]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/youtube.py
0919bef0e06ae1b839a7debc38bdf457dafae192
[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\.)?hooktube\.com/|
336 (?:www\.)?yourepeat\.com/|
337 tube\.majestyc\.net/|
338 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
339 (?:.*?\#/)? # handle anchor (#/) redirect urls
340 (?: # the various things that can precede the ID:
341 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
342 |(?: # or the v= param in all its forms
343 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
344 (?:\?|\#!?) # the params delimiter ? or # or #!
345 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
346 v=
347 )
348 ))
349 |(?:
350 youtu\.be| # just youtu.be/xxxx
351 vid\.plus| # or vid.plus/xxxx
352 zwearz\.com/watch| # or zwearz.com/watch/xxxx
353 )/
354 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
355 )
356 )? # all until now is optional -> you can pass the naked ID
357 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
358 (?!.*?\blist=
359 (?:
360 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
361 WL # WL are handled by the watch later IE
362 )
363 )
364 (?(1).+)? # if we found the ID, everything can follow
365 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
366 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
367 _formats = {
368 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
369 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
370 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
371 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
372 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
373 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
374 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
375 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
376 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
377 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
378 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
379 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
380 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
381 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
382 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
383 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
384 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
385 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
386
387
388 # 3D videos
389 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
390 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
391 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
392 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
393 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
394 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
395 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
396
397 # Apple HTTP Live Streaming
398 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
399 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
400 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
401 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
402 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
403 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
404 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
405 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
406
407 # DASH mp4 video
408 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
409 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
410 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
411 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
412 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
413 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
414 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
415 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
416 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
417 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
418 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
419 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
420
421 # Dash mp4 audio
422 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
423 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
424 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
425 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
426 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
427 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
428 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
429
430 # Dash webm
431 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
432 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
433 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
434 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
435 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
436 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
437 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
438 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
439 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
440 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
441 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
442 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
443 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
444 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
445 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
446 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
447 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
448 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
449 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
450 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
451 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
452 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
453
454 # Dash webm audio
455 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
456 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
457
458 # Dash webm audio with opus inside
459 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
460 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
461 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
462
463 # RTMP (unnamed)
464 '_rtmp': {'protocol': 'rtmp'},
465 }
466 _SUBTITLE_FORMATS = ('ttml', 'vtt')
467
468 _GEO_BYPASS = False
469
470 IE_NAME = 'youtube'
471 _TESTS = [
472 {
473 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
474 'info_dict': {
475 'id': 'BaW_jenozKc',
476 'ext': 'mp4',
477 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
478 'uploader': 'Philipp Hagemeister',
479 'uploader_id': 'phihag',
480 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
481 'upload_date': '20121002',
482 'license': 'Standard YouTube License',
483 '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 .',
484 'categories': ['Science & Technology'],
485 'tags': ['youtube-dl'],
486 'duration': 10,
487 'like_count': int,
488 'dislike_count': int,
489 'start_time': 1,
490 'end_time': 9,
491 }
492 },
493 {
494 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
495 'note': 'Test generic use_cipher_signature video (#897)',
496 'info_dict': {
497 'id': 'UxxajLWwzqY',
498 'ext': 'mp4',
499 'upload_date': '20120506',
500 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
501 'alt_title': 'I Love It (feat. Charli XCX)',
502 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
503 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
504 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
505 'iconic ep', 'iconic', 'love', 'it'],
506 'duration': 180,
507 'uploader': 'Icona Pop',
508 'uploader_id': 'IconaPop',
509 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
510 'license': 'Standard YouTube License',
511 'creator': 'Icona Pop',
512 }
513 },
514 {
515 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
516 'note': 'Test VEVO video with age protection (#956)',
517 'info_dict': {
518 'id': '07FYdnEawAQ',
519 'ext': 'mp4',
520 'upload_date': '20130703',
521 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
522 'alt_title': 'Tunnel Vision',
523 'description': 'md5:64249768eec3bc4276236606ea996373',
524 'duration': 419,
525 'uploader': 'justintimberlakeVEVO',
526 'uploader_id': 'justintimberlakeVEVO',
527 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
528 'license': 'Standard YouTube License',
529 'creator': 'Justin Timberlake',
530 'age_limit': 18,
531 }
532 },
533 {
534 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
535 'note': 'Embed-only video (#1746)',
536 'info_dict': {
537 'id': 'yZIXLfi8CZQ',
538 'ext': 'mp4',
539 'upload_date': '20120608',
540 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
541 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
542 'uploader': 'SET India',
543 'uploader_id': 'setindia',
544 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
545 'license': 'Standard YouTube License',
546 'age_limit': 18,
547 }
548 },
549 {
550 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
551 'note': 'Use the first video ID in the URL',
552 'info_dict': {
553 'id': 'BaW_jenozKc',
554 'ext': 'mp4',
555 'title': 'youtube-dl test video "\'/\\Ƥā†­š•',
556 'uploader': 'Philipp Hagemeister',
557 'uploader_id': 'phihag',
558 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
559 'upload_date': '20121002',
560 'license': 'Standard YouTube License',
561 '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 .',
562 'categories': ['Science & Technology'],
563 'tags': ['youtube-dl'],
564 'duration': 10,
565 'like_count': int,
566 'dislike_count': int,
567 },
568 'params': {
569 'skip_download': True,
570 },
571 },
572 {
573 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
574 'note': '256k DASH audio (format 141) via DASH manifest',
575 'info_dict': {
576 'id': 'a9LDPn-MO4I',
577 'ext': 'm4a',
578 'upload_date': '20121002',
579 'uploader_id': '8KVIDEO',
580 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
581 'description': '',
582 'uploader': '8KVIDEO',
583 'license': 'Standard YouTube License',
584 'title': 'UHDTV TEST 8K VIDEO.mp4'
585 },
586 'params': {
587 'youtube_include_dash_manifest': True,
588 'format': '141',
589 },
590 'skip': 'format 141 not served anymore',
591 },
592 # DASH manifest with encrypted signature
593 {
594 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
595 'info_dict': {
596 'id': 'IB3lcPjvWLA',
597 'ext': 'm4a',
598 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
599 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
600 'duration': 244,
601 'uploader': 'AfrojackVEVO',
602 'uploader_id': 'AfrojackVEVO',
603 'upload_date': '20131011',
604 'license': 'Standard YouTube License',
605 },
606 'params': {
607 'youtube_include_dash_manifest': True,
608 'format': '141/bestaudio[ext=m4a]',
609 },
610 },
611 # JS player signature function name containing $
612 {
613 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
614 'info_dict': {
615 'id': 'nfWlot6h_JM',
616 'ext': 'm4a',
617 'title': 'Taylor Swift - Shake It Off',
618 'alt_title': 'Shake It Off',
619 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
620 'duration': 242,
621 'uploader': 'TaylorSwiftVEVO',
622 'uploader_id': 'TaylorSwiftVEVO',
623 'upload_date': '20140818',
624 'license': 'Standard YouTube License',
625 'creator': 'Taylor Swift',
626 },
627 'params': {
628 'youtube_include_dash_manifest': True,
629 'format': '141/bestaudio[ext=m4a]',
630 },
631 },
632 # Controversy video
633 {
634 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
635 'info_dict': {
636 'id': 'T4XJQO3qol8',
637 'ext': 'mp4',
638 'duration': 219,
639 'upload_date': '20100909',
640 'uploader': 'The Amazing Atheist',
641 'uploader_id': 'TheAmazingAtheist',
642 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
643 'license': 'Standard YouTube License',
644 'title': 'Burning Everyone\'s Koran',
645 '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',
646 }
647 },
648 # Normal age-gate video (No vevo, embed allowed)
649 {
650 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
651 'info_dict': {
652 'id': 'HtVdAasjOgU',
653 'ext': 'mp4',
654 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
655 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
656 'duration': 142,
657 'uploader': 'The Witcher',
658 'uploader_id': 'WitcherGame',
659 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
660 'upload_date': '20140605',
661 'license': 'Standard YouTube License',
662 'age_limit': 18,
663 },
664 },
665 # Age-gate video with encrypted signature
666 {
667 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
668 'info_dict': {
669 'id': '6kLq3WMV1nU',
670 'ext': 'mp4',
671 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
672 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
673 'duration': 247,
674 'uploader': 'LloydVEVO',
675 'uploader_id': 'LloydVEVO',
676 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
677 'upload_date': '20110629',
678 'license': 'Standard YouTube License',
679 'age_limit': 18,
680 },
681 },
682 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
683 # YouTube Red ad is not captured for creator
684 {
685 'url': '__2ABJjxzNo',
686 'info_dict': {
687 'id': '__2ABJjxzNo',
688 'ext': 'mp4',
689 'duration': 266,
690 'upload_date': '20100430',
691 'uploader_id': 'deadmau5',
692 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
693 'creator': 'deadmau5',
694 'description': 'md5:12c56784b8032162bb936a5f76d55360',
695 'uploader': 'deadmau5',
696 'license': 'Standard YouTube License',
697 'title': 'Deadmau5 - Some Chords (HD)',
698 'alt_title': 'Some Chords',
699 },
700 'expected_warnings': [
701 'DASH manifest missing',
702 ]
703 },
704 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
705 {
706 'url': 'lqQg6PlCWgI',
707 'info_dict': {
708 'id': 'lqQg6PlCWgI',
709 'ext': 'mp4',
710 'duration': 6085,
711 'upload_date': '20150827',
712 'uploader_id': 'olympic',
713 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
714 'license': 'Standard YouTube License',
715 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
716 'uploader': 'Olympic',
717 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
718 },
719 'params': {
720 'skip_download': 'requires avconv',
721 }
722 },
723 # Non-square pixels
724 {
725 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
726 'info_dict': {
727 'id': '_b-2C3KPAM0',
728 'ext': 'mp4',
729 'stretched_ratio': 16 / 9.,
730 'duration': 85,
731 'upload_date': '20110310',
732 'uploader_id': 'AllenMeow',
733 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
734 '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',
735 'uploader': 'å­«č‰¾å€«',
736 'license': 'Standard YouTube License',
737 'title': '[A-made] č®Šę…‹å¦å­—å¹•ē‰ˆ å¤Ŗ妍 ęˆ‘å°±ę˜Æ這ęØ£ēš„äŗŗ',
738 },
739 },
740 # url_encoded_fmt_stream_map is empty string
741 {
742 'url': 'qEJwOuvDf7I',
743 'info_dict': {
744 'id': 'qEJwOuvDf7I',
745 'ext': 'webm',
746 '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',
747 'description': '',
748 'upload_date': '20150404',
749 'uploader_id': 'spbelect',
750 'uploader': 'ŠŠ°Š±Š»ŃŽŠ“Š°Ń‚ŠµŠ»Šø ŠŸŠµŃ‚ŠµŃ€Š±ŃƒŃ€Š³Š°',
751 },
752 'params': {
753 'skip_download': 'requires avconv',
754 },
755 'skip': 'This live event has ended.',
756 },
757 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
758 {
759 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
760 'info_dict': {
761 'id': 'FIl7x6_3R5Y',
762 'ext': 'mp4',
763 'title': 'md5:7b81415841e02ecd4313668cde88737a',
764 'description': 'md5:116377fd2963b81ec4ce64b542173306',
765 'duration': 220,
766 'upload_date': '20150625',
767 'uploader_id': 'dorappi2000',
768 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
769 'uploader': 'dorappi2000',
770 'license': 'Standard YouTube License',
771 'formats': 'mincount:32',
772 },
773 },
774 # DASH manifest with segment_list
775 {
776 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
777 'md5': '8ce563a1d667b599d21064e982ab9e31',
778 'info_dict': {
779 'id': 'CsmdDsKjzN8',
780 'ext': 'mp4',
781 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
782 'uploader': 'Airtek',
783 'description': 'RetransmisiĆ³n en directo de la XVIII media maratĆ³n de Zaragoza.',
784 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
785 'license': 'Standard YouTube License',
786 'title': 'RetransmisiĆ³n XVIII Media maratĆ³n Zaragoza 2015',
787 },
788 'params': {
789 'youtube_include_dash_manifest': True,
790 'format': '135', # bestvideo
791 },
792 'skip': 'This live event has ended.',
793 },
794 {
795 # Multifeed videos (multiple cameras), URL is for Main Camera
796 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
797 'info_dict': {
798 'id': 'jqWvoWXjCVs',
799 'title': 'teamPGP: Rocket League Noob Stream',
800 'description': 'md5:dc7872fb300e143831327f1bae3af010',
801 },
802 'playlist': [{
803 'info_dict': {
804 'id': 'jqWvoWXjCVs',
805 'ext': 'mp4',
806 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
807 'description': 'md5:dc7872fb300e143831327f1bae3af010',
808 'duration': 7335,
809 'upload_date': '20150721',
810 'uploader': 'Beer Games Beer',
811 'uploader_id': 'beergamesbeer',
812 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
813 'license': 'Standard YouTube License',
814 },
815 }, {
816 'info_dict': {
817 'id': '6h8e8xoXJzg',
818 'ext': 'mp4',
819 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
820 'description': 'md5:dc7872fb300e143831327f1bae3af010',
821 'duration': 7337,
822 'upload_date': '20150721',
823 'uploader': 'Beer Games Beer',
824 'uploader_id': 'beergamesbeer',
825 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
826 'license': 'Standard YouTube License',
827 },
828 }, {
829 'info_dict': {
830 'id': 'PUOgX5z9xZw',
831 'ext': 'mp4',
832 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
833 'description': 'md5:dc7872fb300e143831327f1bae3af010',
834 'duration': 7337,
835 'upload_date': '20150721',
836 'uploader': 'Beer Games Beer',
837 'uploader_id': 'beergamesbeer',
838 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
839 'license': 'Standard YouTube License',
840 },
841 }, {
842 'info_dict': {
843 'id': 'teuwxikvS5k',
844 'ext': 'mp4',
845 'title': 'teamPGP: Rocket League Noob Stream (zim)',
846 'description': 'md5:dc7872fb300e143831327f1bae3af010',
847 'duration': 7334,
848 'upload_date': '20150721',
849 'uploader': 'Beer Games Beer',
850 'uploader_id': 'beergamesbeer',
851 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
852 'license': 'Standard YouTube License',
853 },
854 }],
855 'params': {
856 'skip_download': True,
857 },
858 },
859 {
860 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
861 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
862 'info_dict': {
863 'id': 'gVfLd0zydlo',
864 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
865 },
866 'playlist_count': 2,
867 'skip': 'Not multifeed anymore',
868 },
869 {
870 'url': 'https://vid.plus/FlRa-iH7PGw',
871 'only_matching': True,
872 },
873 {
874 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
875 'only_matching': True,
876 },
877 {
878 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
879 # Also tests cut-off URL expansion in video description (see
880 # https://github.com/rg3/youtube-dl/issues/1892,
881 # https://github.com/rg3/youtube-dl/issues/8164)
882 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
883 'info_dict': {
884 'id': 'lsguqyKfVQg',
885 'ext': 'mp4',
886 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
887 'alt_title': 'Dark Walk',
888 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
889 'duration': 133,
890 'upload_date': '20151119',
891 'uploader_id': 'IronSoulElf',
892 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
893 'uploader': 'IronSoulElf',
894 'license': 'Standard YouTube License',
895 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
896 },
897 'params': {
898 'skip_download': True,
899 },
900 },
901 {
902 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
903 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
904 'only_matching': True,
905 },
906 {
907 # Video with yt:stretch=17:0
908 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
909 'info_dict': {
910 'id': 'Q39EVAstoRM',
911 'ext': 'mp4',
912 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
913 'description': 'md5:ee18a25c350637c8faff806845bddee9',
914 'upload_date': '20151107',
915 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
916 'uploader': 'CH GAMER DROID',
917 },
918 'params': {
919 'skip_download': True,
920 },
921 'skip': 'This video does not exist.',
922 },
923 {
924 # Video licensed under Creative Commons
925 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
926 'info_dict': {
927 'id': 'M4gD1WSo5mA',
928 'ext': 'mp4',
929 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
930 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
931 'duration': 721,
932 'upload_date': '20150127',
933 'uploader_id': 'BerkmanCenter',
934 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
935 'uploader': 'The Berkman Klein Center for Internet & Society',
936 'license': 'Creative Commons Attribution license (reuse allowed)',
937 },
938 'params': {
939 'skip_download': True,
940 },
941 },
942 {
943 # Channel-like uploader_url
944 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
945 'info_dict': {
946 'id': 'eQcmzGIKrzg',
947 'ext': 'mp4',
948 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
949 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
950 'duration': 4060,
951 'upload_date': '20151119',
952 'uploader': 'Bernie 2016',
953 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
954 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
955 'license': 'Creative Commons Attribution license (reuse allowed)',
956 },
957 'params': {
958 'skip_download': True,
959 },
960 },
961 {
962 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
963 'only_matching': True,
964 },
965 {
966 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
967 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
968 'only_matching': True,
969 },
970 {
971 # Rental video preview
972 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
973 'info_dict': {
974 'id': 'uGpuVWrhIzE',
975 'ext': 'mp4',
976 'title': 'Piku - Trailer',
977 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
978 'upload_date': '20150811',
979 'uploader': 'FlixMatrix',
980 'uploader_id': 'FlixMatrixKaravan',
981 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
982 'license': 'Standard YouTube License',
983 },
984 'params': {
985 'skip_download': True,
986 },
987 },
988 {
989 # YouTube Red video with episode data
990 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
991 'info_dict': {
992 'id': 'iqKdEhx-dD4',
993 'ext': 'mp4',
994 'title': 'Isolation - Mind Field (Ep 1)',
995 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
996 'duration': 2085,
997 'upload_date': '20170118',
998 'uploader': 'Vsauce',
999 'uploader_id': 'Vsauce',
1000 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1001 'license': 'Standard YouTube License',
1002 'series': 'Mind Field',
1003 'season_number': 1,
1004 'episode_number': 1,
1005 },
1006 'params': {
1007 'skip_download': True,
1008 },
1009 'expected_warnings': [
1010 'Skipping DASH manifest',
1011 ],
1012 },
1013 {
1014 # The following content has been identified by the YouTube community
1015 # as inappropriate or offensive to some audiences.
1016 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1017 'info_dict': {
1018 'id': '6SJNVb0GnPI',
1019 'ext': 'mp4',
1020 'title': 'Race Differences in Intelligence',
1021 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1022 'duration': 965,
1023 'upload_date': '20140124',
1024 'uploader': 'New Century Foundation',
1025 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1026 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1027 'license': 'Standard YouTube License',
1028 'view_count': int,
1029 },
1030 'params': {
1031 'skip_download': True,
1032 },
1033 },
1034 {
1035 # itag 212
1036 'url': '1t24XAntNCY',
1037 'only_matching': True,
1038 },
1039 {
1040 # geo restricted to JP
1041 'url': 'sJL6WA-aGkQ',
1042 'only_matching': True,
1043 },
1044 {
1045 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1046 'only_matching': True,
1047 },
1048 ]
1049
1050 def __init__(self, *args, **kwargs):
1051 super(YoutubeIE, self).__init__(*args, **kwargs)
1052 self._player_cache = {}
1053
1054 def report_video_info_webpage_download(self, video_id):
1055 """Report attempt to download video info webpage."""
1056 self.to_screen('%s: Downloading video info webpage' % video_id)
1057
1058 def report_information_extraction(self, video_id):
1059 """Report attempt to extract video information."""
1060 self.to_screen('%s: Extracting video information' % video_id)
1061
1062 def report_unavailable_format(self, video_id, format):
1063 """Report extracted video URL."""
1064 self.to_screen('%s: Format %s not available' % (video_id, format))
1065
1066 def report_rtmp_download(self):
1067 """Indicate the download will use the RTMP protocol."""
1068 self.to_screen('RTMP download detected')
1069
1070 def _signature_cache_id(self, example_sig):
1071 """ Return a string representation of a signature """
1072 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1073
1074 def _extract_signature_function(self, video_id, player_url, example_sig):
1075 id_m = re.match(
1076 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1077 player_url)
1078 if not id_m:
1079 raise ExtractorError('Cannot identify player %r' % player_url)
1080 player_type = id_m.group('ext')
1081 player_id = id_m.group('id')
1082
1083 # Read from filesystem cache
1084 func_id = '%s_%s_%s' % (
1085 player_type, player_id, self._signature_cache_id(example_sig))
1086 assert os.path.basename(func_id) == func_id
1087
1088 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1089 if cache_spec is not None:
1090 return lambda s: ''.join(s[i] for i in cache_spec)
1091
1092 download_note = (
1093 'Downloading player %s' % player_url
1094 if self._downloader.params.get('verbose') else
1095 'Downloading %s player %s' % (player_type, player_id)
1096 )
1097 if player_type == 'js':
1098 code = self._download_webpage(
1099 player_url, video_id,
1100 note=download_note,
1101 errnote='Download of %s failed' % player_url)
1102 res = self._parse_sig_js(code)
1103 elif player_type == 'swf':
1104 urlh = self._request_webpage(
1105 player_url, video_id,
1106 note=download_note,
1107 errnote='Download of %s failed' % player_url)
1108 code = urlh.read()
1109 res = self._parse_sig_swf(code)
1110 else:
1111 assert False, 'Invalid player type %r' % player_type
1112
1113 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1114 cache_res = res(test_string)
1115 cache_spec = [ord(c) for c in cache_res]
1116
1117 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1118 return res
1119
1120 def _print_sig_code(self, func, example_sig):
1121 def gen_sig_code(idxs):
1122 def _genslice(start, end, step):
1123 starts = '' if start == 0 else str(start)
1124 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1125 steps = '' if step == 1 else (':%d' % step)
1126 return 's[%s%s%s]' % (starts, ends, steps)
1127
1128 step = None
1129 # Quelch pyflakes warnings - start will be set when step is set
1130 start = '(Never used)'
1131 for i, prev in zip(idxs[1:], idxs[:-1]):
1132 if step is not None:
1133 if i - prev == step:
1134 continue
1135 yield _genslice(start, prev, step)
1136 step = None
1137 continue
1138 if i - prev in [-1, 1]:
1139 step = i - prev
1140 start = prev
1141 continue
1142 else:
1143 yield 's[%d]' % prev
1144 if step is None:
1145 yield 's[%d]' % i
1146 else:
1147 yield _genslice(start, i, step)
1148
1149 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1150 cache_res = func(test_string)
1151 cache_spec = [ord(c) for c in cache_res]
1152 expr_code = ' + '.join(gen_sig_code(cache_spec))
1153 signature_id_tuple = '(%s)' % (
1154 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1155 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1156 ' return %s\n') % (signature_id_tuple, expr_code)
1157 self.to_screen('Extracted signature function:\n' + code)
1158
1159 def _parse_sig_js(self, jscode):
1160 funcname = self._search_regex(
1161 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1162 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1163 jscode, 'Initial JS player signature function name', group='sig')
1164
1165 jsi = JSInterpreter(jscode)
1166 initial_function = jsi.extract_function(funcname)
1167 return lambda s: initial_function([s])
1168
1169 def _parse_sig_swf(self, file_contents):
1170 swfi = SWFInterpreter(file_contents)
1171 TARGET_CLASSNAME = 'SignatureDecipher'
1172 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1173 initial_function = swfi.extract_function(searched_class, 'decipher')
1174 return lambda s: initial_function([s])
1175
1176 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1177 """Turn the encrypted s field into a working signature"""
1178
1179 if player_url is None:
1180 raise ExtractorError('Cannot decrypt signature without player_url')
1181
1182 if player_url.startswith('//'):
1183 player_url = 'https:' + player_url
1184 elif not re.match(r'https?://', player_url):
1185 player_url = compat_urlparse.urljoin(
1186 'https://www.youtube.com', player_url)
1187 try:
1188 player_id = (player_url, self._signature_cache_id(s))
1189 if player_id not in self._player_cache:
1190 func = self._extract_signature_function(
1191 video_id, player_url, s
1192 )
1193 self._player_cache[player_id] = func
1194 func = self._player_cache[player_id]
1195 if self._downloader.params.get('youtube_print_sig_code'):
1196 self._print_sig_code(func, s)
1197 return func(s)
1198 except Exception as e:
1199 tb = traceback.format_exc()
1200 raise ExtractorError(
1201 'Signature extraction failed: ' + tb, cause=e)
1202
1203 def _get_subtitles(self, video_id, webpage):
1204 try:
1205 subs_doc = self._download_xml(
1206 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1207 video_id, note=False)
1208 except ExtractorError as err:
1209 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1210 return {}
1211
1212 sub_lang_list = {}
1213 for track in subs_doc.findall('track'):
1214 lang = track.attrib['lang_code']
1215 if lang in sub_lang_list:
1216 continue
1217 sub_formats = []
1218 for ext in self._SUBTITLE_FORMATS:
1219 params = compat_urllib_parse_urlencode({
1220 'lang': lang,
1221 'v': video_id,
1222 'fmt': ext,
1223 'name': track.attrib['name'].encode('utf-8'),
1224 })
1225 sub_formats.append({
1226 'url': 'https://www.youtube.com/api/timedtext?' + params,
1227 'ext': ext,
1228 })
1229 sub_lang_list[lang] = sub_formats
1230 if not sub_lang_list:
1231 self._downloader.report_warning('video doesn\'t have subtitles')
1232 return {}
1233 return sub_lang_list
1234
1235 def _get_ytplayer_config(self, video_id, webpage):
1236 patterns = (
1237 # User data may contain arbitrary character sequences that may affect
1238 # JSON extraction with regex, e.g. when '};' is contained the second
1239 # regex won't capture the whole JSON. Yet working around by trying more
1240 # concrete regex first keeping in mind proper quoted string handling
1241 # to be implemented in future that will replace this workaround (see
1242 # https://github.com/rg3/youtube-dl/issues/7468,
1243 # https://github.com/rg3/youtube-dl/pull/7599)
1244 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1245 r';ytplayer\.config\s*=\s*({.+?});',
1246 )
1247 config = self._search_regex(
1248 patterns, webpage, 'ytplayer.config', default=None)
1249 if config:
1250 return self._parse_json(
1251 uppercase_escape(config), video_id, fatal=False)
1252
1253 def _get_automatic_captions(self, video_id, webpage):
1254 """We need the webpage for getting the captions url, pass it as an
1255 argument to speed up the process."""
1256 self.to_screen('%s: Looking for automatic captions' % video_id)
1257 player_config = self._get_ytplayer_config(video_id, webpage)
1258 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1259 if not player_config:
1260 self._downloader.report_warning(err_msg)
1261 return {}
1262 try:
1263 args = player_config['args']
1264 caption_url = args.get('ttsurl')
1265 if caption_url:
1266 timestamp = args['timestamp']
1267 # We get the available subtitles
1268 list_params = compat_urllib_parse_urlencode({
1269 'type': 'list',
1270 'tlangs': 1,
1271 'asrs': 1,
1272 })
1273 list_url = caption_url + '&' + list_params
1274 caption_list = self._download_xml(list_url, video_id)
1275 original_lang_node = caption_list.find('track')
1276 if original_lang_node is None:
1277 self._downloader.report_warning('Video doesn\'t have automatic captions')
1278 return {}
1279 original_lang = original_lang_node.attrib['lang_code']
1280 caption_kind = original_lang_node.attrib.get('kind', '')
1281
1282 sub_lang_list = {}
1283 for lang_node in caption_list.findall('target'):
1284 sub_lang = lang_node.attrib['lang_code']
1285 sub_formats = []
1286 for ext in self._SUBTITLE_FORMATS:
1287 params = compat_urllib_parse_urlencode({
1288 'lang': original_lang,
1289 'tlang': sub_lang,
1290 'fmt': ext,
1291 'ts': timestamp,
1292 'kind': caption_kind,
1293 })
1294 sub_formats.append({
1295 'url': caption_url + '&' + params,
1296 'ext': ext,
1297 })
1298 sub_lang_list[sub_lang] = sub_formats
1299 return sub_lang_list
1300
1301 def make_captions(sub_url, sub_langs):
1302 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1303 caption_qs = compat_parse_qs(parsed_sub_url.query)
1304 captions = {}
1305 for sub_lang in sub_langs:
1306 sub_formats = []
1307 for ext in self._SUBTITLE_FORMATS:
1308 caption_qs.update({
1309 'tlang': [sub_lang],
1310 'fmt': [ext],
1311 })
1312 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1313 query=compat_urllib_parse_urlencode(caption_qs, True)))
1314 sub_formats.append({
1315 'url': sub_url,
1316 'ext': ext,
1317 })
1318 captions[sub_lang] = sub_formats
1319 return captions
1320
1321 # New captions format as of 22.06.2017
1322 player_response = args.get('player_response')
1323 if player_response and isinstance(player_response, compat_str):
1324 player_response = self._parse_json(
1325 player_response, video_id, fatal=False)
1326 if player_response:
1327 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1328 base_url = renderer['captionTracks'][0]['baseUrl']
1329 sub_lang_list = []
1330 for lang in renderer['translationLanguages']:
1331 lang_code = lang.get('languageCode')
1332 if lang_code:
1333 sub_lang_list.append(lang_code)
1334 return make_captions(base_url, sub_lang_list)
1335
1336 # Some videos don't provide ttsurl but rather caption_tracks and
1337 # caption_translation_languages (e.g. 20LmZk1hakA)
1338 # Does not used anymore as of 22.06.2017
1339 caption_tracks = args['caption_tracks']
1340 caption_translation_languages = args['caption_translation_languages']
1341 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1342 sub_lang_list = []
1343 for lang in caption_translation_languages.split(','):
1344 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1345 sub_lang = lang_qs.get('lc', [None])[0]
1346 if sub_lang:
1347 sub_lang_list.append(sub_lang)
1348 return make_captions(caption_url, sub_lang_list)
1349 # An extractor error can be raise by the download process if there are
1350 # no automatic captions but there are subtitles
1351 except (KeyError, IndexError, ExtractorError):
1352 self._downloader.report_warning(err_msg)
1353 return {}
1354
1355 def _mark_watched(self, video_id, video_info):
1356 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1357 if not playback_url:
1358 return
1359 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1360 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1361
1362 # cpn generation algorithm is reverse engineered from base.js.
1363 # In fact it works even with dummy cpn.
1364 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1365 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1366
1367 qs.update({
1368 'ver': ['2'],
1369 'cpn': [cpn],
1370 })
1371 playback_url = compat_urlparse.urlunparse(
1372 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1373
1374 self._download_webpage(
1375 playback_url, video_id, 'Marking watched',
1376 'Unable to mark watched', fatal=False)
1377
1378 @staticmethod
1379 def _extract_urls(webpage):
1380 # Embedded YouTube player
1381 entries = [
1382 unescapeHTML(mobj.group('url'))
1383 for mobj in re.finditer(r'''(?x)
1384 (?:
1385 <iframe[^>]+?src=|
1386 data-video-url=|
1387 <embed[^>]+?src=|
1388 embedSWF\(?:\s*|
1389 <object[^>]+data=|
1390 new\s+SWFObject\(
1391 )
1392 (["\'])
1393 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1394 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1395 \1''', webpage)]
1396
1397 # lazyYT YouTube embed
1398 entries.extend(list(map(
1399 unescapeHTML,
1400 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1401
1402 # Wordpress "YouTube Video Importer" plugin
1403 matches = re.findall(r'''(?x)<div[^>]+
1404 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1405 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1406 entries.extend(m[-1] for m in matches)
1407
1408 return entries
1409
1410 @staticmethod
1411 def _extract_url(webpage):
1412 urls = YoutubeIE._extract_urls(webpage)
1413 return urls[0] if urls else None
1414
1415 @classmethod
1416 def extract_id(cls, url):
1417 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1418 if mobj is None:
1419 raise ExtractorError('Invalid URL: %s' % url)
1420 video_id = mobj.group(2)
1421 return video_id
1422
1423 def _extract_annotations(self, video_id):
1424 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1425 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1426
1427 @staticmethod
1428 def _extract_chapters(description, duration):
1429 if not description:
1430 return None
1431 chapter_lines = re.findall(
1432 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1433 description)
1434 if not chapter_lines:
1435 return None
1436 chapters = []
1437 for next_num, (chapter_line, time_point) in enumerate(
1438 chapter_lines, start=1):
1439 start_time = parse_duration(time_point)
1440 if start_time is None:
1441 continue
1442 if start_time > duration:
1443 break
1444 end_time = (duration if next_num == len(chapter_lines)
1445 else parse_duration(chapter_lines[next_num][1]))
1446 if end_time is None:
1447 continue
1448 if end_time > duration:
1449 end_time = duration
1450 if start_time > end_time:
1451 break
1452 chapter_title = re.sub(
1453 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1454 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1455 chapters.append({
1456 'start_time': start_time,
1457 'end_time': end_time,
1458 'title': chapter_title,
1459 })
1460 return chapters
1461
1462 def _real_extract(self, url):
1463 url, smuggled_data = unsmuggle_url(url, {})
1464
1465 proto = (
1466 'http' if self._downloader.params.get('prefer_insecure', False)
1467 else 'https')
1468
1469 start_time = None
1470 end_time = None
1471 parsed_url = compat_urllib_parse_urlparse(url)
1472 for component in [parsed_url.fragment, parsed_url.query]:
1473 query = compat_parse_qs(component)
1474 if start_time is None and 't' in query:
1475 start_time = parse_duration(query['t'][0])
1476 if start_time is None and 'start' in query:
1477 start_time = parse_duration(query['start'][0])
1478 if end_time is None and 'end' in query:
1479 end_time = parse_duration(query['end'][0])
1480
1481 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1482 mobj = re.search(self._NEXT_URL_RE, url)
1483 if mobj:
1484 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1485 video_id = self.extract_id(url)
1486
1487 # Get video webpage
1488 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1489 video_webpage = self._download_webpage(url, video_id)
1490
1491 # Attempt to extract SWF player URL
1492 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1493 if mobj is not None:
1494 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1495 else:
1496 player_url = None
1497
1498 dash_mpds = []
1499
1500 def add_dash_mpd(video_info):
1501 dash_mpd = video_info.get('dashmpd')
1502 if dash_mpd and dash_mpd[0] not in dash_mpds:
1503 dash_mpds.append(dash_mpd[0])
1504
1505 is_live = None
1506 view_count = None
1507
1508 def extract_view_count(v_info):
1509 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1510
1511 # Get video info
1512 embed_webpage = None
1513 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1514 age_gate = True
1515 # We simulate the access to the video from www.youtube.com/v/{video_id}
1516 # this can be viewed without login into Youtube
1517 url = proto + '://www.youtube.com/embed/%s' % video_id
1518 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1519 data = compat_urllib_parse_urlencode({
1520 'video_id': video_id,
1521 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1522 'sts': self._search_regex(
1523 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1524 })
1525 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1526 video_info_webpage = self._download_webpage(
1527 video_info_url, video_id,
1528 note='Refetching age-gated info webpage',
1529 errnote='unable to download video info webpage')
1530 video_info = compat_parse_qs(video_info_webpage)
1531 add_dash_mpd(video_info)
1532 else:
1533 age_gate = False
1534 video_info = None
1535 sts = None
1536 # Try looking directly into the video webpage
1537 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1538 if ytplayer_config:
1539 args = ytplayer_config['args']
1540 if args.get('url_encoded_fmt_stream_map'):
1541 # Convert to the same format returned by compat_parse_qs
1542 video_info = dict((k, [v]) for k, v in args.items())
1543 add_dash_mpd(video_info)
1544 # Rental video is not rented but preview is available (e.g.
1545 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1546 # https://github.com/rg3/youtube-dl/issues/10532)
1547 if not video_info and args.get('ypc_vid'):
1548 return self.url_result(
1549 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1550 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1551 is_live = True
1552 sts = ytplayer_config.get('sts')
1553 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1554 # We also try looking in get_video_info since it may contain different dashmpd
1555 # URL that points to a DASH manifest with possibly different itag set (some itags
1556 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1557 # manifest pointed by get_video_info's dashmpd).
1558 # The general idea is to take a union of itags of both DASH manifests (for example
1559 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1560 self.report_video_info_webpage_download(video_id)
1561 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1562 query = {
1563 'video_id': video_id,
1564 'ps': 'default',
1565 'eurl': '',
1566 'gl': 'US',
1567 'hl': 'en',
1568 }
1569 if el:
1570 query['el'] = el
1571 if sts:
1572 query['sts'] = sts
1573 video_info_webpage = self._download_webpage(
1574 '%s://www.youtube.com/get_video_info' % proto,
1575 video_id, note=False,
1576 errnote='unable to download video info webpage',
1577 fatal=False, query=query)
1578 if not video_info_webpage:
1579 continue
1580 get_video_info = compat_parse_qs(video_info_webpage)
1581 add_dash_mpd(get_video_info)
1582 if view_count is None:
1583 view_count = extract_view_count(get_video_info)
1584 if not video_info:
1585 video_info = get_video_info
1586 if 'token' in get_video_info:
1587 # Different get_video_info requests may report different results, e.g.
1588 # some may report video unavailability, but some may serve it without
1589 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1590 # the original webpage as well as el=info and el=embedded get_video_info
1591 # requests report video unavailability due to geo restriction while
1592 # el=detailpage succeeds and returns valid data). This is probably
1593 # due to YouTube measures against IP ranges of hosting providers.
1594 # Working around by preferring the first succeeded video_info containing
1595 # the token if no such video_info yet was found.
1596 if 'token' not in video_info:
1597 video_info = get_video_info
1598 break
1599 if 'token' not in video_info:
1600 if 'reason' in video_info:
1601 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1602 regions_allowed = self._html_search_meta(
1603 'regionsAllowed', video_webpage, default=None)
1604 countries = regions_allowed.split(',') if regions_allowed else None
1605 self.raise_geo_restricted(
1606 msg=video_info['reason'][0], countries=countries)
1607 raise ExtractorError(
1608 'YouTube said: %s' % video_info['reason'][0],
1609 expected=True, video_id=video_id)
1610 else:
1611 raise ExtractorError(
1612 '"token" parameter not in video info for unknown reason',
1613 video_id=video_id)
1614
1615 # title
1616 if 'title' in video_info:
1617 video_title = video_info['title'][0]
1618 else:
1619 self._downloader.report_warning('Unable to extract video title')
1620 video_title = '_'
1621
1622 # description
1623 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1624 if video_description:
1625
1626 def replace_url(m):
1627 redir_url = compat_urlparse.urljoin(url, m.group(1))
1628 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1629 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1630 qs = compat_parse_qs(parsed_redir_url.query)
1631 q = qs.get('q')
1632 if q and q[0]:
1633 return q[0]
1634 return redir_url
1635
1636 description_original = video_description = re.sub(r'''(?x)
1637 <a\s+
1638 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1639 (?:title|href)="([^"]+)"\s+
1640 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1641 class="[^"]*"[^>]*>
1642 [^<]+\.{3}\s*
1643 </a>
1644 ''', replace_url, video_description)
1645 video_description = clean_html(video_description)
1646 else:
1647 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1648 if fd_mobj:
1649 video_description = unescapeHTML(fd_mobj.group(1))
1650 else:
1651 video_description = ''
1652
1653 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1654 if not self._downloader.params.get('noplaylist'):
1655 entries = []
1656 feed_ids = []
1657 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1658 for feed in multifeed_metadata_list.split(','):
1659 # Unquote should take place before split on comma (,) since textual
1660 # fields may contain comma as well (see
1661 # https://github.com/rg3/youtube-dl/issues/8536)
1662 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1663 entries.append({
1664 '_type': 'url_transparent',
1665 'ie_key': 'Youtube',
1666 'url': smuggle_url(
1667 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1668 {'force_singlefeed': True}),
1669 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1670 })
1671 feed_ids.append(feed_data['id'][0])
1672 self.to_screen(
1673 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1674 % (', '.join(feed_ids), video_id))
1675 return self.playlist_result(entries, video_id, video_title, video_description)
1676 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1677
1678 if view_count is None:
1679 view_count = extract_view_count(video_info)
1680
1681 # Check for "rental" videos
1682 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1683 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1684
1685 # Start extracting information
1686 self.report_information_extraction(video_id)
1687
1688 # uploader
1689 if 'author' not in video_info:
1690 raise ExtractorError('Unable to extract uploader name')
1691 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1692
1693 # uploader_id
1694 video_uploader_id = None
1695 video_uploader_url = None
1696 mobj = re.search(
1697 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1698 video_webpage)
1699 if mobj is not None:
1700 video_uploader_id = mobj.group('uploader_id')
1701 video_uploader_url = mobj.group('uploader_url')
1702 else:
1703 self._downloader.report_warning('unable to extract uploader nickname')
1704
1705 # thumbnail image
1706 # We try first to get a high quality image:
1707 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1708 video_webpage, re.DOTALL)
1709 if m_thumb is not None:
1710 video_thumbnail = m_thumb.group(1)
1711 elif 'thumbnail_url' not in video_info:
1712 self._downloader.report_warning('unable to extract video thumbnail')
1713 video_thumbnail = None
1714 else: # don't panic if we can't find it
1715 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1716
1717 # upload date
1718 upload_date = self._html_search_meta(
1719 'datePublished', video_webpage, 'upload date', default=None)
1720 if not upload_date:
1721 upload_date = self._search_regex(
1722 [r'(?s)id="eow-date.*?>(.*?)</span>',
1723 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1724 video_webpage, 'upload date', default=None)
1725 upload_date = unified_strdate(upload_date)
1726
1727 video_license = self._html_search_regex(
1728 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1729 video_webpage, 'license', default=None)
1730
1731 m_music = re.search(
1732 r'''(?x)
1733 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1734 <ul[^>]*>\s*
1735 <li>(?P<title>.+?)
1736 by (?P<creator>.+?)
1737 (?:
1738 \(.+?\)|
1739 <a[^>]*
1740 (?:
1741 \bhref=["\']/red[^>]*>| # drop possible
1742 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1743 )
1744 .*?
1745 )?</li
1746 ''',
1747 video_webpage)
1748 if m_music:
1749 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1750 video_creator = clean_html(m_music.group('creator'))
1751 else:
1752 video_alt_title = video_creator = None
1753
1754 m_episode = re.search(
1755 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*ā€¢\s*E(?P<episode>\d+)</span>',
1756 video_webpage)
1757 if m_episode:
1758 series = m_episode.group('series')
1759 season_number = int(m_episode.group('season'))
1760 episode_number = int(m_episode.group('episode'))
1761 else:
1762 series = season_number = episode_number = None
1763
1764 m_cat_container = self._search_regex(
1765 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1766 video_webpage, 'categories', default=None)
1767 if m_cat_container:
1768 category = self._html_search_regex(
1769 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1770 default=None)
1771 video_categories = None if category is None else [category]
1772 else:
1773 video_categories = None
1774
1775 video_tags = [
1776 unescapeHTML(m.group('content'))
1777 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1778
1779 def _extract_count(count_name):
1780 return str_to_int(self._search_regex(
1781 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1782 % re.escape(count_name),
1783 video_webpage, count_name, default=None))
1784
1785 like_count = _extract_count('like')
1786 dislike_count = _extract_count('dislike')
1787
1788 # subtitles
1789 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1790 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1791
1792 video_duration = try_get(
1793 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1794 if not video_duration:
1795 video_duration = parse_duration(self._html_search_meta(
1796 'duration', video_webpage, 'video duration'))
1797
1798 # annotations
1799 video_annotations = None
1800 if self._downloader.params.get('writeannotations', False):
1801 video_annotations = self._extract_annotations(video_id)
1802
1803 chapters = self._extract_chapters(description_original, video_duration)
1804
1805 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1806 self.report_rtmp_download()
1807 formats = [{
1808 'format_id': '_rtmp',
1809 'protocol': 'rtmp',
1810 'url': video_info['conn'][0],
1811 'player_url': player_url,
1812 }]
1813 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1814 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1815 if 'rtmpe%3Dyes' in encoded_url_map:
1816 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1817 formats_spec = {}
1818 fmt_list = video_info.get('fmt_list', [''])[0]
1819 if fmt_list:
1820 for fmt in fmt_list.split(','):
1821 spec = fmt.split('/')
1822 if len(spec) > 1:
1823 width_height = spec[1].split('x')
1824 if len(width_height) == 2:
1825 formats_spec[spec[0]] = {
1826 'resolution': spec[1],
1827 'width': int_or_none(width_height[0]),
1828 'height': int_or_none(width_height[1]),
1829 }
1830 formats = []
1831 for url_data_str in encoded_url_map.split(','):
1832 url_data = compat_parse_qs(url_data_str)
1833 if 'itag' not in url_data or 'url' not in url_data:
1834 continue
1835 format_id = url_data['itag'][0]
1836 url = url_data['url'][0]
1837
1838 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1839 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1840 jsplayer_url_json = self._search_regex(
1841 ASSETS_RE,
1842 embed_webpage if age_gate else video_webpage,
1843 'JS player URL (1)', default=None)
1844 if not jsplayer_url_json and not age_gate:
1845 # We need the embed website after all
1846 if embed_webpage is None:
1847 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1848 embed_webpage = self._download_webpage(
1849 embed_url, video_id, 'Downloading embed webpage')
1850 jsplayer_url_json = self._search_regex(
1851 ASSETS_RE, embed_webpage, 'JS player URL')
1852
1853 player_url = json.loads(jsplayer_url_json)
1854 if player_url is None:
1855 player_url_json = self._search_regex(
1856 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1857 video_webpage, 'age gate player URL')
1858 player_url = json.loads(player_url_json)
1859
1860 if 'sig' in url_data:
1861 url += '&signature=' + url_data['sig'][0]
1862 elif 's' in url_data:
1863 encrypted_sig = url_data['s'][0]
1864
1865 if self._downloader.params.get('verbose'):
1866 if player_url is None:
1867 player_version = 'unknown'
1868 player_desc = 'unknown'
1869 else:
1870 if player_url.endswith('swf'):
1871 player_version = self._search_regex(
1872 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1873 'flash player', fatal=False)
1874 player_desc = 'flash player %s' % player_version
1875 else:
1876 player_version = self._search_regex(
1877 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1878 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1879 player_url,
1880 'html5 player', fatal=False)
1881 player_desc = 'html5 player %s' % player_version
1882
1883 parts_sizes = self._signature_cache_id(encrypted_sig)
1884 self.to_screen('{%s} signature length %s, %s' %
1885 (format_id, parts_sizes, player_desc))
1886
1887 signature = self._decrypt_signature(
1888 encrypted_sig, video_id, player_url, age_gate)
1889 url += '&signature=' + signature
1890 if 'ratebypass' not in url:
1891 url += '&ratebypass=yes'
1892
1893 dct = {
1894 'format_id': format_id,
1895 'url': url,
1896 'player_url': player_url,
1897 }
1898 if format_id in self._formats:
1899 dct.update(self._formats[format_id])
1900 if format_id in formats_spec:
1901 dct.update(formats_spec[format_id])
1902
1903 # Some itags are not included in DASH manifest thus corresponding formats will
1904 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1905 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1906 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1907 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1908
1909 more_fields = {
1910 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1911 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1912 'width': width,
1913 'height': height,
1914 'fps': int_or_none(url_data.get('fps', [None])[0]),
1915 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1916 }
1917 for key, value in more_fields.items():
1918 if value:
1919 dct[key] = value
1920 type_ = url_data.get('type', [None])[0]
1921 if type_:
1922 type_split = type_.split(';')
1923 kind_ext = type_split[0].split('/')
1924 if len(kind_ext) == 2:
1925 kind, _ = kind_ext
1926 dct['ext'] = mimetype2ext(type_split[0])
1927 if kind in ('audio', 'video'):
1928 codecs = None
1929 for mobj in re.finditer(
1930 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1931 if mobj.group('key') == 'codecs':
1932 codecs = mobj.group('val')
1933 break
1934 if codecs:
1935 dct.update(parse_codecs(codecs))
1936 formats.append(dct)
1937 elif video_info.get('hlsvp'):
1938 manifest_url = video_info['hlsvp'][0]
1939 formats = []
1940 m3u8_formats = self._extract_m3u8_formats(
1941 manifest_url, video_id, 'mp4', fatal=False)
1942 for a_format in m3u8_formats:
1943 itag = self._search_regex(
1944 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1945 if itag:
1946 a_format['format_id'] = itag
1947 if itag in self._formats:
1948 dct = self._formats[itag].copy()
1949 dct.update(a_format)
1950 a_format = dct
1951 a_format['player_url'] = player_url
1952 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1953 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1954 formats.append(a_format)
1955 else:
1956 unavailable_message = self._html_search_regex(
1957 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1958 video_webpage, 'unavailable message', default=None)
1959 if unavailable_message:
1960 raise ExtractorError(unavailable_message, expected=True)
1961 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1962
1963 # Look for the DASH manifest
1964 if self._downloader.params.get('youtube_include_dash_manifest', True):
1965 dash_mpd_fatal = True
1966 for mpd_url in dash_mpds:
1967 dash_formats = {}
1968 try:
1969 def decrypt_sig(mobj):
1970 s = mobj.group(1)
1971 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1972 return '/signature/%s' % dec_s
1973
1974 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1975
1976 for df in self._extract_mpd_formats(
1977 mpd_url, video_id, fatal=dash_mpd_fatal,
1978 formats_dict=self._formats):
1979 # Do not overwrite DASH format found in some previous DASH manifest
1980 if df['format_id'] not in dash_formats:
1981 dash_formats[df['format_id']] = df
1982 # Additional DASH manifests may end up in HTTP Error 403 therefore
1983 # allow them to fail without bug report message if we already have
1984 # some DASH manifest succeeded. This is temporary workaround to reduce
1985 # burst of bug reports until we figure out the reason and whether it
1986 # can be fixed at all.
1987 dash_mpd_fatal = False
1988 except (ExtractorError, KeyError) as e:
1989 self.report_warning(
1990 'Skipping DASH manifest: %r' % e, video_id)
1991 if dash_formats:
1992 # Remove the formats we found through non-DASH, they
1993 # contain less info and it can be wrong, because we use
1994 # fixed values (for example the resolution). See
1995 # https://github.com/rg3/youtube-dl/issues/5774 for an
1996 # example.
1997 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1998 formats.extend(dash_formats.values())
1999
2000 # Check for malformed aspect ratio
2001 stretched_m = re.search(
2002 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2003 video_webpage)
2004 if stretched_m:
2005 w = float(stretched_m.group('w'))
2006 h = float(stretched_m.group('h'))
2007 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2008 # We will only process correct ratios.
2009 if w > 0 and h > 0:
2010 ratio = w / h
2011 for f in formats:
2012 if f.get('vcodec') != 'none':
2013 f['stretched_ratio'] = ratio
2014
2015 self._sort_formats(formats)
2016
2017 self.mark_watched(video_id, video_info)
2018
2019 return {
2020 'id': video_id,
2021 'uploader': video_uploader,
2022 'uploader_id': video_uploader_id,
2023 'uploader_url': video_uploader_url,
2024 'upload_date': upload_date,
2025 'license': video_license,
2026 'creator': video_creator,
2027 'title': video_title,
2028 'alt_title': video_alt_title,
2029 'thumbnail': video_thumbnail,
2030 'description': video_description,
2031 'categories': video_categories,
2032 'tags': video_tags,
2033 'subtitles': video_subtitles,
2034 'automatic_captions': automatic_captions,
2035 'duration': video_duration,
2036 'age_limit': 18 if age_gate else 0,
2037 'annotations': video_annotations,
2038 'chapters': chapters,
2039 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2040 'view_count': view_count,
2041 'like_count': like_count,
2042 'dislike_count': dislike_count,
2043 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2044 'formats': formats,
2045 'is_live': is_live,
2046 'start_time': start_time,
2047 'end_time': end_time,
2048 'series': series,
2049 'season_number': season_number,
2050 'episode_number': episode_number,
2051 }
2052
2053
2054 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2055 IE_DESC = 'YouTube.com playlists'
2056 _VALID_URL = r"""(?x)(?:
2057 (?:https?://)?
2058 (?:\w+\.)?
2059 (?:
2060 youtube\.com/
2061 (?:
2062 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2063 \? (?:.*?[&;])*? (?:p|a|list)=
2064 | p/
2065 )|
2066 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2067 )
2068 (
2069 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2070 # Top tracks, they can also include dots
2071 |(?:MC)[\w\.]*
2072 )
2073 .*
2074 |
2075 (%(playlist_id)s)
2076 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2077 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2078 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2079 IE_NAME = 'youtube:playlist'
2080 _TESTS = [{
2081 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2082 'info_dict': {
2083 'title': 'ytdl test PL',
2084 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2085 },
2086 'playlist_count': 3,
2087 }, {
2088 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2089 'info_dict': {
2090 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2091 'title': 'YDL_Empty_List',
2092 },
2093 'playlist_count': 0,
2094 'skip': 'This playlist is private',
2095 }, {
2096 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2097 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2098 'info_dict': {
2099 'title': '29C3: Not my department',
2100 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2101 },
2102 'playlist_count': 95,
2103 }, {
2104 'note': 'issue #673',
2105 'url': 'PLBB231211A4F62143',
2106 'info_dict': {
2107 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2108 'id': 'PLBB231211A4F62143',
2109 },
2110 'playlist_mincount': 26,
2111 }, {
2112 'note': 'Large playlist',
2113 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2114 'info_dict': {
2115 'title': 'Uploads from Cauchemar',
2116 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2117 },
2118 'playlist_mincount': 799,
2119 }, {
2120 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2121 'info_dict': {
2122 'title': 'YDL_safe_search',
2123 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2124 },
2125 'playlist_count': 2,
2126 'skip': 'This playlist is private',
2127 }, {
2128 'note': 'embedded',
2129 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2130 'playlist_count': 4,
2131 'info_dict': {
2132 'title': 'JODA15',
2133 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2134 }
2135 }, {
2136 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2137 'playlist_mincount': 485,
2138 'info_dict': {
2139 'title': '2017 čÆčŖžęœ€ę–°å–®ę›² (2/24ꛓꖰ)',
2140 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2141 }
2142 }, {
2143 'note': 'Embedded SWF player',
2144 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2145 'playlist_count': 4,
2146 'info_dict': {
2147 'title': 'JODA7',
2148 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2149 }
2150 }, {
2151 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2152 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2153 'info_dict': {
2154 'title': 'Uploads from Interstellar Movie',
2155 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2156 },
2157 'playlist_mincount': 21,
2158 }, {
2159 # Playlist URL that does not actually serve a playlist
2160 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2161 'info_dict': {
2162 'id': 'FqZTN594JQw',
2163 'ext': 'webm',
2164 'title': "Smiley's People 01 detective, Adventure Series, Action",
2165 'uploader': 'STREEM',
2166 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2167 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2168 'upload_date': '20150526',
2169 'license': 'Standard YouTube License',
2170 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2171 'categories': ['People & Blogs'],
2172 'tags': list,
2173 'like_count': int,
2174 'dislike_count': int,
2175 },
2176 'params': {
2177 'skip_download': True,
2178 },
2179 'add_ie': [YoutubeIE.ie_key()],
2180 }, {
2181 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2182 'info_dict': {
2183 'id': 'yeWKywCrFtk',
2184 'ext': 'mp4',
2185 'title': 'Small Scale Baler and Braiding Rugs',
2186 'uploader': 'Backus-Page House Museum',
2187 'uploader_id': 'backuspagemuseum',
2188 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2189 'upload_date': '20161008',
2190 'license': 'Standard YouTube License',
2191 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2192 'categories': ['Nonprofits & Activism'],
2193 'tags': list,
2194 'like_count': int,
2195 'dislike_count': int,
2196 },
2197 'params': {
2198 'noplaylist': True,
2199 'skip_download': True,
2200 },
2201 }, {
2202 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2203 'only_matching': True,
2204 }, {
2205 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2206 'only_matching': True,
2207 }]
2208
2209 def _real_initialize(self):
2210 self._login()
2211
2212 def _extract_mix(self, playlist_id):
2213 # The mixes are generated from a single video
2214 # the id of the playlist is just 'RD' + video_id
2215 ids = []
2216 last_id = playlist_id[-11:]
2217 for n in itertools.count(1):
2218 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2219 webpage = self._download_webpage(
2220 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2221 new_ids = orderedSet(re.findall(
2222 r'''(?xs)data-video-username=".*?".*?
2223 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2224 webpage))
2225 # Fetch new pages until all the videos are repeated, it seems that
2226 # there are always 51 unique videos.
2227 new_ids = [_id for _id in new_ids if _id not in ids]
2228 if not new_ids:
2229 break
2230 ids.extend(new_ids)
2231 last_id = ids[-1]
2232
2233 url_results = self._ids_to_results(ids)
2234
2235 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2236 title_span = (
2237 search_title('playlist-title') or
2238 search_title('title long-title') or
2239 search_title('title'))
2240 title = clean_html(title_span)
2241
2242 return self.playlist_result(url_results, playlist_id, title)
2243
2244 def _extract_playlist(self, playlist_id):
2245 url = self._TEMPLATE_URL % playlist_id
2246 page = self._download_webpage(url, playlist_id)
2247
2248 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2249 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2250 match = match.strip()
2251 # Check if the playlist exists or is private
2252 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2253 if mobj:
2254 reason = mobj.group('reason')
2255 message = 'This playlist %s' % reason
2256 if 'private' in reason:
2257 message += ', use --username or --netrc to access it'
2258 message += '.'
2259 raise ExtractorError(message, expected=True)
2260 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2261 raise ExtractorError(
2262 'Invalid parameters. Maybe URL is incorrect.',
2263 expected=True)
2264 elif re.match(r'[^<]*Choose your language[^<]*', match):
2265 continue
2266 else:
2267 self.report_warning('Youtube gives an alert message: ' + match)
2268
2269 playlist_title = self._html_search_regex(
2270 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2271 page, 'title', default=None)
2272
2273 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2274 uploader = self._search_regex(
2275 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2276 page, 'uploader', default=None)
2277 mobj = re.search(
2278 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2279 page)
2280 if mobj:
2281 uploader_id = mobj.group('uploader_id')
2282 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2283 else:
2284 uploader_id = uploader_url = None
2285
2286 has_videos = True
2287
2288 if not playlist_title:
2289 try:
2290 # Some playlist URLs don't actually serve a playlist (e.g.
2291 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2292 next(self._entries(page, playlist_id))
2293 except StopIteration:
2294 has_videos = False
2295
2296 playlist = self.playlist_result(
2297 self._entries(page, playlist_id), playlist_id, playlist_title)
2298 playlist.update({
2299 'uploader': uploader,
2300 'uploader_id': uploader_id,
2301 'uploader_url': uploader_url,
2302 })
2303
2304 return has_videos, playlist
2305
2306 def _check_download_just_video(self, url, playlist_id):
2307 # Check if it's a video-specific URL
2308 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2309 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2310 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2311 'video id', default=None)
2312 if video_id:
2313 if self._downloader.params.get('noplaylist'):
2314 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2315 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2316 else:
2317 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2318 return video_id, None
2319 return None, None
2320
2321 def _real_extract(self, url):
2322 # Extract playlist id
2323 mobj = re.match(self._VALID_URL, url)
2324 if mobj is None:
2325 raise ExtractorError('Invalid URL: %s' % url)
2326 playlist_id = mobj.group(1) or mobj.group(2)
2327
2328 video_id, video = self._check_download_just_video(url, playlist_id)
2329 if video:
2330 return video
2331
2332 if playlist_id.startswith(('RD', 'UL', 'PU')):
2333 # Mixes require a custom extraction process
2334 return self._extract_mix(playlist_id)
2335
2336 has_videos, playlist = self._extract_playlist(playlist_id)
2337 if has_videos or not video_id:
2338 return playlist
2339
2340 # Some playlist URLs don't actually serve a playlist (see
2341 # https://github.com/rg3/youtube-dl/issues/10537).
2342 # Fallback to plain video extraction if there is a video id
2343 # along with playlist id.
2344 return self.url_result(video_id, 'Youtube', video_id=video_id)
2345
2346
2347 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2348 IE_DESC = 'YouTube.com channels'
2349 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2350 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2351 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2352 IE_NAME = 'youtube:channel'
2353 _TESTS = [{
2354 'note': 'paginated channel',
2355 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2356 'playlist_mincount': 91,
2357 'info_dict': {
2358 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2359 'title': 'Uploads from lex will',
2360 }
2361 }, {
2362 'note': 'Age restricted channel',
2363 # from https://www.youtube.com/user/DeusExOfficial
2364 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2365 'playlist_mincount': 64,
2366 'info_dict': {
2367 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2368 'title': 'Uploads from Deus Ex',
2369 },
2370 }]
2371
2372 @classmethod
2373 def suitable(cls, url):
2374 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2375 else super(YoutubeChannelIE, cls).suitable(url))
2376
2377 def _build_template_url(self, url, channel_id):
2378 return self._TEMPLATE_URL % channel_id
2379
2380 def _real_extract(self, url):
2381 channel_id = self._match_id(url)
2382
2383 url = self._build_template_url(url, channel_id)
2384
2385 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2386 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2387 # otherwise fallback on channel by page extraction
2388 channel_page = self._download_webpage(
2389 url + '?view=57', channel_id,
2390 'Downloading channel page', fatal=False)
2391 if channel_page is False:
2392 channel_playlist_id = False
2393 else:
2394 channel_playlist_id = self._html_search_meta(
2395 'channelId', channel_page, 'channel id', default=None)
2396 if not channel_playlist_id:
2397 channel_url = self._html_search_meta(
2398 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2399 channel_page, 'channel url', default=None)
2400 if channel_url:
2401 channel_playlist_id = self._search_regex(
2402 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2403 channel_url, 'channel id', default=None)
2404 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2405 playlist_id = 'UU' + channel_playlist_id[2:]
2406 return self.url_result(
2407 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2408
2409 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2410 autogenerated = re.search(r'''(?x)
2411 class="[^"]*?(?:
2412 channel-header-autogenerated-label|
2413 yt-channel-title-autogenerated
2414 )[^"]*"''', channel_page) is not None
2415
2416 if autogenerated:
2417 # The videos are contained in a single page
2418 # the ajax pages can't be used, they are empty
2419 entries = [
2420 self.url_result(
2421 video_id, 'Youtube', video_id=video_id,
2422 video_title=video_title)
2423 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2424 return self.playlist_result(entries, channel_id)
2425
2426 try:
2427 next(self._entries(channel_page, channel_id))
2428 except StopIteration:
2429 alert_message = self._html_search_regex(
2430 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2431 channel_page, 'alert', default=None, group='alert')
2432 if alert_message:
2433 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2434
2435 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2436
2437
2438 class YoutubeUserIE(YoutubeChannelIE):
2439 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2440 _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_-]+)'
2441 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2442 IE_NAME = 'youtube:user'
2443
2444 _TESTS = [{
2445 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2446 'playlist_mincount': 320,
2447 'info_dict': {
2448 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2449 'title': 'Uploads from The Linux Foundation',
2450 }
2451 }, {
2452 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2453 # but not https://www.youtube.com/user/12minuteathlete/videos
2454 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2455 'playlist_mincount': 249,
2456 'info_dict': {
2457 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2458 'title': 'Uploads from 12 Minute Athlete',
2459 }
2460 }, {
2461 'url': 'ytuser:phihag',
2462 'only_matching': True,
2463 }, {
2464 'url': 'https://www.youtube.com/c/gametrailers',
2465 'only_matching': True,
2466 }, {
2467 'url': 'https://www.youtube.com/gametrailers',
2468 'only_matching': True,
2469 }, {
2470 # This channel is not available, geo restricted to JP
2471 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2472 'only_matching': True,
2473 }]
2474
2475 @classmethod
2476 def suitable(cls, url):
2477 # Don't return True if the url can be extracted with other youtube
2478 # extractor, the regex would is too permissive and it would match.
2479 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2480 if any(ie.suitable(url) for ie in other_yt_ies):
2481 return False
2482 else:
2483 return super(YoutubeUserIE, cls).suitable(url)
2484
2485 def _build_template_url(self, url, channel_id):
2486 mobj = re.match(self._VALID_URL, url)
2487 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2488
2489
2490 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2491 IE_DESC = 'YouTube.com live streams'
2492 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2493 IE_NAME = 'youtube:live'
2494
2495 _TESTS = [{
2496 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2497 'info_dict': {
2498 'id': 'a48o2S1cPoo',
2499 'ext': 'mp4',
2500 'title': 'The Young Turks - Live Main Show',
2501 'uploader': 'The Young Turks',
2502 'uploader_id': 'TheYoungTurks',
2503 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2504 'upload_date': '20150715',
2505 'license': 'Standard YouTube License',
2506 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2507 'categories': ['News & Politics'],
2508 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2509 'like_count': int,
2510 'dislike_count': int,
2511 },
2512 'params': {
2513 'skip_download': True,
2514 },
2515 }, {
2516 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2517 'only_matching': True,
2518 }, {
2519 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2520 'only_matching': True,
2521 }, {
2522 'url': 'https://www.youtube.com/TheYoungTurks/live',
2523 'only_matching': True,
2524 }]
2525
2526 def _real_extract(self, url):
2527 mobj = re.match(self._VALID_URL, url)
2528 channel_id = mobj.group('id')
2529 base_url = mobj.group('base_url')
2530 webpage = self._download_webpage(url, channel_id, fatal=False)
2531 if webpage:
2532 page_type = self._og_search_property(
2533 'type', webpage, 'page type', default=None)
2534 video_id = self._html_search_meta(
2535 'videoId', webpage, 'video id', default=None)
2536 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2537 return self.url_result(video_id, YoutubeIE.ie_key())
2538 return self.url_result(base_url)
2539
2540
2541 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2542 IE_DESC = 'YouTube.com user/channel playlists'
2543 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2544 IE_NAME = 'youtube:playlists'
2545
2546 _TESTS = [{
2547 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2548 'playlist_mincount': 4,
2549 'info_dict': {
2550 'id': 'ThirstForScience',
2551 'title': 'Thirst for Science',
2552 },
2553 }, {
2554 # with "Load more" button
2555 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2556 'playlist_mincount': 70,
2557 'info_dict': {
2558 'id': 'igorkle1',
2559 'title': 'Š˜Š³Š¾Ń€ŃŒ ŠšŠ»ŠµŠ¹Š½ŠµŃ€',
2560 },
2561 }, {
2562 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2563 'playlist_mincount': 17,
2564 'info_dict': {
2565 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2566 'title': 'Chem Player',
2567 },
2568 }]
2569
2570
2571 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2572 IE_DESC = 'YouTube.com searches'
2573 # there doesn't appear to be a real limit, for example if you search for
2574 # 'python' you get more than 8.000.000 results
2575 _MAX_RESULTS = float('inf')
2576 IE_NAME = 'youtube:search'
2577 _SEARCH_KEY = 'ytsearch'
2578 _EXTRA_QUERY_ARGS = {}
2579 _TESTS = []
2580
2581 def _get_n_results(self, query, n):
2582 """Get a specified number of results for a query"""
2583
2584 videos = []
2585 limit = n
2586
2587 url_query = {
2588 'search_query': query.encode('utf-8'),
2589 }
2590 url_query.update(self._EXTRA_QUERY_ARGS)
2591 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2592
2593 for pagenum in itertools.count(1):
2594 data = self._download_json(
2595 result_url, video_id='query "%s"' % query,
2596 note='Downloading page %s' % pagenum,
2597 errnote='Unable to download API page',
2598 query={'spf': 'navigate'})
2599 html_content = data[1]['body']['content']
2600
2601 if 'class="search-message' in html_content:
2602 raise ExtractorError(
2603 '[youtube] No video results', expected=True)
2604
2605 new_videos = self._ids_to_results(orderedSet(re.findall(
2606 r'href="/watch\?v=(.{11})', html_content)))
2607 videos += new_videos
2608 if not new_videos or len(videos) > limit:
2609 break
2610 next_link = self._html_search_regex(
2611 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2612 html_content, 'next link', default=None)
2613 if next_link is None:
2614 break
2615 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2616
2617 if len(videos) > n:
2618 videos = videos[:n]
2619 return self.playlist_result(videos, query)
2620
2621
2622 class YoutubeSearchDateIE(YoutubeSearchIE):
2623 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2624 _SEARCH_KEY = 'ytsearchdate'
2625 IE_DESC = 'YouTube.com searches, newest videos first'
2626 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2627
2628
2629 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2630 IE_DESC = 'YouTube.com search URLs'
2631 IE_NAME = 'youtube:search_url'
2632 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2633 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2634 _TESTS = [{
2635 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2636 'playlist_mincount': 5,
2637 'info_dict': {
2638 'title': 'youtube-dl test video',
2639 }
2640 }, {
2641 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2642 'only_matching': True,
2643 }]
2644
2645 def _real_extract(self, url):
2646 mobj = re.match(self._VALID_URL, url)
2647 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2648 webpage = self._download_webpage(url, query)
2649 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2650
2651
2652 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2653 IE_DESC = 'YouTube.com (multi-season) shows'
2654 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2655 IE_NAME = 'youtube:show'
2656 _TESTS = [{
2657 'url': 'https://www.youtube.com/show/airdisasters',
2658 'playlist_mincount': 5,
2659 'info_dict': {
2660 'id': 'airdisasters',
2661 'title': 'Air Disasters',
2662 }
2663 }]
2664
2665 def _real_extract(self, url):
2666 playlist_id = self._match_id(url)
2667 return super(YoutubeShowIE, self)._real_extract(
2668 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2669
2670
2671 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2672 """
2673 Base class for feed extractors
2674 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2675 """
2676 _LOGIN_REQUIRED = True
2677
2678 @property
2679 def IE_NAME(self):
2680 return 'youtube:%s' % self._FEED_NAME
2681
2682 def _real_initialize(self):
2683 self._login()
2684
2685 def _real_extract(self, url):
2686 page = self._download_webpage(
2687 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2688
2689 # The extraction process is the same as for playlists, but the regex
2690 # for the video ids doesn't contain an index
2691 ids = []
2692 more_widget_html = content_html = page
2693 for page_num in itertools.count(1):
2694 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2695
2696 # 'recommended' feed has infinite 'load more' and each new portion spins
2697 # the same videos in (sometimes) slightly different order, so we'll check
2698 # for unicity and break when portion has no new videos
2699 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2700 if not new_ids:
2701 break
2702
2703 ids.extend(new_ids)
2704
2705 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2706 if not mobj:
2707 break
2708
2709 more = self._download_json(
2710 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2711 'Downloading page #%s' % page_num,
2712 transform_source=uppercase_escape)
2713 content_html = more['content_html']
2714 more_widget_html = more['load_more_widget_html']
2715
2716 return self.playlist_result(
2717 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2718
2719
2720 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2721 IE_NAME = 'youtube:watchlater'
2722 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2723 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2724
2725 _TESTS = [{
2726 'url': 'https://www.youtube.com/playlist?list=WL',
2727 'only_matching': True,
2728 }, {
2729 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2730 'only_matching': True,
2731 }]
2732
2733 def _real_extract(self, url):
2734 _, video = self._check_download_just_video(url, 'WL')
2735 if video:
2736 return video
2737 _, playlist = self._extract_playlist('WL')
2738 return playlist
2739
2740
2741 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2742 IE_NAME = 'youtube:favorites'
2743 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2744 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2745 _LOGIN_REQUIRED = True
2746
2747 def _real_extract(self, url):
2748 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2749 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2750 return self.url_result(playlist_id, 'YoutubePlaylist')
2751
2752
2753 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2754 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2755 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2756 _FEED_NAME = 'recommended'
2757 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2758
2759
2760 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2761 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2762 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2763 _FEED_NAME = 'subscriptions'
2764 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2765
2766
2767 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2768 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2769 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2770 _FEED_NAME = 'history'
2771 _PLAYLIST_TITLE = 'Youtube History'
2772
2773
2774 class YoutubeTruncatedURLIE(InfoExtractor):
2775 IE_NAME = 'youtube:truncated_url'
2776 IE_DESC = False # Do not list
2777 _VALID_URL = r'''(?x)
2778 (?:https?://)?
2779 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2780 (?:watch\?(?:
2781 feature=[a-z_]+|
2782 annotation_id=annotation_[^&]+|
2783 x-yt-cl=[0-9]+|
2784 hl=[^&]*|
2785 t=[0-9]+
2786 )?
2787 |
2788 attribution_link\?a=[^&]+
2789 )
2790 $
2791 '''
2792
2793 _TESTS = [{
2794 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2795 'only_matching': True,
2796 }, {
2797 'url': 'https://www.youtube.com/watch?',
2798 'only_matching': True,
2799 }, {
2800 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2801 'only_matching': True,
2802 }, {
2803 'url': 'https://www.youtube.com/watch?feature=foo',
2804 'only_matching': True,
2805 }, {
2806 'url': 'https://www.youtube.com/watch?hl=en-GB',
2807 'only_matching': True,
2808 }, {
2809 'url': 'https://www.youtube.com/watch?t=2372',
2810 'only_matching': True,
2811 }]
2812
2813 def _real_extract(self, url):
2814 raise ExtractorError(
2815 'Did you forget to quote the URL? Remember that & is a meta '
2816 'character in most shells, so you want to put the URL in quotes, '
2817 'like youtube-dl '
2818 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2819 ' or simply youtube-dl BaW_jenozKc .',
2820 expected=True)
2821
2822
2823 class YoutubeTruncatedIDIE(InfoExtractor):
2824 IE_NAME = 'youtube:truncated_id'
2825 IE_DESC = False # Do not list
2826 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2827
2828 _TESTS = [{
2829 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2830 'only_matching': True,
2831 }]
2832
2833 def _real_extract(self, url):
2834 video_id = self._match_id(url)
2835 raise ExtractorError(
2836 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2837 expected=True)