 assert False, 'Invalid player type %r' % player_type
1111
1112 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1113 cache_res = res(test_string)
1114 cache_spec = [ord(c) for c in cache_res]
1115
1116 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1117 return res
1118
1119 def _print_sig_code(self, func, example_sig):
1120 def gen_sig_code(idxs):
1121 def _genslice(start, end, step):
1122 starts = '' if start == 0 else str(start)
1123 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1124 steps = '' if step == 1 else (':%d' % step)
1125 return 's[%s%s%s]' % (starts, ends, steps)
1126
1127 step = None
1128 # Quelch pyflakes warnings - start will be set when step is set
1129 start = '(Never used)'
1130 for i, prev in zip(idxs[1:], idxs[:-1]):
1131 if step is not None:
1132 if i - prev == step:
1133 continue
1134 yield _genslice(start, prev, step)
1135 step = None
1136 continue
1137 if i - prev in [-1, 1]:
1138 step = i - prev
1139 start = prev
1140 continue
1141 else:
1142 yield 's[%d]' % prev
1143 if step is None:
1144 yield 's[%d]' % i
1145 else:
1146 yield _genslice(start, i, step)
1147
1148 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1149 cache_res = func(test_string)
1150 cache_spec = [ord(c) for c in cache_res]
1151 expr_code = ' + '.join(gen_sig_code(cache_spec))
1152 signature_id_tuple = '(%s)' % (
1153 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1154 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1155 ' return %s\n') % (signature_id_tuple, expr_code)
1156 self.to_screen('Extracted signature function:\n' + code)
1157
1158 def _parse_sig_js(self, jscode):
1159 funcname = self._search_regex(
1160 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1161 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1162 jscode, 'Initial JS player signature function name', group='sig')
1163
1164 jsi = JSInterpreter(jscode)
1165 initial_function = jsi.extract_function(funcname)
1166 return lambda s: initial_function([s])
1167
1168 def _parse_sig_swf(self, file_contents):
1169 swfi = SWFInterpreter(file_contents)
1170 TARGET_CLASSNAME = 'SignatureDecipher'
1171 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1172 initial_function = swfi.extract_function(searched_class, 'decipher')
1173 return lambda s: initial_function([s])
1174
1175 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1176 """Turn the encrypted s field into a working signature"""
1177
1178 if player_url is None:
1179 raise ExtractorError('Cannot decrypt signature without player_url')
1180
1181 if player_url.startswith('//'):
1182 player_url = 'https:' + player_url
1183 elif not re.match(r'https?://', player_url):
1184 player_url = compat_urlparse.urljoin(
1185 'https://www.youtube.com', player_url)
1186 try:
1187 player_id = (player_url, self._signature_cache_id(s))
1188 if player_id not in self._player_cache:
1189 func = self._extract_signature_function(
1190 video_id, player_url, s
1191 )
1192 self._player_cache[player_id] = func
1193 func = self._player_cache[player_id]
1194 if self._downloader.params.get('youtube_print_sig_code'):
1195 self._print_sig_code(func, s)
1196 return func(s)
1197 except Exception as e:
1198 tb = traceback.format_exc()
1199 raise ExtractorError(
1200 'Signature extraction failed: ' + tb, cause=e)
1201
1202 def _get_subtitles(self, video_id, webpage):
1203 try:
1204 subs_doc = self._download_xml(
1205 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1206 video_id, note=False)
1207 except ExtractorError as err:
1208 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1209 return {}
1210
1211 sub_lang_list = {}
1212 for track in subs_doc.findall('track'):
1213 lang = track.attrib['lang_code']
1214 if lang in sub_lang_list:
1215 continue
1216 sub_formats = []
1217 for ext in self._SUBTITLE_FORMATS:
1218 params = compat_urllib_parse_urlencode({
1219 'lang': lang,
1220 'v': video_id,
1221 'fmt': ext,
1222 'name': track.attrib['name'].encode('utf-8'),
1223 })
1224 sub_formats.append({
1225 'url': 'https://www.youtube.com/api/timedtext?' + params,
1226 'ext': ext,
1227 })
1228 sub_lang_list[lang] = sub_formats
1229 if not sub_lang_list:
1230 self._downloader.report_warning('video doesn\'t have subtitles')
1231 return {}
1232 return sub_lang_list
1233
1234 def _get_ytplayer_config(self, video_id, webpage):
1235 patterns = (
1236 # User data may contain arbitrary character sequences that may affect
1237 # JSON extraction with regex, e.g. when '};' is contained the second
1238 # regex won't capture the whole JSON. Yet working around by trying more
1239 # concrete regex first keeping in mind proper quoted string handling
1240 # to be implemented in future that will replace this workaround (see
1241 # https://github.com/rg3/youtube-dl/issues/7468,
1242 # https://github.com/rg3/youtube-dl/pull/7599)
1243 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1244 r';ytplayer\.config\s*=\s*({.+?});',
1245 )
1246 config = self._search_regex(
1247 patterns, webpage, 'ytplayer.config', default=None)
1248 if config:
1249 return self._parse_json(
1250 uppercase_escape(config), video_id, fatal=False)
1251
1252 def _get_automatic_captions(self, video_id, webpage):
1253 """We need the webpage for getting the captions url, pass it as an
1254 argument to speed up the process."""
1255 self.to_screen('%s: Looking for automatic captions' % video_id)
1256 player_config = self._get_ytplayer_config(video_id, webpage)
1257 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1258 if not player_config:
1259 self._downloader.report_warning(err_msg)
1260 return {}
1261 try:
1262 args = player_config['args']
1263 caption_url = args.get('ttsurl')
1264 if caption_url:
1265 timestamp = args['timestamp']
1266 # We get the available subtitles
1267 list_params = compat_urllib_parse_urlencode({
1268 'type': 'list',
1269 'tlangs': 1,
1270 'asrs': 1,
1271 })
1272 list_url = caption_url + '&' + list_params
1273 caption_list = self._download_xml(list_url, video_id)
1274 original_lang_node = caption_list.find('track')
1275 if original_lang_node is None:
1276 self._downloader.report_warning('Video doesn\'t have automatic captions')
1277 return {}
1278 original_lang = original_lang_node.attrib['lang_code']
1279 caption_kind = original_lang_node.attrib.get('kind', '')
1280
1281 sub_lang_list = {}
1282 for lang_node in caption_list.findall('target'):
1283 sub_lang = lang_node.attrib['lang_code']
1284 sub_formats = []
1285 for ext in self._SUBTITLE_FORMATS:
1286 params = compat_urllib_parse_urlencode({
1287 'lang': original_lang,
1288 'tlang': sub_lang,
1289 'fmt': ext,
1290 'ts': timestamp,
1291 'kind': caption_kind,
1292 })
1293 sub_formats.append({
1294 'url': caption_url + '&' + params,
1295 'ext': ext,
1296 })
1297 sub_lang_list[sub_lang] = sub_formats
1298 return sub_lang_list
1299
1300 def make_captions(sub_url, sub_langs):
1301 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1302 caption_qs = compat_parse_qs(parsed_sub_url.query)
1303 captions = {}
1304 for sub_lang in sub_langs:
1305 sub_formats = []
1306 for ext in self._SUBTITLE_FORMATS:
1307 caption_qs.update({
1308 'tlang': [sub_lang],
1309 'fmt': [ext],
1310 })
1311 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1312 query=compat_urllib_parse_urlencode(caption_qs, True)))
1313 sub_formats.append({
1314 'url': sub_url,
1315 'ext': ext,
1316 })
1317 captions[sub_lang] = sub_formats
1318 return captions
1319
1320 # New captions format as of 22.06.2017
1321 player_response = args.get('player_response')
1322 if player_response and isinstance(player_response, compat_str):
1323 player_response = self._parse_json(
1324 player_response, video_id, fatal=False)
1325 if player_response:
1326 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1327 base_url = renderer['captionTracks'][0]['baseUrl']
1328 sub_lang_list = []
1329 for lang in renderer['translationLanguages']:
1330 lang_code = lang.get('languageCode')
1331 if lang_code:
1332 sub_lang_list.append(lang_code)
1333 return make_captions(base_url, sub_lang_list)
1334
1335 # Some videos don't provide ttsurl but rather caption_tracks and
1336 # caption_translation_languages (e.g. 20LmZk1hakA)
1337 # Does not used anymore as of 22.06.2017
1338 caption_tracks = args['caption_tracks']
1339 caption_translation_languages = args['caption_translation_languages']
1340 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1341 sub_lang_list = []
1342 for lang in caption_translation_languages.split(','):
1343 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1344 sub_lang = lang_qs.get('lc', [None])[0]
1345 if sub_lang:
1346 sub_lang_list.append(sub_lang)
1347 return make_captions(caption_url, sub_lang_list)
1348 # An extractor error can be raise by the download process if there are
1349 # no automatic captions but there are subtitles
1350 except (KeyError, IndexError, ExtractorError):
1351 self._downloader.report_warning(err_msg)
1352 return {}
1353
1354 def _mark_watched(self, video_id, video_info):
1355 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1356 if not playback_url:
1357 return
1358 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1359 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1360
1361 # cpn generation algorithm is reverse engineered from base.js.
1362 # In fact it works even with dummy cpn.
1363 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1364 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1365
1366 qs.update({
1367 'ver': ['2'],
1368 'cpn': [cpn],
1369 })
1370 playback_url = compat_urlparse.urlunparse(
1371 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1372
1373 self._download_webpage(
1374 playback_url, video_id, 'Marking watched',
1375 'Unable to mark watched', fatal=False)
1376
1377 @staticmethod
1378 def _extract_urls(webpage):
1379 # Embedded YouTube player
1380 entries = [
1381 unescapeHTML(mobj.group('url'))
1382 for mobj in re.finditer(r'''(?x)
1383 (?:
1384 <iframe[^>]+?src=|
1385 data-video-url=|
1386 <embed[^>]+?src=|
1387 embedSWF\(?:\s*|
1388 <object[^>]+data=|
1389 new\s+SWFObject\(
1390 )
1391 (["\'])
1392 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1393 (?:embed|v|p)/.+?)
1394 \1''', webpage)]
1395
1396 # lazyYT YouTube embed
1397 entries.extend(list(map(
1398 unescapeHTML,
1399 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1400
1401 # Wordpress "YouTube Video Importer" plugin
1402 matches = re.findall(r'''(?x)<div[^>]+
1403 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1404 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1405 entries.extend(m[-1] for m in matches)
1406
1407 return entries
1408
1409 @staticmethod
1410 def _extract_url(webpage):
1411 urls = YoutubeIE._extract_urls(webpage)
1412 return urls[0] if urls else None
1413
1414 @classmethod
1415 def extract_id(cls, url):
1416 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1417 if mobj is None:
1418 raise ExtractorError('Invalid URL: %s' % url)
1419 video_id = mobj.group(2)
1420 return video_id
1421
1422 def _extract_annotations(self, video_id):
1423 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1424 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1425
1426 @staticmethod
1427 def _extract_chapters(description, duration):
1428 if not description:
1429 return None
1430 chapter_lines = re.findall(
1431 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1432 description)
1433 if not chapter_lines:
1434 return None
1435 chapters = []
1436 for next_num, (chapter_line, time_point) in enumerate(
1437 chapter_lines, start=1):
1438 start_time = parse_duration(time_point)
1439 if start_time is None:
1440 continue
1441 if start_time > duration:
1442 break
1443 end_time = (duration if next_num == len(chapter_lines)
1444 else parse_duration(chapter_lines[next_num][1]))
1445 if end_time is None:
1446 continue
1447 if end_time > duration:
1448 end_time = duration
1449 if start_time > end_time:
1450 break
1451 chapter_title = re.sub(
1452 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1453 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1454 chapters.append({
1455 'start_time': start_time,
1456 'end_time': end_time,
1457 'title': chapter_title,
1458 })
1459 return chapters
1460
1461 def _real_extract(self, url):
1462 url, smuggled_data = unsmuggle_url(url, {})
1463
1464 proto = (
1465 'http' if self._downloader.params.get('prefer_insecure', False)
1466 else 'https')
1467
1468 start_time = None
1469 end_time = None
1470 parsed_url = compat_urllib_parse_urlparse(url)
1471 for component in [parsed_url.fragment, parsed_url.query]:
1472 query = compat_parse_qs(component)
1473 if start_time is None and 't' in query:
1474 start_time = parse_duration(query['t'][0])
1475 if start_time is None and 'start' in query:
1476 start_time = parse_duration(query['start'][0])
1477 if end_time is None and 'end' in query:
1478 end_time = parse_duration(query['end'][0])
1479
1480 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1481 mobj = re.search(self._NEXT_URL_RE, url)
1482 if mobj:
1483 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1484 video_id = self.extract_id(url)
1485
1486 # Get video webpage
1487 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1488 video_webpage = self._download_webpage(url, video_id)
1489
1490 # Attempt to extract SWF player URL
1491 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1492 if mobj is not None:
1493 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1494 else:
1495 player_url = None
1496
1497 dash_mpds = []
1498
1499 def add_dash_mpd(video_info):
1500 dash_mpd = video_info.get('dashmpd')
1501 if dash_mpd and dash_mpd[0] not in dash_mpds:
1502 dash_mpds.append(dash_mpd[0])
1503
1504 is_live = None
1505 view_count = None
1506
1507 def extract_view_count(v_info):
1508 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1509
1510 # Get video info
1511 embed_webpage = None
1512 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1513 age_gate = True
1514 # We simulate the access to the video from www.youtube.com/v/{video_id}
1515 # this can be viewed without login into Youtube
1516 url = proto + '://www.youtube.com/embed/%s' % video_id
1517 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1518 data = compat_urllib_parse_urlencode({
1519 'video_id': video_id,
1520 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1521 'sts': self._search_regex(
1522 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1523 })
1524 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1525 video_info_webpage = self._download_webpage(
1526 video_info_url, video_id,
1527 note='Refetching age-gated info webpage',
1528 errnote='unable to download video info webpage')
1529 video_info = compat_parse_qs(video_info_webpage)
1530 add_dash_mpd(video_info)
1531 else:
1532 age_gate = False
1533 video_info = None
1534 sts = None
1535 # Try looking directly into the video webpage
1536 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1537 if ytplayer_config:
1538 args = ytplayer_config['args']
1539 if args.get('url_encoded_fmt_stream_map'):
1540 # Convert to the same format returned by compat_parse_qs
1541 video_info = dict((k, [v]) for k, v in args.items())
1542 add_dash_mpd(video_info)
1543 # Rental video is not rented but preview is available (e.g.
1544 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1545 # https://github.com/rg3/youtube-dl/issues/10532)
1546 if not video_info and args.get('ypc_vid'):
1547 return self.url_result(
1548 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1549 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1550 is_live = True
1551 sts = ytplayer_config.get('sts')
1552 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1553 # We also try looking in get_video_info since it may contain different dashmpd
1554 # URL that points to a DASH manifest with possibly different itag set (some itags
1555 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1556 # manifest pointed by get_video_info's dashmpd).
1557 # The general idea is to take a union of itags of both DASH manifests (for example
1558 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1559 self.report_video_info_webpage_download(video_id)
1560 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1561 query = {
1562 'video_id': video_id,
1563 'ps': 'default',
1564 'eurl': '',
1565 'gl': 'US',
1566 'hl': 'en',
1567 }
1568 if el:
1569 query['el'] = el
1570 if sts:
1571 query['sts'] = sts
1572 video_info_webpage = self._download_webpage(
1573 '%s://www.youtube.com/get_video_info' % proto,
1574 video_id, note=False,
1575 errnote='unable to download video info webpage',
1576 fatal=False, query=query)
1577 if not video_info_webpage:
1578 continue
1579 get_video_info = compat_parse_qs(video_info_webpage)
1580 add_dash_mpd(get_video_info)
1581 if view_count is None:
1582 view_count = extract_view_count(get_video_info)
1583 if not video_info:
1584 video_info = get_video_info
1585 if 'token' in get_video_info:
1586 # Different get_video_info requests may report different results, e.g.
1587 # some may report video unavailability, but some may serve it without
1588 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1589 # the original webpage as well as el=info and el=embedded get_video_info
1590 # requests report video unavailability due to geo restriction while
1591 # el=detailpage succeeds and returns valid data). This is probably
1592 # due to YouTube measures against IP ranges of hosting providers.
1593 # Working around by preferring the first succeeded video_info containing
1594 # the token if no such video_info yet was found.
1595 if 'token' not in video_info:
1596 video_info = get_video_info
1597 break
1598 if 'token' not in video_info:
1599 if 'reason' in video_info:
1600 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1601 regions_allowed = self._html_search_meta(
1602 'regionsAllowed', video_webpage, default=None)
1603 countries = regions_allowed.split(',') if regions_allowed else None
1604 self.raise_geo_restricted(
1605 msg=video_info['reason'][0], countries=countries)
1606 raise ExtractorError(
1607 'YouTube said: %s' % video_info['reason'][0],
1608 expected=True, video_id=video_id)
1609 else:
1610 raise ExtractorError(
1611 '"token" parameter not in video info for unknown reason',
1612 video_id=video_id)
1613
1614 # title
1615 if 'title' in video_info:
1616 video_title = video_info['title'][0]
1617 else:
1618 self._downloader.report_warning('Unable to extract video title')
1619 video_title = '_'
1620
1621 # description
1622 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1623 if video_description:
1624 description_original = video_description = re.sub(r'''(?x)
1625 <a\s+
1626 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1627 (?:title|href)="([^"]+)"\s+
1628 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1629 class="[^"]*"[^>]*>
1630 [^<]+\.{3}\s*
1631 </a>
1632 ''', r'\1', video_description)
1633 video_description = clean_html(video_description)
1634 else:
1635 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1636 if fd_mobj:
1637 video_description = unescapeHTML(fd_mobj.group(1))
1638 else:
1639 video_description = ''
1640
1641 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1642 if not self._downloader.params.get('noplaylist'):
1643 entries = []
1644 feed_ids = []
1645 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1646 for feed in multifeed_metadata_list.split(','):
1647 # Unquote should take place before split on comma (,) since textual
1648 # fields may contain comma as well (see
1649 # https://github.com/rg3/youtube-dl/issues/8536)
1650 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1651 entries.append({
1652 '_type': 'url_transparent',
1653 'ie_key': 'Youtube',
1654 'url': smuggle_url(
1655 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1656 {'force_singlefeed': True}),
1657 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1658 })
1659 feed_ids.append(feed_data['id'][0])
1660 self.to_screen(
1661 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1662 % (', '.join(feed_ids), video_id))
1663 return self.playlist_result(entries, video_id, video_title, video_description)
1664 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1665
1666 if view_count is None:
1667 view_count = extract_view_count(video_info)
1668
1669 # Check for "rental" videos
1670 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1671 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1672
1673 # Start extracting information
1674 self.report_information_extraction(video_id)
1675
1676 # uploader
1677 if 'author' not in video_info:
1678 raise ExtractorError('Unable to extract uploader name')
1679 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1680
1681 # uploader_id
1682 video_uploader_id = None
1683 video_uploader_url = None
1684 mobj = re.search(
1685 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1686 video_webpage)
1687 if mobj is not None:
1688 video_uploader_id = mobj.group('uploader_id')
1689 video_uploader_url = mobj.group('uploader_url')
1690 else:
1691 self._downloader.report_warning('unable to extract uploader nickname')
1692
1693 # thumbnail image
1694 # We try first to get a high quality image:
1695 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1696 video_webpage, re.DOTALL)
1697 if m_thumb is not None:
1698 video_thumbnail = m_thumb.group(1)
1699 elif 'thumbnail_url' not in video_info:
1700 self._downloader.report_warning('unable to extract video thumbnail')
1701 video_thumbnail = None
1702 else: # don't panic if we can't find it
1703 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1704
1705 # upload date
1706 upload_date = self._html_search_meta(
1707 'datePublished', video_webpage, 'upload date', default=None)
1708 if not upload_date:
1709 upload_date = self._search_regex(
1710 [r'(?s)id="eow-date.*?>(.*?)</span>',
1711 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1712 video_webpage, 'upload date', default=None)
1713 upload_date = unified_strdate(upload_date)
1714
1715 video_license = self._html_search_regex(
1716 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1717 video_webpage, 'license', default=None)
1718
1719 m_music = re.search(
1720 r'''(?x)
1721 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1722 <ul[^>]*>\s*
1723 <li>(?P<title>.+?)
1724 by (?P<creator>.+?)
1725 (?:
1726 \(.+?\)|
1727 <a[^>]*
1728 (?:
1729 \bhref=["\']/red[^>]*>| # drop possible
1730 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1731 )
1732 .*?
1733 )?</li
1734 ''',
1735 video_webpage)
1736 if m_music:
1737 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1738 video_creator = clean_html(m_music.group('creator'))
1739 else:
1740 video_alt_title = video_creator = None
1741
1742 m_episode = re.search(
1743 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
1744 video_webpage)
1745 if m_episode:
1746 series = m_episode.group('series')
1747 season_number = int(m_episode.group('season'))
1748 episode_number = int(m_episode.group('episode'))
1749 else:
1750 series = season_number = episode_number = None
1751
1752 m_cat_container = self._search_regex(
1753 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1754 video_webpage, 'categories', default=None)
1755 if m_cat_container:
1756 category = self._html_search_regex(
1757 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1758 default=None)
1759 video_categories = None if category is None else [category]
1760 else:
1761 video_categories = None
1762
1763 video_tags = [
1764 unescapeHTML(m.group('content'))
1765 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1766
1767 def _extract_count(count_name):
1768 return str_to_int(self._search_regex(
1769 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1770 % re.escape(count_name),
1771 video_webpage, count_name, default=None))
1772
1773 like_count = _extract_count('like')
1774 dislike_count = _extract_count('dislike')
1775
1776 # subtitles
1777 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1778 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1779
1780 video_duration = try_get(
1781 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1782 if not video_duration:
1783 video_duration = parse_duration(self._html_search_meta(
1784 'duration', video_webpage, 'video duration'))
1785
1786 # annotations
1787 video_annotations = None
1788 if self._downloader.params.get('writeannotations', False):
1789 video_annotations = self._extract_annotations(video_id)
1790
1791 chapters = self._extract_chapters(description_original, video_duration)
1792
1793 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1794 self.report_rtmp_download()
1795 formats = [{
1796 'format_id': '_rtmp',
1797 'protocol': 'rtmp',
1798 'url': video_info['conn'][0],
1799 'player_url': player_url,
1800 }]
1801 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1802 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1803 if 'rtmpe%3Dyes' in encoded_url_map:
1804 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1805 formats_spec = {}
1806 fmt_list = video_info.get('fmt_list', [''])[0]
1807 if fmt_list:
1808 for fmt in fmt_list.split(','):
1809 spec = fmt.split('/')
1810 if len(spec) > 1:
1811 width_height = spec[1].split('x')
1812 if len(width_height) == 2:
1813 formats_spec[spec[0]] = {
1814 'resolution': spec[1],
1815 'width': int_or_none(width_height[0]),
1816 'height': int_or_none(width_height[1]),
1817 }
1818 formats = []
1819 for url_data_str in encoded_url_map.split(','):
1820 url_data = compat_parse_qs(url_data_str)
1821 if 'itag' not in url_data or 'url' not in url_data:
1822 continue
1823 format_id = url_data['itag'][0]
1824 url = url_data['url'][0]
1825
1826 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1827 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1828 jsplayer_url_json = self._search_regex(
1829 ASSETS_RE,
1830 embed_webpage if age_gate else video_webpage,
1831 'JS player URL (1)', default=None)
1832 if not jsplayer_url_json and not age_gate:
1833 # We need the embed website after all
1834 if embed_webpage is None:
1835 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1836 embed_webpage = self._download_webpage(
1837 embed_url, video_id, 'Downloading embed webpage')
1838 jsplayer_url_json = self._search_regex(
1839 ASSETS_RE, embed_webpage, 'JS player URL')
1840
1841 player_url = json.loads(jsplayer_url_json)
1842 if player_url is None:
1843 player_url_json = self._search_regex(
1844 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1845 video_webpage, 'age gate player URL')
1846 player_url = json.loads(player_url_json)
1847
1848 if 'sig' in url_data:
1849 url += '&signature=' + url_data['sig'][0]
1850 elif 's' in url_data:
1851 encrypted_sig = url_data['s'][0]
1852
1853 if self._downloader.params.get('verbose'):
1854 if player_url is None:
1855 player_version = 'unknown'
1856 player_desc = 'unknown'
1857 else:
1858 if player_url.endswith('swf'):
1859 player_version = self._search_regex(
1860 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1861 'flash player', fatal=False)
1862 player_desc = 'flash player %s' % player_version
1863 else:
1864 player_version = self._search_regex(
1865 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1866 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1867 player_url,
1868 'html5 player', fatal=False)
1869 player_desc = 'html5 player %s' % player_version
1870
1871 parts_sizes = self._signature_cache_id(encrypted_sig)
1872 self.to_screen('{%s} signature length %s, %s' %
1873 (format_id, parts_sizes, player_desc))
1874
1875 signature = self._decrypt_signature(
1876 encrypted_sig, video_id, player_url, age_gate)
1877 url += '&signature=' + signature
1878 if 'ratebypass' not in url:
1879 url += '&ratebypass=yes'
1880
1881 dct = {
1882 'format_id': format_id,
1883 'url': url,
1884 'player_url': player_url,
1885 }
1886 if format_id in self._formats:
1887 dct.update(self._formats[format_id])
1888 if format_id in formats_spec:
1889 dct.update(formats_spec[format_id])
1890
1891 # Some itags are not included in DASH manifest thus corresponding formats will
1892 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1893 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1894 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1895 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1896
1897 more_fields = {
1898 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1899 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1900 'width': width,
1901 'height': height,
1902 'fps': int_or_none(url_data.get('fps', [None])[0]),
1903 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1904 }
1905 for key, value in more_fields.items():
1906 if value:
1907 dct[key] = value
1908 type_ = url_data.get('type', [None])[0]
1909 if type_:
1910 type_split = type_.split(';')
1911 kind_ext = type_split[0].split('/')
1912 if len(kind_ext) == 2:
1913 kind, _ = kind_ext
1914 dct['ext'] = mimetype2ext(type_split[0])
1915 if kind in ('audio', 'video'):
1916 codecs = None
1917 for mobj in re.finditer(
1918 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1919 if mobj.group('key') == 'codecs':
1920 codecs = mobj.group('val')
1921 break
1922 if codecs:
1923 dct.update(parse_codecs(codecs))
1924 formats.append(dct)
1925 elif video_info.get('hlsvp'):
1926 manifest_url = video_info['hlsvp'][0]
1927 formats = []
1928 m3u8_formats = self._extract_m3u8_formats(
1929 manifest_url, video_id, 'mp4', fatal=False)
1930 for a_format in m3u8_formats:
1931 itag = self._search_regex(
1932 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1933 if itag:
1934 a_format['format_id'] = itag
1935 if itag in self._formats:
1936 dct = self._formats[itag].copy()
1937 dct.update(a_format)
1938 a_format = dct
1939 a_format['player_url'] = player_url
1940 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1941 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1942 formats.append(a_format)
1943 else:
1944 unavailable_message = self._html_search_regex(
1945 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1946 video_webpage, 'unavailable message', default=None)
1947 if unavailable_message:
1948 raise ExtractorError(unavailable_message, expected=True)
1949 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1950
1951 # Look for the DASH manifest
1952 if self._downloader.params.get('youtube_include_dash_manifest', True):
1953 dash_mpd_fatal = True
1954 for mpd_url in dash_mpds:
1955 dash_formats = {}
1956 try:
1957 def decrypt_sig(mobj):
1958 s = mobj.group(1)
1959 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1960 return '/signature/%s' % dec_s
1961
1962 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1963
1964 for df in self._extract_mpd_formats(
1965 mpd_url, video_id, fatal=dash_mpd_fatal,
1966 formats_dict=self._formats):
1967 # Do not overwrite DASH format found in some previous DASH manifest
1968 if df['format_id'] not in dash_formats:
1969 dash_formats[df['format_id']] = df
1970 # Additional DASH manifests may end up in HTTP Error 403 therefore
1971 # allow them to fail without bug report message if we already have
1972 # some DASH manifest succeeded. This is temporary workaround to reduce
1973 # burst of bug reports until we figure out the reason and whether it
1974 # can be fixed at all.
1975 dash_mpd_fatal = False
1976 except (ExtractorError, KeyError) as e:
1977 self.report_warning(
1978 'Skipping DASH manifest: %r' % e, video_id)
1979 if dash_formats:
1980 # Remove the formats we found through non-DASH, they
1981 # contain less info and it can be wrong, because we use
1982 # fixed values (for example the resolution). See
1983 # https://github.com/rg3/youtube-dl/issues/5774 for an
1984 # example.
1985 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1986 formats.extend(dash_formats.values())
1987
1988 # Check for malformed aspect ratio
1989 stretched_m = re.search(
1990 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1991 video_webpage)
1992 if stretched_m:
1993 w = float(stretched_m.group('w'))
1994 h = float(stretched_m.group('h'))
1995 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1996 # We will only process correct ratios.
1997 if w > 0 and h > 0:
1998 ratio = w / h
1999 for f in formats:
2000 if f.get('vcodec') != 'none':
2001 f['stretched_ratio'] = ratio
2002
2003 self._sort_formats(formats)
2004
2005 self.mark_watched(video_id, video_info)
2006
2007 return {
2008 'id': video_id,
2009 'uploader': video_uploader,
2010 'uploader_id': video_uploader_id,
2011 'uploader_url': video_uploader_url,
2012 'upload_date': upload_date,
2013 'license': video_license,
2014 'creator': video_creator,
2015 'title': video_title,
2016 'alt_title': video_alt_title,
2017 'thumbnail': video_thumbnail,
2018 'description': video_description,
2019 'categories': video_categories,
2020 'tags': video_tags,
2021 'subtitles': video_subtitles,
2022 'automatic_captions': automatic_captions,
2023 'duration': video_duration,
2024 'age_limit': 18 if age_gate else 0,
2025 'annotations': video_annotations,
2026 'chapters': chapters,
2027 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2028 'view_count': view_count,
2029 'like_count': like_count,
2030 'dislike_count': dislike_count,
2031 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2032 'formats': formats,
2033 'is_live': is_live,
2034 'start_time': start_time,
2035 'end_time': end_time,
2036 'series': series,
2037 'season_number': season_number,
2038 'episode_number': episode_number,
2039 }
2040
2041
2042 class YoutubeSharedVideoIE(InfoExtractor):
2043 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
2044 IE_NAME = 'youtube:shared'
2045
2046 _TEST = {
2047 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
2048 'info_dict': {
2049 'id': 'uPDB5I9wfp8',
2050 'ext': 'webm',
2051 'title': 'Pocoyo: 90 minutos de episĆ³dios completos PortuguĆŖs para crianƧas - PARTE 3',
2052 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
2053 'upload_date': '20160219',
2054 'uploader': 'Pocoyo - PortuguĆŖs (BR)',
2055 'uploader_id': 'PocoyoBrazil',
2056 },
2057 'add_ie': ['Youtube'],
2058 'params': {
2059 # There are already too many Youtube downloads
2060 'skip_download': True,
2061 },
2062 }
2063
2064 def _real_extract(self, url):
2065 video_id = self._match_id(url)
2066
2067 webpage = self._download_webpage(url, video_id)
2068
2069 real_video_id = self._html_search_meta(
2070 'videoId', webpage, 'YouTube video id', fatal=True)
2071
2072 return self.url_result(real_video_id, YoutubeIE.ie_key())
2073
2074
2075 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2076 IE_DESC = 'YouTube.com playlists'
2077 _VALID_URL = r"""(?x)(?:
2078 (?:https?://)?
2079 (?:\w+\.)?
2080 (?:
2081 youtube\.com/
2082 (?:
2083 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2084 \? (?:.*?[&;])*? (?:p|a|list)=
2085 | p/
2086 )|
2087 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2088 )
2089 (
2090 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2091 # Top tracks, they can also include dots
2092 |(?:MC)[\w\.]*
2093 )
2094 .*
2095 |
2096 (%(playlist_id)s)
2097 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2098 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2099 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2100 IE_NAME = 'youtube:playlist'
2101 _TESTS = [{
2102 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2103 'info_dict': {
2104 'title': 'ytdl test PL',
2105 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2106 },
2107 'playlist_count': 3,
2108 }, {
2109 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2110 'info_dict': {
2111 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2112 'title': 'YDL_Empty_List',
2113 },
2114 'playlist_count': 0,
2115 'skip': 'This playlist is private',
2116 }, {
2117 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2118 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2119 'info_dict': {
2120 'title': '29C3: Not my department',
2121 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2122 },
2123 'playlist_count': 95,
2124 }, {
2125 'note': 'issue #673',
2126 'url': 'PLBB231211A4F62143',
2127 'info_dict': {
2128 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2129 'id': 'PLBB231211A4F62143',
2130 },
2131 'playlist_mincount': 26,
2132 }, {
2133 'note': 'Large playlist',
2134 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2135 'info_dict': {
2136 'title': 'Uploads from Cauchemar',
2137 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2138 },
2139 'playlist_mincount': 799,
2140 }, {
2141 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2142 'info_dict': {
2143 'title': 'YDL_safe_search',
2144 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2145 },
2146 'playlist_count': 2,
2147 'skip': 'This playlist is private',
2148 }, {
2149 'note': 'embedded',
2150 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2151 'playlist_count': 4,
2152 'info_dict': {
2153 'title': 'JODA15',
2154 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2155 }
2156 }, {
2157 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2158 'playlist_mincount': 485,
2159 'info_dict': {
2160 'title': '2017 čÆčŖžęœ€ę–°å–®ę›² (2/24ꛓꖰ)',
2161 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2162 }
2163 }, {
2164 'note': 'Embedded SWF player',
2165 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2166 'playlist_count': 4,
2167 'info_dict': {
2168 'title': 'JODA7',
2169 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2170 }
2171 }, {
2172 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2173 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2174 'info_dict': {
2175 'title': 'Uploads from Interstellar Movie',
2176 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2177 },
2178 'playlist_mincount': 21,
2179 }, {
2180 # Playlist URL that does not actually serve a playlist
2181 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2182 'info_dict': {
2183 'id': 'FqZTN594JQw',
2184 'ext': 'webm',
2185 'title': "Smiley's People 01 detective, Adventure Series, Action",
2186 'uploader': 'STREEM',
2187 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2188 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2189 'upload_date': '20150526',
2190 'license': 'Standard YouTube License',
2191 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2192 'categories': ['People & Blogs'],
2193 'tags': list,
2194 'like_count': int,
2195 'dislike_count': int,
2196 },
2197 'params': {
2198 'skip_download': True,
2199 },
2200 'add_ie': [YoutubeIE.ie_key()],
2201 }, {
2202 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2203 'info_dict': {
2204 'id': 'yeWKywCrFtk',
2205 'ext': 'mp4',
2206 'title': 'Small Scale Baler and Braiding Rugs',
2207 'uploader': 'Backus-Page House Museum',
2208 'uploader_id': 'backuspagemuseum',
2209 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2210 'upload_date': '20161008',
2211 'license': 'Standard YouTube License',
2212 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2213 'categories': ['Nonprofits & Activism'],
2214 'tags': list,
2215 'like_count': int,
2216 'dislike_count': int,
2217 },
2218 'params': {
2219 'noplaylist': True,
2220 'skip_download': True,
2221 },
2222 }, {
2223 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2224 'only_matching': True,
2225 }, {
2226 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2227 'only_matching': True,
2228 }]
2229
2230 def _real_initialize(self):
2231 self._login()
2232
2233 def _extract_mix(self, playlist_id):
2234 # The mixes are generated from a single video
2235 # the id of the playlist is just 'RD' + video_id
2236 ids = []
2237 last_id = playlist_id[-11:]
2238 for n in itertools.count(1):
2239 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2240 webpage = self._download_webpage(
2241 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2242 new_ids = orderedSet(re.findall(
2243 r'''(?xs)data-video-username=".*?".*?
2244 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2245 webpage))
2246 # Fetch new pages until all the videos are repeated, it seems that
2247 # there are always 51 unique videos.
2248 new_ids = [_id for _id in new_ids if _id not in ids]
2249 if not new_ids:
2250 break
2251 ids.extend(new_ids)
2252 last_id = ids[-1]
2253
2254 url_results = self._ids_to_results(ids)
2255
2256 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2257 title_span = (
2258 search_title('playlist-title') or
2259 search_title('title long-title') or
2260 search_title('title'))
2261 title = clean_html(title_span)
2262
2263 return self.playlist_result(url_results, playlist_id, title)
2264
2265 def _extract_playlist(self, playlist_id):
2266 url = self._TEMPLATE_URL % playlist_id
2267 page = self._download_webpage(url, playlist_id)
2268
2269 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2270 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2271 match = match.strip()
2272 # Check if the playlist exists or is private
2273 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2274 if mobj:
2275 reason = mobj.group('reason')
2276 message = 'This playlist %s' % reason
2277 if 'private' in reason:
2278 message += ', use --username or --netrc to access it'
2279 message += '.'
2280 raise ExtractorError(message, expected=True)
2281 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2282 raise ExtractorError(
2283 'Invalid parameters. Maybe URL is incorrect.',
2284 expected=True)
2285 elif re.match(r'[^<]*Choose your language[^<]*', match):
2286 continue
2287 else:
2288 self.report_warning('Youtube gives an alert message: ' + match)
2289
2290 playlist_title = self._html_search_regex(
2291 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2292 page, 'title', default=None)
2293
2294 has_videos = True
2295
2296 if not playlist_title:
2297 try:
2298 # Some playlist URLs don't actually serve a playlist (e.g.
2299 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2300 next(self._entries(page, playlist_id))
2301 except StopIteration:
2302 has_videos = False
2303
2304 return has_videos, self.playlist_result(
2305 self._entries(page, playlist_id), playlist_id, playlist_title)
2306
2307 def _check_download_just_video(self, url, playlist_id):
2308 # Check if it's a video-specific URL
2309 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2310 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2311 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2312 'video id', default=None)
2313 if video_id:
2314 if self._downloader.params.get('noplaylist'):
2315 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2316 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2317 else:
2318 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2319 return video_id, None
2320 return None, None
2321
2322 def _real_extract(self, url):
2323 # Extract playlist id
2324 mobj = re.match(self._VALID_URL, url)
2325 if mobj is None:
2326 raise ExtractorError('Invalid URL: %s' % url)
2327 playlist_id = mobj.group(1) or mobj.group(2)
2328
2329 video_id, video = self._check_download_just_video(url, playlist_id)
2330 if video:
2331 return video
2332
2333 if playlist_id.startswith(('RD', 'UL', 'PU')):
2334 # Mixes require a custom extraction process
2335 return self._extract_mix(playlist_id)
2336
2337 has_videos, playlist = self._extract_playlist(playlist_id)
2338 if has_videos or not video_id:
2339 return playlist
2340
2341 # Some playlist URLs don't actually serve a playlist (see
2342 # https://github.com/rg3/youtube-dl/issues/10537).
2343 # Fallback to plain video extraction if there is a video id
2344 # along with playlist id.
2345 return self.url_result(video_id, 'Youtube', video_id=video_id)
2346
2347
2348 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2349 IE_DESC = 'YouTube.com channels'
2350 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2351 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2352 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2353 IE_NAME = 'youtube:channel'
2354 _TESTS = [{
2355 'note': 'paginated channel',
2356 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2357 'playlist_mincount': 91,
2358 'info_dict': {
2359 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2360 'title': 'Uploads from lex will',
2361 }
2362 }, {
2363 'note': 'Age restricted channel',
2364 # from https://www.youtube.com/user/DeusExOfficial
2365 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2366 'playlist_mincount': 64,
2367 'info_dict': {
2368 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2369 'title': 'Uploads from Deus Ex',
2370 },
2371 }]
2372
2373 @classmethod
2374 def suitable(cls, url):
2375 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2376 else super(YoutubeChannelIE, cls).suitable(url))
2377
2378 def _build_template_url(self, url, channel_id):
2379 return self._TEMPLATE_URL % channel_id
2380
2381 def _real_extract(self, url):
2382 channel_id = self._match_id(url)
2383
2384 url = self._build_template_url(url, channel_id)
2385
2386 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2387 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2388 # otherwise fallback on channel by page extraction
2389 channel_page = self._download_webpage(
2390 url + '?view=57', channel_id,
2391 'Downloading channel page', fatal=False)
2392 if channel_page is False:
2393 channel_playlist_id = False
2394 else:
2395 channel_playlist_id = self._html_search_meta(
2396 'channelId', channel_page, 'channel id', default=None)
2397 if not channel_playlist_id:
2398 channel_url = self._html_search_meta(
2399 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2400 channel_page, 'channel url', default=None)
2401 if channel_url:
2402 channel_playlist_id = self._search_regex(
2403 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2404 channel_url, 'channel id', default=None)
2405 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2406 playlist_id = 'UU' + channel_playlist_id[2:]
2407 return self.url_result(
2408 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2409
2410 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2411 autogenerated = re.search(r'''(?x)
2412 class="[^"]*?(?:
2413 channel-header-autogenerated-label|
2414 yt-channel-title-autogenerated
2415 )[^"]*"''', channel_page) is not None
2416
2417 if autogenerated:
2418 # The videos are contained in a single page
2419 # the ajax pages can't be used, they are empty
2420 entries = [
2421 self.url_result(
2422 video_id, 'Youtube', video_id=video_id,
2423 video_title=video_title)
2424 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2425 return self.playlist_result(entries, channel_id)
2426
2427 try:
2428 next(self._entries(channel_page, channel_id))
2429 except StopIteration:
2430 alert_message = self._html_search_regex(
2431 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2432 channel_page, 'alert', default=None, group='alert')
2433 if alert_message:
2434 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2435
2436 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2437
2438
2439 class YoutubeUserIE(YoutubeChannelIE):
2440 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2441 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2442 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2443 IE_NAME = 'youtube:user'
2444
2445 _TESTS = [{
2446 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2447 'playlist_mincount': 320,
2448 'info_dict': {
2449 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2450 'title': 'Uploads from The Linux Foundation',
2451 }
2452 }, {
2453 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2454 # but not https://www.youtube.com/user/12minuteathlete/videos
2455 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2456 'playlist_mincount': 249,
2457 'info_dict': {
2458 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2459 'title': 'Uploads from 12 Minute Athlete',
2460 }
2461 }, {
2462 'url': 'ytuser:phihag',
2463 'only_matching': True,
2464 }, {
2465 'url': 'https://www.youtube.com/c/gametrailers',
2466 'only_matching': True,
2467 }, {
2468 'url': 'https://www.youtube.com/gametrailers',
2469 'only_matching': True,
2470 }, {
2471 # This channel is not available, geo restricted to JP
2472 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2473 'only_matching': True,
2474 }]
2475
2476 @classmethod
2477 def suitable(cls, url):
2478 # Don't return True if the url can be extracted with other youtube
2479 # extractor, the regex would is too permissive and it would match.
2480 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2481 if any(ie.suitable(url) for ie in other_yt_ies):
2482 return False
2483 else:
2484 return super(YoutubeUserIE, cls).suitable(url)
2485
2486 def _build_template_url(self, url, channel_id):
2487 mobj = re.match(self._VALID_URL, url)
2488 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2489
2490
2491 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2492 IE_DESC = 'YouTube.com live streams'
2493 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2494 IE_NAME = 'youtube:live'
2495
2496 _TESTS = [{
2497 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2498 'info_dict': {
2499 'id': 'a48o2S1cPoo',
2500 'ext': 'mp4',
2501 'title': 'The Young Turks - Live Main Show',
2502 'uploader': 'The Young Turks',
2503 'uploader_id': 'TheYoungTurks',
2504 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2505 'upload_date': '20150715',
2506 'license': 'Standard YouTube License',
2507 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2508 'categories': ['News & Politics'],
2509 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2510 'like_count': int,
2511 'dislike_count': int,
2512 },
2513 'params': {
2514 'skip_download': True,
2515 },
2516 }, {
2517 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2518 'only_matching': True,
2519 }, {
2520 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2521 'only_matching': True,
2522 }, {
2523 'url': 'https://www.youtube.com/TheYoungTurks/live',
2524 'only_matching': True,
2525 }]
2526
2527 def _real_extract(self, url):
2528 mobj = re.match(self._VALID_URL, url)
2529 channel_id = mobj.group('id')
2530 base_url = mobj.group('base_url')
2531 webpage = self._download_webpage(url, channel_id, fatal=False)
2532 if webpage:
2533 page_type = self._og_search_property(
2534 'type', webpage, 'page type', default=None)
2535 video_id = self._html_search_meta(
2536 'videoId', webpage, 'video id', default=None)
2537 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2538 return self.url_result(video_id, YoutubeIE.ie_key())
2539 return self.url_result(base_url)
2540
2541
2542 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2543 IE_DESC = 'YouTube.com user/channel playlists'
2544 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2545 IE_NAME = 'youtube:playlists'
2546
2547 _TESTS = [{
2548 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2549 'playlist_mincount': 4,
2550 'info_dict': {
2551 'id': 'ThirstForScience',
2552 'title': 'Thirst for Science',
2553 },
2554 }, {
2555 # with "Load more" button
2556 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2557 'playlist_mincount': 70,
2558 'info_dict': {
2559 'id': 'igorkle1',
2560 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
2561 },
2562 }, {
2563 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2564 'playlist_mincount': 17,
2565 'info_dict': {
2566 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2567 'title': 'Chem Player',
2568 },
2569 }]
2570
2571
2572 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2573 IE_DESC = 'YouTube.com searches'
2574 # there doesn't appear to be a real limit, for example if you search for
2575 # 'python' you get more than 8.000.000 results
2576 _MAX_RESULTS = float('inf')
2577 IE_NAME = 'youtube:search'
2578 _SEARCH_KEY = 'ytsearch'
2579 _EXTRA_QUERY_ARGS = {}
2580 _TESTS = []
2581
2582 def _get_n_results(self, query, n):
2583 """Get a specified number of results for a query"""
2584
2585 videos = []
2586 limit = n
2587
2588 url_query = {
2589 'search_query': query.encode('utf-8'),
2590 }
2591 url_query.update(self._EXTRA_QUERY_ARGS)
2592 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2593
2594 for pagenum in itertools.count(1):
2595 data = self._download_json(
2596 result_url, video_id='query "%s"' % query,
2597 note='Downloading page %s' % pagenum,
2598 errnote='Unable to download API page',
2599 query={'spf': 'navigate'})
2600 html_content = data[1]['body']['content']
2601
2602 if 'class="search-message' in html_content:
2603 raise ExtractorError(
2604 '[youtube] No video results', expected=True)
2605
2606 new_videos = self._ids_to_results(orderedSet(re.findall(
2607 r'href="/watch\?v=(.{11})', html_content)))
2608 videos += new_videos
2609 if not new_videos or len(videos) > limit:
2610 break
2611 next_link = self._html_search_regex(
2612 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2613 html_content, 'next link', default=None)
2614 if next_link is None:
2615 break
2616 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2617
2618 if len(videos) > n:
2619 videos = videos[:n]
2620 return self.playlist_result(videos, query)
2621
2622
2623 class YoutubeSearchDateIE(YoutubeSearchIE):
2624 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2625 _SEARCH_KEY = 'ytsearchdate'
2626 IE_DESC = 'YouTube.com searches, newest videos first'
2627 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2628
2629
2630 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2631 IE_DESC = 'YouTube.com search URLs'
2632 IE_NAME = 'youtube:search_url'
2633 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2634 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2635 _TESTS = [{
2636 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2637 'playlist_mincount': 5,
2638 'info_dict': {
2639 'title': 'youtube-dl test video',
2640 }
2641 }, {
2642 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2643 'only_matching': True,
2644 }]
2645
2646 def _real_extract(self, url):
2647 mobj = re.match(self._VALID_URL, url)
2648 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2649 webpage = self._download_webpage(url, query)
2650 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2651
2652
2653 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2654 IE_DESC = 'YouTube.com (multi-season) shows'
2655 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2656 IE_NAME = 'youtube:show'
2657 _TESTS = [{
2658 'url': 'https://www.youtube.com/show/airdisasters',
2659 'playlist_mincount': 5,
2660 'info_dict': {
2661 'id': 'airdisasters',
2662 'title': 'Air Disasters',
2663 }
2664 }]
2665
2666 def _real_extract(self, url):
2667 playlist_id = self._match_id(url)
2668 return super(YoutubeShowIE, self)._real_extract(
2669 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2670
2671
2672 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2673 """
2674 Base class for feed extractors
2675 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2676 """
2677 _LOGIN_REQUIRED = True
2678
2679 @property
2680 def IE_NAME(self):
2681 return 'youtube:%s' % self._FEED_NAME
2682
2683 def _real_initialize(self):
2684 self._login()
2685
2686 def _real_extract(self, url):
2687 page = self._download_webpage(
2688 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2689
2690 # The extraction process is the same as for playlists, but the regex
2691 # for the video ids doesn't contain an index
2692 ids = []
2693 more_widget_html = content_html = page
2694 for page_num in itertools.count(1):
2695 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2696
2697 # 'recommended' feed has infinite 'load more' and each new portion spins
2698 # the same videos in (sometimes) slightly different order, so we'll check
2699 # for unicity and break when portion has no new videos
2700 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2701 if not new_ids:
2702 break
2703
2704 ids.extend(new_ids)
2705
2706 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2707 if not mobj:
2708 break
2709
2710 more = self._download_json(
2711 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2712 'Downloading page #%s' % page_num,
2713 transform_source=uppercase_escape)
2714 content_html = more['content_html']
2715 more_widget_html = more['load_more_widget_html']
2716
2717 return self.playlist_result(
2718 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2719
2720
2721 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2722 IE_NAME = 'youtube:watchlater'
2723 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2724 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2725
2726 _TESTS = [{
2727 'url': 'https://www.youtube.com/playlist?list=WL',
2728 'only_matching': True,
2729 }, {
2730 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2731 'only_matching': True,
2732 }]
2733
2734 def _real_extract(self, url):
2735 _, video = self._check_download_just_video(url, 'WL')
2736 if video:
2737 return video
2738 _, playlist = self._extract_playlist('WL')
2739 return playlist
2740
2741
2742 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2743 IE_NAME = 'youtube:favorites'
2744 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2745 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2746 _LOGIN_REQUIRED = True
2747
2748 def _real_extract(self, url):
2749 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2750 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2751 return self.url_result(playlist_id, 'YoutubePlaylist')
2752
2753
2754 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2755 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2756 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2757 _FEED_NAME = 'recommended'
2758 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2759
2760
2761 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2762 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2763 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2764 _FEED_NAME = 'subscriptions'
2765 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2766
2767
2768 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2769 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2770 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2771 _FEED_NAME = 'history'
2772 _PLAYLIST_TITLE = 'Youtube History'
2773
2774
2775 class YoutubeTruncatedURLIE(InfoExtractor):
2776 IE_NAME = 'youtube:truncated_url'
2777 IE_DESC = False # Do not list
2778 _VALID_URL = r'''(?x)
2779 (?:https?://)?
2780 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2781 (?:watch\?(?:
2782 feature=[a-z_]+|
2783 annotation_id=annotation_[^&]+|
2784 x-yt-cl=[0-9]+|
2785 hl=[^&]*|
2786 t=[0-9]+
2787 )?
2788 |
2789 attribution_link\?a=[^&]+
2790 )
2791 $
2792 '''
2793
2794 _TESTS = [{
2795 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2796 'only_matching': True,
2797 }, {
2798 'url': 'https://www.youtube.com/watch?',
2799 'only_matching': True,
2800 }, {
2801 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2802 'only_matching': True,
2803 }, {
2804 'url': 'https://www.youtube.com/watch?feature=foo',
2805 'only_matching': True,
2806 }, {
2807 'url': 'https://www.youtube.com/watch?hl=en-GB',
2808 'only_matching': True,
2809 }, {
2810 'url': 'https://www.youtube.com/watch?t=2372',
2811 'only_matching': True,
2812 }]
2813
2814 def _real_extract(self, url):
2815 raise ExtractorError(
2816 'Did you forget to quote the URL? Remember that & is a meta '
2817 'character in most shells, so you want to put the URL in quotes, '
2818 'like youtube-dl '
2819 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2820 ' or simply youtube-dl BaW_jenozKc .',
2821 expected=True)
2822
2823
2824 class YoutubeTruncatedIDIE(InfoExtractor):
2825 IE_NAME = 'youtube:truncated_id'
2826 IE_DESC = False # Do not list
2827 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2828
2829 _TESTS = [{
2830 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2831 'only_matching': True,
2832 }]
2833
2834 def _real_extract(self, url):
2835 video_id = self._match_id(url)
2836 raise ExtractorError(
2837 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2838 expected=True)
Backup clone of https://github.com/ytdl-org unavailable since RIAA fucking raid...