]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[youtube] Fix authentication (closes #12927)
[yt-dlp.git] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5
6 import itertools
7 import json
8 import os.path
9 import random
10 import re
11 import time
12 import traceback
13
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
18 compat_chr,
19 compat_parse_qs,
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
24 compat_urlparse,
25 compat_str,
26 )
27 from ..utils import (
28 clean_html,
29 error_to_compat_str,
30 ExtractorError,
31 float_or_none,
32 get_element_by_attribute,
33 get_element_by_id,
34 int_or_none,
35 mimetype2ext,
36 orderedSet,
37 parse_codecs,
38 parse_duration,
39 remove_quotes,
40 # remove_start,
41 smuggle_url,
42 str_to_int,
43 try_get,
44 unescapeHTML,
45 unified_strdate,
46 unsmuggle_url,
47 uppercase_escape,
48 urlencode_postdata,
49 )
50
51
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
56
57 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
58 _LOOKUP_REQ_TEMPLATE = '["{0}",null,[],null,"US",null,null,2,false,true,[null,null,[2,1,null,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",null,[],4],1,[null,null,[]],null,null,null,true],"{0}"]'
59
60 _PASSWORD_CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
61 _PASSWORD_CHALLENGE_REQ_TEMPLATE = '["{0}",null,1,null,[1,null,null,null,["{1}",null,true]],[null,null,[2,1,null,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",null,[],4],1,[null,null,[]],null,null,null,true]]'
62
63 _TFA_URL = 'https://accounts.google.com/_/signin/challenge'
64 _TFA_REQ_TEMPLATE = '["{0}",null,2,null,[9,null,null,null,null,null,null,null,[null,"{1}",false,2]]]'
65
66 _NETRC_MACHINE = 'youtube'
67 # If True it will raise an error if no login info is provided
68 _LOGIN_REQUIRED = False
69
70 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
71
72 def _set_language(self):
73 self._set_cookie(
74 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
75 # YouTube sets the expire time to about two months
76 expire_time=time.time() + 2 * 30 * 24 * 3600)
77
78 def _ids_to_results(self, ids):
79 return [
80 self.url_result(vid_id, 'Youtube', video_id=vid_id)
81 for vid_id in ids]
82
83 def _login(self):
84 """
85 Attempt to log in to YouTube.
86 True is returned if successful or skipped.
87 False is returned if login failed.
88
89 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
90 """
91 (username, password) = self._get_login_info()
92 # No authentication to be performed
93 if username is None:
94 if self._LOGIN_REQUIRED:
95 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
96 return True
97
98 login_page = self._download_webpage(
99 self._LOGIN_URL, None,
100 note='Downloading login page',
101 errnote='unable to fetch login page', fatal=False)
102 if login_page is False:
103 return
104
105 login_form = self._hidden_inputs(login_page)
106
107 def req(url, f_req, note, errnote):
108 data = login_form.copy()
109 data.update({
110 'pstMsg': 1,
111 'checkConnection': 'youtube',
112 'checkedDomains': 'youtube',
113 'hl': 'en',
114 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
115 'f.req': f_req,
116 'flowName': 'GlifWebSignIn',
117 'flowEntry': 'ServiceLogin',
118 })
119 return self._download_json(
120 url, None, note=note, errnote=errnote,
121 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
122 fatal=False,
123 data=urlencode_postdata(data), headers={
124 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
125 'Google-Accounts-XSRF': 1,
126 })
127
128 lookup_results = req(
129 self._LOOKUP_URL, self._LOOKUP_REQ_TEMPLATE.format(username),
130 'Looking up account info', 'Unable to look up account info')
131
132 if lookup_results is False:
133 return False
134
135 user_hash = lookup_results[0][2]
136
137 password_challenge_results = req(
138 self._PASSWORD_CHALLENGE_URL,
139 self._PASSWORD_CHALLENGE_REQ_TEMPLATE.format(user_hash, password),
140 'Logging in', 'Unable to log in')[0]
141
142 if password_challenge_results is False:
143 return
144
145 msg = password_challenge_results[5]
146 if msg is not None and isinstance(msg, list):
147 raise ExtractorError('Unable to login: %s' % msg[5], expected=True)
148
149 password_challenge_results = password_challenge_results[-1]
150
151 # tfa = password_challenge_results[0]
152 # if isinstance(tfa, list) and tfa[0][2] == 'TWO_STEP_VERIFICATION':
153 # tfa_code = self._get_tfa_info('2-step verification code')
154 #
155 # if not tfa_code:
156 # self._downloader.report_warning(
157 # 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
158 # '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
159 # return False
160 #
161 # tfa_code = remove_start(tfa_code, 'G-')
162 # print('tfa', tfa_code)
163 # tfa_results = req(
164 # self._TFA_URL,
165 # self._TFA_REQ_TEMPLATE.format(user_hash, tfa_code),
166 # 'Submitting TFA code', 'Unable to submit TFA code')
167 #
168 # TODO
169
170 check_cookie_results = self._download_webpage(
171 password_challenge_results[2], None, 'Checking cookie')
172
173 if '>Sign out<' not in check_cookie_results:
174 self._downloader.report_warning('Unable to log in')
175 return False
176
177 return True
178
179 def _real_initialize(self):
180 if self._downloader is None:
181 return
182 self._set_language()
183 if not self._login():
184 return
185
186
187 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
188 # Extract entries from page with "Load more" button
189 def _entries(self, page, playlist_id):
190 more_widget_html = content_html = page
191 for page_num in itertools.count(1):
192 for entry in self._process_page(content_html):
193 yield entry
194
195 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
196 if not mobj:
197 break
198
199 more = self._download_json(
200 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
201 'Downloading page #%s' % page_num,
202 transform_source=uppercase_escape)
203 content_html = more['content_html']
204 if not content_html.strip():
205 # Some webpages show a "Load more" button but they don't
206 # have more videos
207 break
208 more_widget_html = more['load_more_widget_html']
209
210
211 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
212 def _process_page(self, content):
213 for video_id, video_title in self.extract_videos_from_page(content):
214 yield self.url_result(video_id, 'Youtube', video_id, video_title)
215
216 def extract_videos_from_page(self, page):
217 ids_in_page = []
218 titles_in_page = []
219 for mobj in re.finditer(self._VIDEO_RE, page):
220 # The link with index 0 is not the first video of the playlist (not sure if still actual)
221 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
222 continue
223 video_id = mobj.group('id')
224 video_title = unescapeHTML(mobj.group('title'))
225 if video_title:
226 video_title = video_title.strip()
227 try:
228 idx = ids_in_page.index(video_id)
229 if video_title and not titles_in_page[idx]:
230 titles_in_page[idx] = video_title
231 except ValueError:
232 ids_in_page.append(video_id)
233 titles_in_page.append(video_title)
234 return zip(ids_in_page, titles_in_page)
235
236
237 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
238 def _process_page(self, content):
239 for playlist_id in orderedSet(re.findall(
240 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
241 content)):
242 yield self.url_result(
243 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
244
245 def _real_extract(self, url):
246 playlist_id = self._match_id(url)
247 webpage = self._download_webpage(url, playlist_id)
248 title = self._og_search_title(webpage, fatal=False)
249 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
250
251
252 class YoutubeIE(YoutubeBaseInfoExtractor):
253 IE_DESC = 'YouTube.com'
254 _VALID_URL = r"""(?x)^
255 (
256 (?:https?://|//) # http(s):// or protocol-independent URL
257 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
258 (?:www\.)?deturl\.com/www\.youtube\.com/|
259 (?:www\.)?pwnyoutube\.com/|
260 (?:www\.)?yourepeat\.com/|
261 tube\.majestyc\.net/|
262 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
263 (?:.*?\#/)? # handle anchor (#/) redirect urls
264 (?: # the various things that can precede the ID:
265 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
266 |(?: # or the v= param in all its forms
267 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
268 (?:\?|\#!?) # the params delimiter ? or # or #!
269 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
270 v=
271 )
272 ))
273 |(?:
274 youtu\.be| # just youtu.be/xxxx
275 vid\.plus| # or vid.plus/xxxx
276 zwearz\.com/watch| # or zwearz.com/watch/xxxx
277 )/
278 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
279 )
280 )? # all until now is optional -> you can pass the naked ID
281 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
282 (?!.*?\blist=
283 (?:
284 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
285 WL # WL are handled by the watch later IE
286 )
287 )
288 (?(1).+)? # if we found the ID, everything can follow
289 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
290 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
291 _formats = {
292 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
293 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
294 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
295 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
296 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
297 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
298 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
300 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
301 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
302 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
303 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
304 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
305 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
306 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
307 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
308 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
309 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
310
311
312 # 3D videos
313 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
314 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
315 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
316 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
317 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
318 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
319 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
320
321 # Apple HTTP Live Streaming
322 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
323 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
324 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
325 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
326 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
327 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
328 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
329 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
330
331 # DASH mp4 video
332 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
333 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
334 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
335 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
336 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
337 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
338 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
339 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
340 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
341 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
342 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
343 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
344
345 # Dash mp4 audio
346 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
347 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
348 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
349 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
350 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
351 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
352 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
353
354 # Dash webm
355 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
356 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
357 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
358 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
359 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
360 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
361 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
362 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
363 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
364 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
365 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
366 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
367 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
368 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
369 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
370 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
371 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
372 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
373 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
374 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
375 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
376 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
377
378 # Dash webm audio
379 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
380 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
381
382 # Dash webm audio with opus inside
383 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
384 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
385 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
386
387 # RTMP (unnamed)
388 '_rtmp': {'protocol': 'rtmp'},
389 }
390 _SUBTITLE_FORMATS = ('ttml', 'vtt')
391
392 _GEO_BYPASS = False
393
394 IE_NAME = 'youtube'
395 _TESTS = [
396 {
397 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
398 'info_dict': {
399 'id': 'BaW_jenozKc',
400 'ext': 'mp4',
401 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
402 'uploader': 'Philipp Hagemeister',
403 'uploader_id': 'phihag',
404 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
405 'upload_date': '20121002',
406 'license': 'Standard YouTube License',
407 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
408 'categories': ['Science & Technology'],
409 'tags': ['youtube-dl'],
410 'duration': 10,
411 'like_count': int,
412 'dislike_count': int,
413 'start_time': 1,
414 'end_time': 9,
415 }
416 },
417 {
418 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
419 'note': 'Test generic use_cipher_signature video (#897)',
420 'info_dict': {
421 'id': 'UxxajLWwzqY',
422 'ext': 'mp4',
423 'upload_date': '20120506',
424 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
425 'alt_title': 'I Love It (feat. Charli XCX)',
426 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
427 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
428 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
429 'iconic ep', 'iconic', 'love', 'it'],
430 'duration': 180,
431 'uploader': 'Icona Pop',
432 'uploader_id': 'IconaPop',
433 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
434 'license': 'Standard YouTube License',
435 'creator': 'Icona Pop',
436 }
437 },
438 {
439 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
440 'note': 'Test VEVO video with age protection (#956)',
441 'info_dict': {
442 'id': '07FYdnEawAQ',
443 'ext': 'mp4',
444 'upload_date': '20130703',
445 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
446 'alt_title': 'Tunnel Vision',
447 'description': 'md5:64249768eec3bc4276236606ea996373',
448 'duration': 419,
449 'uploader': 'justintimberlakeVEVO',
450 'uploader_id': 'justintimberlakeVEVO',
451 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
452 'license': 'Standard YouTube License',
453 'creator': 'Justin Timberlake',
454 'age_limit': 18,
455 }
456 },
457 {
458 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
459 'note': 'Embed-only video (#1746)',
460 'info_dict': {
461 'id': 'yZIXLfi8CZQ',
462 'ext': 'mp4',
463 'upload_date': '20120608',
464 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
465 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
466 'uploader': 'SET India',
467 'uploader_id': 'setindia',
468 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
469 'license': 'Standard YouTube License',
470 'age_limit': 18,
471 }
472 },
473 {
474 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
475 'note': 'Use the first video ID in the URL',
476 'info_dict': {
477 'id': 'BaW_jenozKc',
478 'ext': 'mp4',
479 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
480 'uploader': 'Philipp Hagemeister',
481 'uploader_id': 'phihag',
482 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
483 'upload_date': '20121002',
484 'license': 'Standard YouTube License',
485 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
486 'categories': ['Science & Technology'],
487 'tags': ['youtube-dl'],
488 'duration': 10,
489 'like_count': int,
490 'dislike_count': int,
491 },
492 'params': {
493 'skip_download': True,
494 },
495 },
496 {
497 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
498 'note': '256k DASH audio (format 141) via DASH manifest',
499 'info_dict': {
500 'id': 'a9LDPn-MO4I',
501 'ext': 'm4a',
502 'upload_date': '20121002',
503 'uploader_id': '8KVIDEO',
504 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
505 'description': '',
506 'uploader': '8KVIDEO',
507 'license': 'Standard YouTube License',
508 'title': 'UHDTV TEST 8K VIDEO.mp4'
509 },
510 'params': {
511 'youtube_include_dash_manifest': True,
512 'format': '141',
513 },
514 'skip': 'format 141 not served anymore',
515 },
516 # DASH manifest with encrypted signature
517 {
518 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
519 'info_dict': {
520 'id': 'IB3lcPjvWLA',
521 'ext': 'm4a',
522 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
523 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
524 'duration': 244,
525 'uploader': 'AfrojackVEVO',
526 'uploader_id': 'AfrojackVEVO',
527 'upload_date': '20131011',
528 'license': 'Standard YouTube License',
529 },
530 'params': {
531 'youtube_include_dash_manifest': True,
532 'format': '141/bestaudio[ext=m4a]',
533 },
534 },
535 # JS player signature function name containing $
536 {
537 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
538 'info_dict': {
539 'id': 'nfWlot6h_JM',
540 'ext': 'm4a',
541 'title': 'Taylor Swift - Shake It Off',
542 'alt_title': 'Shake It Off',
543 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
544 'duration': 242,
545 'uploader': 'TaylorSwiftVEVO',
546 'uploader_id': 'TaylorSwiftVEVO',
547 'upload_date': '20140818',
548 'license': 'Standard YouTube License',
549 'creator': 'Taylor Swift',
550 },
551 'params': {
552 'youtube_include_dash_manifest': True,
553 'format': '141/bestaudio[ext=m4a]',
554 },
555 },
556 # Controversy video
557 {
558 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
559 'info_dict': {
560 'id': 'T4XJQO3qol8',
561 'ext': 'mp4',
562 'duration': 219,
563 'upload_date': '20100909',
564 'uploader': 'The Amazing Atheist',
565 'uploader_id': 'TheAmazingAtheist',
566 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
567 'license': 'Standard YouTube License',
568 'title': 'Burning Everyone\'s Koran',
569 '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',
570 }
571 },
572 # Normal age-gate video (No vevo, embed allowed)
573 {
574 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
575 'info_dict': {
576 'id': 'HtVdAasjOgU',
577 'ext': 'mp4',
578 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
579 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
580 'duration': 142,
581 'uploader': 'The Witcher',
582 'uploader_id': 'WitcherGame',
583 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
584 'upload_date': '20140605',
585 'license': 'Standard YouTube License',
586 'age_limit': 18,
587 },
588 },
589 # Age-gate video with encrypted signature
590 {
591 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
592 'info_dict': {
593 'id': '6kLq3WMV1nU',
594 'ext': 'mp4',
595 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
596 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
597 'duration': 247,
598 'uploader': 'LloydVEVO',
599 'uploader_id': 'LloydVEVO',
600 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
601 'upload_date': '20110629',
602 'license': 'Standard YouTube License',
603 'age_limit': 18,
604 },
605 },
606 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
607 {
608 'url': '__2ABJjxzNo',
609 'info_dict': {
610 'id': '__2ABJjxzNo',
611 'ext': 'mp4',
612 'duration': 266,
613 'upload_date': '20100430',
614 'uploader_id': 'deadmau5',
615 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
616 'creator': 'deadmau5',
617 'description': 'md5:12c56784b8032162bb936a5f76d55360',
618 'uploader': 'deadmau5',
619 'license': 'Standard YouTube License',
620 'title': 'Deadmau5 - Some Chords (HD)',
621 'alt_title': 'Some Chords',
622 },
623 'expected_warnings': [
624 'DASH manifest missing',
625 ]
626 },
627 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
628 {
629 'url': 'lqQg6PlCWgI',
630 'info_dict': {
631 'id': 'lqQg6PlCWgI',
632 'ext': 'mp4',
633 'duration': 6085,
634 'upload_date': '20150827',
635 'uploader_id': 'olympic',
636 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
637 'license': 'Standard YouTube License',
638 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
639 'uploader': 'Olympic',
640 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
641 },
642 'params': {
643 'skip_download': 'requires avconv',
644 }
645 },
646 # Non-square pixels
647 {
648 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
649 'info_dict': {
650 'id': '_b-2C3KPAM0',
651 'ext': 'mp4',
652 'stretched_ratio': 16 / 9.,
653 'duration': 85,
654 'upload_date': '20110310',
655 'uploader_id': 'AllenMeow',
656 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
657 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
658 'uploader': '孫艾倫',
659 'license': 'Standard YouTube License',
660 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
661 },
662 },
663 # url_encoded_fmt_stream_map is empty string
664 {
665 'url': 'qEJwOuvDf7I',
666 'info_dict': {
667 'id': 'qEJwOuvDf7I',
668 'ext': 'webm',
669 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
670 'description': '',
671 'upload_date': '20150404',
672 'uploader_id': 'spbelect',
673 'uploader': 'Наблюдатели Петербурга',
674 },
675 'params': {
676 'skip_download': 'requires avconv',
677 },
678 'skip': 'This live event has ended.',
679 },
680 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
681 {
682 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
683 'info_dict': {
684 'id': 'FIl7x6_3R5Y',
685 'ext': 'mp4',
686 'title': 'md5:7b81415841e02ecd4313668cde88737a',
687 'description': 'md5:116377fd2963b81ec4ce64b542173306',
688 'duration': 220,
689 'upload_date': '20150625',
690 'uploader_id': 'dorappi2000',
691 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
692 'uploader': 'dorappi2000',
693 'license': 'Standard YouTube License',
694 'formats': 'mincount:32',
695 },
696 },
697 # DASH manifest with segment_list
698 {
699 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
700 'md5': '8ce563a1d667b599d21064e982ab9e31',
701 'info_dict': {
702 'id': 'CsmdDsKjzN8',
703 'ext': 'mp4',
704 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
705 'uploader': 'Airtek',
706 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
707 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
708 'license': 'Standard YouTube License',
709 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
710 },
711 'params': {
712 'youtube_include_dash_manifest': True,
713 'format': '135', # bestvideo
714 },
715 'skip': 'This live event has ended.',
716 },
717 {
718 # Multifeed videos (multiple cameras), URL is for Main Camera
719 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
720 'info_dict': {
721 'id': 'jqWvoWXjCVs',
722 'title': 'teamPGP: Rocket League Noob Stream',
723 'description': 'md5:dc7872fb300e143831327f1bae3af010',
724 },
725 'playlist': [{
726 'info_dict': {
727 'id': 'jqWvoWXjCVs',
728 'ext': 'mp4',
729 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
730 'description': 'md5:dc7872fb300e143831327f1bae3af010',
731 'duration': 7335,
732 'upload_date': '20150721',
733 'uploader': 'Beer Games Beer',
734 'uploader_id': 'beergamesbeer',
735 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
736 'license': 'Standard YouTube License',
737 },
738 }, {
739 'info_dict': {
740 'id': '6h8e8xoXJzg',
741 'ext': 'mp4',
742 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
743 'description': 'md5:dc7872fb300e143831327f1bae3af010',
744 'duration': 7337,
745 'upload_date': '20150721',
746 'uploader': 'Beer Games Beer',
747 'uploader_id': 'beergamesbeer',
748 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
749 'license': 'Standard YouTube License',
750 },
751 }, {
752 'info_dict': {
753 'id': 'PUOgX5z9xZw',
754 'ext': 'mp4',
755 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
756 'description': 'md5:dc7872fb300e143831327f1bae3af010',
757 'duration': 7337,
758 'upload_date': '20150721',
759 'uploader': 'Beer Games Beer',
760 'uploader_id': 'beergamesbeer',
761 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
762 'license': 'Standard YouTube License',
763 },
764 }, {
765 'info_dict': {
766 'id': 'teuwxikvS5k',
767 'ext': 'mp4',
768 'title': 'teamPGP: Rocket League Noob Stream (zim)',
769 'description': 'md5:dc7872fb300e143831327f1bae3af010',
770 'duration': 7334,
771 'upload_date': '20150721',
772 'uploader': 'Beer Games Beer',
773 'uploader_id': 'beergamesbeer',
774 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
775 'license': 'Standard YouTube License',
776 },
777 }],
778 'params': {
779 'skip_download': True,
780 },
781 },
782 {
783 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
784 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
785 'info_dict': {
786 'id': 'gVfLd0zydlo',
787 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
788 },
789 'playlist_count': 2,
790 'skip': 'Not multifeed anymore',
791 },
792 {
793 'url': 'https://vid.plus/FlRa-iH7PGw',
794 'only_matching': True,
795 },
796 {
797 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
798 'only_matching': True,
799 },
800 {
801 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
802 # Also tests cut-off URL expansion in video description (see
803 # https://github.com/rg3/youtube-dl/issues/1892,
804 # https://github.com/rg3/youtube-dl/issues/8164)
805 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
806 'info_dict': {
807 'id': 'lsguqyKfVQg',
808 'ext': 'mp4',
809 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
810 'alt_title': 'Dark Walk',
811 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
812 'duration': 133,
813 'upload_date': '20151119',
814 'uploader_id': 'IronSoulElf',
815 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
816 'uploader': 'IronSoulElf',
817 'license': 'Standard YouTube License',
818 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
819 },
820 'params': {
821 'skip_download': True,
822 },
823 },
824 {
825 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
826 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
827 'only_matching': True,
828 },
829 {
830 # Video with yt:stretch=17:0
831 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
832 'info_dict': {
833 'id': 'Q39EVAstoRM',
834 'ext': 'mp4',
835 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
836 'description': 'md5:ee18a25c350637c8faff806845bddee9',
837 'upload_date': '20151107',
838 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
839 'uploader': 'CH GAMER DROID',
840 },
841 'params': {
842 'skip_download': True,
843 },
844 'skip': 'This video does not exist.',
845 },
846 {
847 # Video licensed under Creative Commons
848 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
849 'info_dict': {
850 'id': 'M4gD1WSo5mA',
851 'ext': 'mp4',
852 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
853 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
854 'duration': 721,
855 'upload_date': '20150127',
856 'uploader_id': 'BerkmanCenter',
857 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
858 'uploader': 'The Berkman Klein Center for Internet & Society',
859 'license': 'Creative Commons Attribution license (reuse allowed)',
860 },
861 'params': {
862 'skip_download': True,
863 },
864 },
865 {
866 # Channel-like uploader_url
867 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
868 'info_dict': {
869 'id': 'eQcmzGIKrzg',
870 'ext': 'mp4',
871 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
872 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
873 'duration': 4060,
874 'upload_date': '20151119',
875 'uploader': 'Bernie 2016',
876 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
877 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
878 'license': 'Creative Commons Attribution license (reuse allowed)',
879 },
880 'params': {
881 'skip_download': True,
882 },
883 },
884 {
885 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
886 'only_matching': True,
887 },
888 {
889 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
890 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
891 'only_matching': True,
892 },
893 {
894 # Rental video preview
895 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
896 'info_dict': {
897 'id': 'uGpuVWrhIzE',
898 'ext': 'mp4',
899 'title': 'Piku - Trailer',
900 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
901 'upload_date': '20150811',
902 'uploader': 'FlixMatrix',
903 'uploader_id': 'FlixMatrixKaravan',
904 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
905 'license': 'Standard YouTube License',
906 },
907 'params': {
908 'skip_download': True,
909 },
910 },
911 {
912 # YouTube Red video with episode data
913 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
914 'info_dict': {
915 'id': 'iqKdEhx-dD4',
916 'ext': 'mp4',
917 'title': 'Isolation - Mind Field (Ep 1)',
918 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
919 'duration': 2085,
920 'upload_date': '20170118',
921 'uploader': 'Vsauce',
922 'uploader_id': 'Vsauce',
923 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
924 'license': 'Standard YouTube License',
925 'series': 'Mind Field',
926 'season_number': 1,
927 'episode_number': 1,
928 },
929 'params': {
930 'skip_download': True,
931 },
932 'expected_warnings': [
933 'Skipping DASH manifest',
934 ],
935 },
936 {
937 # itag 212
938 'url': '1t24XAntNCY',
939 'only_matching': True,
940 },
941 {
942 # geo restricted to JP
943 'url': 'sJL6WA-aGkQ',
944 'only_matching': True,
945 },
946 {
947 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
948 'only_matching': True,
949 },
950 ]
951
952 def __init__(self, *args, **kwargs):
953 super(YoutubeIE, self).__init__(*args, **kwargs)
954 self._player_cache = {}
955
956 def report_video_info_webpage_download(self, video_id):
957 """Report attempt to download video info webpage."""
958 self.to_screen('%s: Downloading video info webpage' % video_id)
959
960 def report_information_extraction(self, video_id):
961 """Report attempt to extract video information."""
962 self.to_screen('%s: Extracting video information' % video_id)
963
964 def report_unavailable_format(self, video_id, format):
965 """Report extracted video URL."""
966 self.to_screen('%s: Format %s not available' % (video_id, format))
967
968 def report_rtmp_download(self):
969 """Indicate the download will use the RTMP protocol."""
970 self.to_screen('RTMP download detected')
971
972 def _signature_cache_id(self, example_sig):
973 """ Return a string representation of a signature """
974 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
975
976 def _extract_signature_function(self, video_id, player_url, example_sig):
977 id_m = re.match(
978 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
979 player_url)
980 if not id_m:
981 raise ExtractorError('Cannot identify player %r' % player_url)
982 player_type = id_m.group('ext')
983 player_id = id_m.group('id')
984
985 # Read from filesystem cache
986 func_id = '%s_%s_%s' % (
987 player_type, player_id, self._signature_cache_id(example_sig))
988 assert os.path.basename(func_id) == func_id
989
990 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
991 if cache_spec is not None:
992 return lambda s: ''.join(s[i] for i in cache_spec)
993
994 download_note = (
995 'Downloading player %s' % player_url
996 if self._downloader.params.get('verbose') else
997 'Downloading %s player %s' % (player_type, player_id)
998 )
999 if player_type == 'js':
1000 code = self._download_webpage(
1001 player_url, video_id,
1002 note=download_note,
1003 errnote='Download of %s failed' % player_url)
1004 res = self._parse_sig_js(code)
1005 elif player_type == 'swf':
1006 urlh = self._request_webpage(
1007 player_url, video_id,
1008 note=download_note,
1009 errnote='Download of %s failed' % player_url)
1010 code = urlh.read()
1011 res = self._parse_sig_swf(code)
1012 else:
1013 assert False, 'Invalid player type %r' % player_type
1014
1015 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1016 cache_res = res(test_string)
1017 cache_spec = [ord(c) for c in cache_res]
1018
1019 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1020 return res
1021
1022 def _print_sig_code(self, func, example_sig):
1023 def gen_sig_code(idxs):
1024 def _genslice(start, end, step):
1025 starts = '' if start == 0 else str(start)
1026 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1027 steps = '' if step == 1 else (':%d' % step)
1028 return 's[%s%s%s]' % (starts, ends, steps)
1029
1030 step = None
1031 # Quelch pyflakes warnings - start will be set when step is set
1032 start = '(Never used)'
1033 for i, prev in zip(idxs[1:], idxs[:-1]):
1034 if step is not None:
1035 if i - prev == step:
1036 continue
1037 yield _genslice(start, prev, step)
1038 step = None
1039 continue
1040 if i - prev in [-1, 1]:
1041 step = i - prev
1042 start = prev
1043 continue
1044 else:
1045 yield 's[%d]' % prev
1046 if step is None:
1047 yield 's[%d]' % i
1048 else:
1049 yield _genslice(start, i, step)
1050
1051 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1052 cache_res = func(test_string)
1053 cache_spec = [ord(c) for c in cache_res]
1054 expr_code = ' + '.join(gen_sig_code(cache_spec))
1055 signature_id_tuple = '(%s)' % (
1056 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1057 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1058 ' return %s\n') % (signature_id_tuple, expr_code)
1059 self.to_screen('Extracted signature function:\n' + code)
1060
1061 def _parse_sig_js(self, jscode):
1062 funcname = self._search_regex(
1063 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1064 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1065 jscode, 'Initial JS player signature function name', group='sig')
1066
1067 jsi = JSInterpreter(jscode)
1068 initial_function = jsi.extract_function(funcname)
1069 return lambda s: initial_function([s])
1070
1071 def _parse_sig_swf(self, file_contents):
1072 swfi = SWFInterpreter(file_contents)
1073 TARGET_CLASSNAME = 'SignatureDecipher'
1074 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1075 initial_function = swfi.extract_function(searched_class, 'decipher')
1076 return lambda s: initial_function([s])
1077
1078 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1079 """Turn the encrypted s field into a working signature"""
1080
1081 if player_url is None:
1082 raise ExtractorError('Cannot decrypt signature without player_url')
1083
1084 if player_url.startswith('//'):
1085 player_url = 'https:' + player_url
1086 elif not re.match(r'https?://', player_url):
1087 player_url = compat_urlparse.urljoin(
1088 'https://www.youtube.com', player_url)
1089 try:
1090 player_id = (player_url, self._signature_cache_id(s))
1091 if player_id not in self._player_cache:
1092 func = self._extract_signature_function(
1093 video_id, player_url, s
1094 )
1095 self._player_cache[player_id] = func
1096 func = self._player_cache[player_id]
1097 if self._downloader.params.get('youtube_print_sig_code'):
1098 self._print_sig_code(func, s)
1099 return func(s)
1100 except Exception as e:
1101 tb = traceback.format_exc()
1102 raise ExtractorError(
1103 'Signature extraction failed: ' + tb, cause=e)
1104
1105 def _get_subtitles(self, video_id, webpage):
1106 try:
1107 subs_doc = self._download_xml(
1108 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1109 video_id, note=False)
1110 except ExtractorError as err:
1111 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1112 return {}
1113
1114 sub_lang_list = {}
1115 for track in subs_doc.findall('track'):
1116 lang = track.attrib['lang_code']
1117 if lang in sub_lang_list:
1118 continue
1119 sub_formats = []
1120 for ext in self._SUBTITLE_FORMATS:
1121 params = compat_urllib_parse_urlencode({
1122 'lang': lang,
1123 'v': video_id,
1124 'fmt': ext,
1125 'name': track.attrib['name'].encode('utf-8'),
1126 })
1127 sub_formats.append({
1128 'url': 'https://www.youtube.com/api/timedtext?' + params,
1129 'ext': ext,
1130 })
1131 sub_lang_list[lang] = sub_formats
1132 if not sub_lang_list:
1133 self._downloader.report_warning('video doesn\'t have subtitles')
1134 return {}
1135 return sub_lang_list
1136
1137 def _get_ytplayer_config(self, video_id, webpage):
1138 patterns = (
1139 # User data may contain arbitrary character sequences that may affect
1140 # JSON extraction with regex, e.g. when '};' is contained the second
1141 # regex won't capture the whole JSON. Yet working around by trying more
1142 # concrete regex first keeping in mind proper quoted string handling
1143 # to be implemented in future that will replace this workaround (see
1144 # https://github.com/rg3/youtube-dl/issues/7468,
1145 # https://github.com/rg3/youtube-dl/pull/7599)
1146 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1147 r';ytplayer\.config\s*=\s*({.+?});',
1148 )
1149 config = self._search_regex(
1150 patterns, webpage, 'ytplayer.config', default=None)
1151 if config:
1152 return self._parse_json(
1153 uppercase_escape(config), video_id, fatal=False)
1154
1155 def _get_automatic_captions(self, video_id, webpage):
1156 """We need the webpage for getting the captions url, pass it as an
1157 argument to speed up the process."""
1158 self.to_screen('%s: Looking for automatic captions' % video_id)
1159 player_config = self._get_ytplayer_config(video_id, webpage)
1160 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1161 if not player_config:
1162 self._downloader.report_warning(err_msg)
1163 return {}
1164 try:
1165 args = player_config['args']
1166 caption_url = args.get('ttsurl')
1167 if caption_url:
1168 timestamp = args['timestamp']
1169 # We get the available subtitles
1170 list_params = compat_urllib_parse_urlencode({
1171 'type': 'list',
1172 'tlangs': 1,
1173 'asrs': 1,
1174 })
1175 list_url = caption_url + '&' + list_params
1176 caption_list = self._download_xml(list_url, video_id)
1177 original_lang_node = caption_list.find('track')
1178 if original_lang_node is None:
1179 self._downloader.report_warning('Video doesn\'t have automatic captions')
1180 return {}
1181 original_lang = original_lang_node.attrib['lang_code']
1182 caption_kind = original_lang_node.attrib.get('kind', '')
1183
1184 sub_lang_list = {}
1185 for lang_node in caption_list.findall('target'):
1186 sub_lang = lang_node.attrib['lang_code']
1187 sub_formats = []
1188 for ext in self._SUBTITLE_FORMATS:
1189 params = compat_urllib_parse_urlencode({
1190 'lang': original_lang,
1191 'tlang': sub_lang,
1192 'fmt': ext,
1193 'ts': timestamp,
1194 'kind': caption_kind,
1195 })
1196 sub_formats.append({
1197 'url': caption_url + '&' + params,
1198 'ext': ext,
1199 })
1200 sub_lang_list[sub_lang] = sub_formats
1201 return sub_lang_list
1202
1203 # Some videos don't provide ttsurl but rather caption_tracks and
1204 # caption_translation_languages (e.g. 20LmZk1hakA)
1205 caption_tracks = args['caption_tracks']
1206 caption_translation_languages = args['caption_translation_languages']
1207 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1208 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1209 caption_qs = compat_parse_qs(parsed_caption_url.query)
1210
1211 sub_lang_list = {}
1212 for lang in caption_translation_languages.split(','):
1213 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1214 sub_lang = lang_qs.get('lc', [None])[0]
1215 if not sub_lang:
1216 continue
1217 sub_formats = []
1218 for ext in self._SUBTITLE_FORMATS:
1219 caption_qs.update({
1220 'tlang': [sub_lang],
1221 'fmt': [ext],
1222 })
1223 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1224 query=compat_urllib_parse_urlencode(caption_qs, True)))
1225 sub_formats.append({
1226 'url': sub_url,
1227 'ext': ext,
1228 })
1229 sub_lang_list[sub_lang] = sub_formats
1230 return sub_lang_list
1231 # An extractor error can be raise by the download process if there are
1232 # no automatic captions but there are subtitles
1233 except (KeyError, ExtractorError):
1234 self._downloader.report_warning(err_msg)
1235 return {}
1236
1237 def _mark_watched(self, video_id, video_info):
1238 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1239 if not playback_url:
1240 return
1241 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1242 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1243
1244 # cpn generation algorithm is reverse engineered from base.js.
1245 # In fact it works even with dummy cpn.
1246 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1247 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1248
1249 qs.update({
1250 'ver': ['2'],
1251 'cpn': [cpn],
1252 })
1253 playback_url = compat_urlparse.urlunparse(
1254 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1255
1256 self._download_webpage(
1257 playback_url, video_id, 'Marking watched',
1258 'Unable to mark watched', fatal=False)
1259
1260 @classmethod
1261 def extract_id(cls, url):
1262 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1263 if mobj is None:
1264 raise ExtractorError('Invalid URL: %s' % url)
1265 video_id = mobj.group(2)
1266 return video_id
1267
1268 def _extract_annotations(self, video_id):
1269 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1270 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1271
1272 @staticmethod
1273 def _extract_chapters(description, duration):
1274 if not description:
1275 return None
1276 chapter_lines = re.findall(
1277 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1278 description)
1279 if not chapter_lines:
1280 return None
1281 chapters = []
1282 for next_num, (chapter_line, time_point) in enumerate(
1283 chapter_lines, start=1):
1284 start_time = parse_duration(time_point)
1285 if start_time is None:
1286 continue
1287 end_time = (duration if next_num == len(chapter_lines)
1288 else parse_duration(chapter_lines[next_num][1]))
1289 if end_time is None:
1290 continue
1291 chapter_title = re.sub(
1292 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1293 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1294 chapters.append({
1295 'start_time': start_time,
1296 'end_time': end_time,
1297 'title': chapter_title,
1298 })
1299 return chapters
1300
1301 def _real_extract(self, url):
1302 url, smuggled_data = unsmuggle_url(url, {})
1303
1304 proto = (
1305 'http' if self._downloader.params.get('prefer_insecure', False)
1306 else 'https')
1307
1308 start_time = None
1309 end_time = None
1310 parsed_url = compat_urllib_parse_urlparse(url)
1311 for component in [parsed_url.fragment, parsed_url.query]:
1312 query = compat_parse_qs(component)
1313 if start_time is None and 't' in query:
1314 start_time = parse_duration(query['t'][0])
1315 if start_time is None and 'start' in query:
1316 start_time = parse_duration(query['start'][0])
1317 if end_time is None and 'end' in query:
1318 end_time = parse_duration(query['end'][0])
1319
1320 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1321 mobj = re.search(self._NEXT_URL_RE, url)
1322 if mobj:
1323 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1324 video_id = self.extract_id(url)
1325
1326 # Get video webpage
1327 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1328 video_webpage = self._download_webpage(url, video_id)
1329
1330 # Attempt to extract SWF player URL
1331 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1332 if mobj is not None:
1333 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1334 else:
1335 player_url = None
1336
1337 dash_mpds = []
1338
1339 def add_dash_mpd(video_info):
1340 dash_mpd = video_info.get('dashmpd')
1341 if dash_mpd and dash_mpd[0] not in dash_mpds:
1342 dash_mpds.append(dash_mpd[0])
1343
1344 # Get video info
1345 embed_webpage = None
1346 is_live = None
1347 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1348 age_gate = True
1349 # We simulate the access to the video from www.youtube.com/v/{video_id}
1350 # this can be viewed without login into Youtube
1351 url = proto + '://www.youtube.com/embed/%s' % video_id
1352 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1353 data = compat_urllib_parse_urlencode({
1354 'video_id': video_id,
1355 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1356 'sts': self._search_regex(
1357 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1358 })
1359 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1360 video_info_webpage = self._download_webpage(
1361 video_info_url, video_id,
1362 note='Refetching age-gated info webpage',
1363 errnote='unable to download video info webpage')
1364 video_info = compat_parse_qs(video_info_webpage)
1365 add_dash_mpd(video_info)
1366 else:
1367 age_gate = False
1368 video_info = None
1369 # Try looking directly into the video webpage
1370 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1371 if ytplayer_config:
1372 args = ytplayer_config['args']
1373 if args.get('url_encoded_fmt_stream_map'):
1374 # Convert to the same format returned by compat_parse_qs
1375 video_info = dict((k, [v]) for k, v in args.items())
1376 add_dash_mpd(video_info)
1377 # Rental video is not rented but preview is available (e.g.
1378 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1379 # https://github.com/rg3/youtube-dl/issues/10532)
1380 if not video_info and args.get('ypc_vid'):
1381 return self.url_result(
1382 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1383 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1384 is_live = True
1385 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1386 # We also try looking in get_video_info since it may contain different dashmpd
1387 # URL that points to a DASH manifest with possibly different itag set (some itags
1388 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1389 # manifest pointed by get_video_info's dashmpd).
1390 # The general idea is to take a union of itags of both DASH manifests (for example
1391 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1392 self.report_video_info_webpage_download(video_id)
1393 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1394 video_info_url = (
1395 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1396 % (proto, video_id, el_type))
1397 video_info_webpage = self._download_webpage(
1398 video_info_url,
1399 video_id, note=False,
1400 errnote='unable to download video info webpage')
1401 get_video_info = compat_parse_qs(video_info_webpage)
1402 if get_video_info.get('use_cipher_signature') != ['True']:
1403 add_dash_mpd(get_video_info)
1404 if not video_info:
1405 video_info = get_video_info
1406 if 'token' in get_video_info:
1407 # Different get_video_info requests may report different results, e.g.
1408 # some may report video unavailability, but some may serve it without
1409 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1410 # the original webpage as well as el=info and el=embedded get_video_info
1411 # requests report video unavailability due to geo restriction while
1412 # el=detailpage succeeds and returns valid data). This is probably
1413 # due to YouTube measures against IP ranges of hosting providers.
1414 # Working around by preferring the first succeeded video_info containing
1415 # the token if no such video_info yet was found.
1416 if 'token' not in video_info:
1417 video_info = get_video_info
1418 break
1419 if 'token' not in video_info:
1420 if 'reason' in video_info:
1421 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1422 regions_allowed = self._html_search_meta(
1423 'regionsAllowed', video_webpage, default=None)
1424 countries = regions_allowed.split(',') if regions_allowed else None
1425 self.raise_geo_restricted(
1426 msg=video_info['reason'][0], countries=countries)
1427 raise ExtractorError(
1428 'YouTube said: %s' % video_info['reason'][0],
1429 expected=True, video_id=video_id)
1430 else:
1431 raise ExtractorError(
1432 '"token" parameter not in video info for unknown reason',
1433 video_id=video_id)
1434
1435 # title
1436 if 'title' in video_info:
1437 video_title = video_info['title'][0]
1438 else:
1439 self._downloader.report_warning('Unable to extract video title')
1440 video_title = '_'
1441
1442 # description
1443 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1444 if video_description:
1445 description_original = video_description = re.sub(r'''(?x)
1446 <a\s+
1447 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1448 (?:title|href)="([^"]+)"\s+
1449 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1450 class="[^"]*"[^>]*>
1451 [^<]+\.{3}\s*
1452 </a>
1453 ''', r'\1', video_description)
1454 video_description = clean_html(video_description)
1455 else:
1456 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1457 if fd_mobj:
1458 video_description = unescapeHTML(fd_mobj.group(1))
1459 else:
1460 video_description = ''
1461
1462 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1463 if not self._downloader.params.get('noplaylist'):
1464 entries = []
1465 feed_ids = []
1466 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1467 for feed in multifeed_metadata_list.split(','):
1468 # Unquote should take place before split on comma (,) since textual
1469 # fields may contain comma as well (see
1470 # https://github.com/rg3/youtube-dl/issues/8536)
1471 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1472 entries.append({
1473 '_type': 'url_transparent',
1474 'ie_key': 'Youtube',
1475 'url': smuggle_url(
1476 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1477 {'force_singlefeed': True}),
1478 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1479 })
1480 feed_ids.append(feed_data['id'][0])
1481 self.to_screen(
1482 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1483 % (', '.join(feed_ids), video_id))
1484 return self.playlist_result(entries, video_id, video_title, video_description)
1485 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1486
1487 if 'view_count' in video_info:
1488 view_count = int(video_info['view_count'][0])
1489 else:
1490 view_count = None
1491
1492 # Check for "rental" videos
1493 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1494 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1495
1496 # Start extracting information
1497 self.report_information_extraction(video_id)
1498
1499 # uploader
1500 if 'author' not in video_info:
1501 raise ExtractorError('Unable to extract uploader name')
1502 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1503
1504 # uploader_id
1505 video_uploader_id = None
1506 video_uploader_url = None
1507 mobj = re.search(
1508 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1509 video_webpage)
1510 if mobj is not None:
1511 video_uploader_id = mobj.group('uploader_id')
1512 video_uploader_url = mobj.group('uploader_url')
1513 else:
1514 self._downloader.report_warning('unable to extract uploader nickname')
1515
1516 # thumbnail image
1517 # We try first to get a high quality image:
1518 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1519 video_webpage, re.DOTALL)
1520 if m_thumb is not None:
1521 video_thumbnail = m_thumb.group(1)
1522 elif 'thumbnail_url' not in video_info:
1523 self._downloader.report_warning('unable to extract video thumbnail')
1524 video_thumbnail = None
1525 else: # don't panic if we can't find it
1526 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1527
1528 # upload date
1529 upload_date = self._html_search_meta(
1530 'datePublished', video_webpage, 'upload date', default=None)
1531 if not upload_date:
1532 upload_date = self._search_regex(
1533 [r'(?s)id="eow-date.*?>(.*?)</span>',
1534 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1535 video_webpage, 'upload date', default=None)
1536 if upload_date:
1537 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1538 upload_date = unified_strdate(upload_date)
1539
1540 video_license = self._html_search_regex(
1541 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1542 video_webpage, 'license', default=None)
1543
1544 m_music = re.search(
1545 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1546 video_webpage)
1547 if m_music:
1548 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1549 video_creator = clean_html(m_music.group('creator'))
1550 else:
1551 video_alt_title = video_creator = None
1552
1553 m_episode = re.search(
1554 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1555 video_webpage)
1556 if m_episode:
1557 series = m_episode.group('series')
1558 season_number = int(m_episode.group('season'))
1559 episode_number = int(m_episode.group('episode'))
1560 else:
1561 series = season_number = episode_number = None
1562
1563 m_cat_container = self._search_regex(
1564 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1565 video_webpage, 'categories', default=None)
1566 if m_cat_container:
1567 category = self._html_search_regex(
1568 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1569 default=None)
1570 video_categories = None if category is None else [category]
1571 else:
1572 video_categories = None
1573
1574 video_tags = [
1575 unescapeHTML(m.group('content'))
1576 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1577
1578 def _extract_count(count_name):
1579 return str_to_int(self._search_regex(
1580 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1581 % re.escape(count_name),
1582 video_webpage, count_name, default=None))
1583
1584 like_count = _extract_count('like')
1585 dislike_count = _extract_count('dislike')
1586
1587 # subtitles
1588 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1589 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1590
1591 video_duration = try_get(
1592 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1593 if not video_duration:
1594 video_duration = parse_duration(self._html_search_meta(
1595 'duration', video_webpage, 'video duration'))
1596
1597 # annotations
1598 video_annotations = None
1599 if self._downloader.params.get('writeannotations', False):
1600 video_annotations = self._extract_annotations(video_id)
1601
1602 chapters = self._extract_chapters(description_original, video_duration)
1603
1604 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1605 self.report_rtmp_download()
1606 formats = [{
1607 'format_id': '_rtmp',
1608 'protocol': 'rtmp',
1609 'url': video_info['conn'][0],
1610 'player_url': player_url,
1611 }]
1612 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1613 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1614 if 'rtmpe%3Dyes' in encoded_url_map:
1615 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1616 formats_spec = {}
1617 fmt_list = video_info.get('fmt_list', [''])[0]
1618 if fmt_list:
1619 for fmt in fmt_list.split(','):
1620 spec = fmt.split('/')
1621 if len(spec) > 1:
1622 width_height = spec[1].split('x')
1623 if len(width_height) == 2:
1624 formats_spec[spec[0]] = {
1625 'resolution': spec[1],
1626 'width': int_or_none(width_height[0]),
1627 'height': int_or_none(width_height[1]),
1628 }
1629 formats = []
1630 for url_data_str in encoded_url_map.split(','):
1631 url_data = compat_parse_qs(url_data_str)
1632 if 'itag' not in url_data or 'url' not in url_data:
1633 continue
1634 format_id = url_data['itag'][0]
1635 url = url_data['url'][0]
1636
1637 if 'sig' in url_data:
1638 url += '&signature=' + url_data['sig'][0]
1639 elif 's' in url_data:
1640 encrypted_sig = url_data['s'][0]
1641 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1642
1643 jsplayer_url_json = self._search_regex(
1644 ASSETS_RE,
1645 embed_webpage if age_gate else video_webpage,
1646 'JS player URL (1)', default=None)
1647 if not jsplayer_url_json and not age_gate:
1648 # We need the embed website after all
1649 if embed_webpage is None:
1650 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1651 embed_webpage = self._download_webpage(
1652 embed_url, video_id, 'Downloading embed webpage')
1653 jsplayer_url_json = self._search_regex(
1654 ASSETS_RE, embed_webpage, 'JS player URL')
1655
1656 player_url = json.loads(jsplayer_url_json)
1657 if player_url is None:
1658 player_url_json = self._search_regex(
1659 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1660 video_webpage, 'age gate player URL')
1661 player_url = json.loads(player_url_json)
1662
1663 if self._downloader.params.get('verbose'):
1664 if player_url is None:
1665 player_version = 'unknown'
1666 player_desc = 'unknown'
1667 else:
1668 if player_url.endswith('swf'):
1669 player_version = self._search_regex(
1670 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1671 'flash player', fatal=False)
1672 player_desc = 'flash player %s' % player_version
1673 else:
1674 player_version = self._search_regex(
1675 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1676 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1677 player_url,
1678 'html5 player', fatal=False)
1679 player_desc = 'html5 player %s' % player_version
1680
1681 parts_sizes = self._signature_cache_id(encrypted_sig)
1682 self.to_screen('{%s} signature length %s, %s' %
1683 (format_id, parts_sizes, player_desc))
1684
1685 signature = self._decrypt_signature(
1686 encrypted_sig, video_id, player_url, age_gate)
1687 url += '&signature=' + signature
1688 if 'ratebypass' not in url:
1689 url += '&ratebypass=yes'
1690
1691 dct = {
1692 'format_id': format_id,
1693 'url': url,
1694 'player_url': player_url,
1695 }
1696 if format_id in self._formats:
1697 dct.update(self._formats[format_id])
1698 if format_id in formats_spec:
1699 dct.update(formats_spec[format_id])
1700
1701 # Some itags are not included in DASH manifest thus corresponding formats will
1702 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1703 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1704 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1705 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1706
1707 more_fields = {
1708 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1709 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1710 'width': width,
1711 'height': height,
1712 'fps': int_or_none(url_data.get('fps', [None])[0]),
1713 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1714 }
1715 for key, value in more_fields.items():
1716 if value:
1717 dct[key] = value
1718 type_ = url_data.get('type', [None])[0]
1719 if type_:
1720 type_split = type_.split(';')
1721 kind_ext = type_split[0].split('/')
1722 if len(kind_ext) == 2:
1723 kind, _ = kind_ext
1724 dct['ext'] = mimetype2ext(type_split[0])
1725 if kind in ('audio', 'video'):
1726 codecs = None
1727 for mobj in re.finditer(
1728 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1729 if mobj.group('key') == 'codecs':
1730 codecs = mobj.group('val')
1731 break
1732 if codecs:
1733 dct.update(parse_codecs(codecs))
1734 formats.append(dct)
1735 elif video_info.get('hlsvp'):
1736 manifest_url = video_info['hlsvp'][0]
1737 formats = []
1738 m3u8_formats = self._extract_m3u8_formats(
1739 manifest_url, video_id, 'mp4', fatal=False)
1740 for a_format in m3u8_formats:
1741 itag = self._search_regex(
1742 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1743 if itag:
1744 a_format['format_id'] = itag
1745 if itag in self._formats:
1746 dct = self._formats[itag].copy()
1747 dct.update(a_format)
1748 a_format = dct
1749 a_format['player_url'] = player_url
1750 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1751 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1752 formats.append(a_format)
1753 else:
1754 unavailable_message = self._html_search_regex(
1755 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1756 video_webpage, 'unavailable message', default=None)
1757 if unavailable_message:
1758 raise ExtractorError(unavailable_message, expected=True)
1759 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1760
1761 # Look for the DASH manifest
1762 if self._downloader.params.get('youtube_include_dash_manifest', True):
1763 dash_mpd_fatal = True
1764 for mpd_url in dash_mpds:
1765 dash_formats = {}
1766 try:
1767 def decrypt_sig(mobj):
1768 s = mobj.group(1)
1769 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1770 return '/signature/%s' % dec_s
1771
1772 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1773
1774 for df in self._extract_mpd_formats(
1775 mpd_url, video_id, fatal=dash_mpd_fatal,
1776 formats_dict=self._formats):
1777 # Do not overwrite DASH format found in some previous DASH manifest
1778 if df['format_id'] not in dash_formats:
1779 dash_formats[df['format_id']] = df
1780 # Additional DASH manifests may end up in HTTP Error 403 therefore
1781 # allow them to fail without bug report message if we already have
1782 # some DASH manifest succeeded. This is temporary workaround to reduce
1783 # burst of bug reports until we figure out the reason and whether it
1784 # can be fixed at all.
1785 dash_mpd_fatal = False
1786 except (ExtractorError, KeyError) as e:
1787 self.report_warning(
1788 'Skipping DASH manifest: %r' % e, video_id)
1789 if dash_formats:
1790 # Remove the formats we found through non-DASH, they
1791 # contain less info and it can be wrong, because we use
1792 # fixed values (for example the resolution). See
1793 # https://github.com/rg3/youtube-dl/issues/5774 for an
1794 # example.
1795 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1796 formats.extend(dash_formats.values())
1797
1798 # Check for malformed aspect ratio
1799 stretched_m = re.search(
1800 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1801 video_webpage)
1802 if stretched_m:
1803 w = float(stretched_m.group('w'))
1804 h = float(stretched_m.group('h'))
1805 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1806 # We will only process correct ratios.
1807 if w > 0 and h > 0:
1808 ratio = w / h
1809 for f in formats:
1810 if f.get('vcodec') != 'none':
1811 f['stretched_ratio'] = ratio
1812
1813 self._sort_formats(formats)
1814
1815 self.mark_watched(video_id, video_info)
1816
1817 return {
1818 'id': video_id,
1819 'uploader': video_uploader,
1820 'uploader_id': video_uploader_id,
1821 'uploader_url': video_uploader_url,
1822 'upload_date': upload_date,
1823 'license': video_license,
1824 'creator': video_creator,
1825 'title': video_title,
1826 'alt_title': video_alt_title,
1827 'thumbnail': video_thumbnail,
1828 'description': video_description,
1829 'categories': video_categories,
1830 'tags': video_tags,
1831 'subtitles': video_subtitles,
1832 'automatic_captions': automatic_captions,
1833 'duration': video_duration,
1834 'age_limit': 18 if age_gate else 0,
1835 'annotations': video_annotations,
1836 'chapters': chapters,
1837 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1838 'view_count': view_count,
1839 'like_count': like_count,
1840 'dislike_count': dislike_count,
1841 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1842 'formats': formats,
1843 'is_live': is_live,
1844 'start_time': start_time,
1845 'end_time': end_time,
1846 'series': series,
1847 'season_number': season_number,
1848 'episode_number': episode_number,
1849 }
1850
1851
1852 class YoutubeSharedVideoIE(InfoExtractor):
1853 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1854 IE_NAME = 'youtube:shared'
1855
1856 _TEST = {
1857 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1858 'info_dict': {
1859 'id': 'uPDB5I9wfp8',
1860 'ext': 'webm',
1861 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1862 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1863 'upload_date': '20160219',
1864 'uploader': 'Pocoyo - Português (BR)',
1865 'uploader_id': 'PocoyoBrazil',
1866 },
1867 'add_ie': ['Youtube'],
1868 'params': {
1869 # There are already too many Youtube downloads
1870 'skip_download': True,
1871 },
1872 }
1873
1874 def _real_extract(self, url):
1875 video_id = self._match_id(url)
1876
1877 webpage = self._download_webpage(url, video_id)
1878
1879 real_video_id = self._html_search_meta(
1880 'videoId', webpage, 'YouTube video id', fatal=True)
1881
1882 return self.url_result(real_video_id, YoutubeIE.ie_key())
1883
1884
1885 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1886 IE_DESC = 'YouTube.com playlists'
1887 _VALID_URL = r"""(?x)(?:
1888 (?:https?://)?
1889 (?:\w+\.)?
1890 (?:
1891 youtube\.com/
1892 (?:
1893 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
1894 \? (?:.*?[&;])*? (?:p|a|list)=
1895 | p/
1896 )|
1897 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
1898 )
1899 (
1900 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
1901 # Top tracks, they can also include dots
1902 |(?:MC)[\w\.]*
1903 )
1904 .*
1905 |
1906 (%(playlist_id)s)
1907 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
1908 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
1909 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1910 IE_NAME = 'youtube:playlist'
1911 _TESTS = [{
1912 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1913 'info_dict': {
1914 'title': 'ytdl test PL',
1915 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1916 },
1917 'playlist_count': 3,
1918 }, {
1919 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1920 'info_dict': {
1921 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1922 'title': 'YDL_Empty_List',
1923 },
1924 'playlist_count': 0,
1925 'skip': 'This playlist is private',
1926 }, {
1927 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1928 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1929 'info_dict': {
1930 'title': '29C3: Not my department',
1931 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1932 },
1933 'playlist_count': 95,
1934 }, {
1935 'note': 'issue #673',
1936 'url': 'PLBB231211A4F62143',
1937 'info_dict': {
1938 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1939 'id': 'PLBB231211A4F62143',
1940 },
1941 'playlist_mincount': 26,
1942 }, {
1943 'note': 'Large playlist',
1944 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1945 'info_dict': {
1946 'title': 'Uploads from Cauchemar',
1947 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1948 },
1949 'playlist_mincount': 799,
1950 }, {
1951 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1952 'info_dict': {
1953 'title': 'YDL_safe_search',
1954 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1955 },
1956 'playlist_count': 2,
1957 'skip': 'This playlist is private',
1958 }, {
1959 'note': 'embedded',
1960 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1961 'playlist_count': 4,
1962 'info_dict': {
1963 'title': 'JODA15',
1964 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1965 }
1966 }, {
1967 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
1968 'playlist_mincount': 485,
1969 'info_dict': {
1970 'title': '2017 華語最新單曲 (2/24更新)',
1971 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
1972 }
1973 }, {
1974 'note': 'Embedded SWF player',
1975 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1976 'playlist_count': 4,
1977 'info_dict': {
1978 'title': 'JODA7',
1979 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1980 }
1981 }, {
1982 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1983 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1984 'info_dict': {
1985 'title': 'Uploads from Interstellar Movie',
1986 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1987 },
1988 'playlist_mincount': 21,
1989 }, {
1990 # Playlist URL that does not actually serve a playlist
1991 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
1992 'info_dict': {
1993 'id': 'FqZTN594JQw',
1994 'ext': 'webm',
1995 'title': "Smiley's People 01 detective, Adventure Series, Action",
1996 'uploader': 'STREEM',
1997 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
1998 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
1999 'upload_date': '20150526',
2000 'license': 'Standard YouTube License',
2001 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2002 'categories': ['People & Blogs'],
2003 'tags': list,
2004 'like_count': int,
2005 'dislike_count': int,
2006 },
2007 'params': {
2008 'skip_download': True,
2009 },
2010 'add_ie': [YoutubeIE.ie_key()],
2011 }, {
2012 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2013 'info_dict': {
2014 'id': 'yeWKywCrFtk',
2015 'ext': 'mp4',
2016 'title': 'Small Scale Baler and Braiding Rugs',
2017 'uploader': 'Backus-Page House Museum',
2018 'uploader_id': 'backuspagemuseum',
2019 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2020 'upload_date': '20161008',
2021 'license': 'Standard YouTube License',
2022 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2023 'categories': ['Nonprofits & Activism'],
2024 'tags': list,
2025 'like_count': int,
2026 'dislike_count': int,
2027 },
2028 'params': {
2029 'noplaylist': True,
2030 'skip_download': True,
2031 },
2032 }, {
2033 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2034 'only_matching': True,
2035 }, {
2036 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2037 'only_matching': True,
2038 }]
2039
2040 def _real_initialize(self):
2041 self._login()
2042
2043 def _extract_mix(self, playlist_id):
2044 # The mixes are generated from a single video
2045 # the id of the playlist is just 'RD' + video_id
2046 ids = []
2047 last_id = playlist_id[-11:]
2048 for n in itertools.count(1):
2049 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2050 webpage = self._download_webpage(
2051 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2052 new_ids = orderedSet(re.findall(
2053 r'''(?xs)data-video-username=".*?".*?
2054 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2055 webpage))
2056 # Fetch new pages until all the videos are repeated, it seems that
2057 # there are always 51 unique videos.
2058 new_ids = [_id for _id in new_ids if _id not in ids]
2059 if not new_ids:
2060 break
2061 ids.extend(new_ids)
2062 last_id = ids[-1]
2063
2064 url_results = self._ids_to_results(ids)
2065
2066 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2067 title_span = (
2068 search_title('playlist-title') or
2069 search_title('title long-title') or
2070 search_title('title'))
2071 title = clean_html(title_span)
2072
2073 return self.playlist_result(url_results, playlist_id, title)
2074
2075 def _extract_playlist(self, playlist_id):
2076 url = self._TEMPLATE_URL % playlist_id
2077 page = self._download_webpage(url, playlist_id)
2078
2079 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2080 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2081 match = match.strip()
2082 # Check if the playlist exists or is private
2083 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2084 if mobj:
2085 reason = mobj.group('reason')
2086 message = 'This playlist %s' % reason
2087 if 'private' in reason:
2088 message += ', use --username or --netrc to access it'
2089 message += '.'
2090 raise ExtractorError(message, expected=True)
2091 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2092 raise ExtractorError(
2093 'Invalid parameters. Maybe URL is incorrect.',
2094 expected=True)
2095 elif re.match(r'[^<]*Choose your language[^<]*', match):
2096 continue
2097 else:
2098 self.report_warning('Youtube gives an alert message: ' + match)
2099
2100 playlist_title = self._html_search_regex(
2101 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2102 page, 'title', default=None)
2103
2104 has_videos = True
2105
2106 if not playlist_title:
2107 try:
2108 # Some playlist URLs don't actually serve a playlist (e.g.
2109 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2110 next(self._entries(page, playlist_id))
2111 except StopIteration:
2112 has_videos = False
2113
2114 return has_videos, self.playlist_result(
2115 self._entries(page, playlist_id), playlist_id, playlist_title)
2116
2117 def _check_download_just_video(self, url, playlist_id):
2118 # Check if it's a video-specific URL
2119 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2120 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2121 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2122 'video id', default=None)
2123 if video_id:
2124 if self._downloader.params.get('noplaylist'):
2125 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2126 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2127 else:
2128 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2129 return video_id, None
2130 return None, None
2131
2132 def _real_extract(self, url):
2133 # Extract playlist id
2134 mobj = re.match(self._VALID_URL, url)
2135 if mobj is None:
2136 raise ExtractorError('Invalid URL: %s' % url)
2137 playlist_id = mobj.group(1) or mobj.group(2)
2138
2139 video_id, video = self._check_download_just_video(url, playlist_id)
2140 if video:
2141 return video
2142
2143 if playlist_id.startswith(('RD', 'UL', 'PU')):
2144 # Mixes require a custom extraction process
2145 return self._extract_mix(playlist_id)
2146
2147 has_videos, playlist = self._extract_playlist(playlist_id)
2148 if has_videos or not video_id:
2149 return playlist
2150
2151 # Some playlist URLs don't actually serve a playlist (see
2152 # https://github.com/rg3/youtube-dl/issues/10537).
2153 # Fallback to plain video extraction if there is a video id
2154 # along with playlist id.
2155 return self.url_result(video_id, 'Youtube', video_id=video_id)
2156
2157
2158 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2159 IE_DESC = 'YouTube.com channels'
2160 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2161 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2162 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2163 IE_NAME = 'youtube:channel'
2164 _TESTS = [{
2165 'note': 'paginated channel',
2166 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2167 'playlist_mincount': 91,
2168 'info_dict': {
2169 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2170 'title': 'Uploads from lex will',
2171 }
2172 }, {
2173 'note': 'Age restricted channel',
2174 # from https://www.youtube.com/user/DeusExOfficial
2175 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2176 'playlist_mincount': 64,
2177 'info_dict': {
2178 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2179 'title': 'Uploads from Deus Ex',
2180 },
2181 }]
2182
2183 @classmethod
2184 def suitable(cls, url):
2185 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2186 else super(YoutubeChannelIE, cls).suitable(url))
2187
2188 def _build_template_url(self, url, channel_id):
2189 return self._TEMPLATE_URL % channel_id
2190
2191 def _real_extract(self, url):
2192 channel_id = self._match_id(url)
2193
2194 url = self._build_template_url(url, channel_id)
2195
2196 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2197 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2198 # otherwise fallback on channel by page extraction
2199 channel_page = self._download_webpage(
2200 url + '?view=57', channel_id,
2201 'Downloading channel page', fatal=False)
2202 if channel_page is False:
2203 channel_playlist_id = False
2204 else:
2205 channel_playlist_id = self._html_search_meta(
2206 'channelId', channel_page, 'channel id', default=None)
2207 if not channel_playlist_id:
2208 channel_url = self._html_search_meta(
2209 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2210 channel_page, 'channel url', default=None)
2211 if channel_url:
2212 channel_playlist_id = self._search_regex(
2213 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2214 channel_url, 'channel id', default=None)
2215 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2216 playlist_id = 'UU' + channel_playlist_id[2:]
2217 return self.url_result(
2218 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2219
2220 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2221 autogenerated = re.search(r'''(?x)
2222 class="[^"]*?(?:
2223 channel-header-autogenerated-label|
2224 yt-channel-title-autogenerated
2225 )[^"]*"''', channel_page) is not None
2226
2227 if autogenerated:
2228 # The videos are contained in a single page
2229 # the ajax pages can't be used, they are empty
2230 entries = [
2231 self.url_result(
2232 video_id, 'Youtube', video_id=video_id,
2233 video_title=video_title)
2234 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2235 return self.playlist_result(entries, channel_id)
2236
2237 try:
2238 next(self._entries(channel_page, channel_id))
2239 except StopIteration:
2240 alert_message = self._html_search_regex(
2241 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2242 channel_page, 'alert', default=None, group='alert')
2243 if alert_message:
2244 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2245
2246 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2247
2248
2249 class YoutubeUserIE(YoutubeChannelIE):
2250 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2251 _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_-]+)'
2252 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2253 IE_NAME = 'youtube:user'
2254
2255 _TESTS = [{
2256 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2257 'playlist_mincount': 320,
2258 'info_dict': {
2259 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2260 'title': 'Uploads from The Linux Foundation',
2261 }
2262 }, {
2263 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2264 # but not https://www.youtube.com/user/12minuteathlete/videos
2265 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2266 'playlist_mincount': 249,
2267 'info_dict': {
2268 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2269 'title': 'Uploads from 12 Minute Athlete',
2270 }
2271 }, {
2272 'url': 'ytuser:phihag',
2273 'only_matching': True,
2274 }, {
2275 'url': 'https://www.youtube.com/c/gametrailers',
2276 'only_matching': True,
2277 }, {
2278 'url': 'https://www.youtube.com/gametrailers',
2279 'only_matching': True,
2280 }, {
2281 # This channel is not available, geo restricted to JP
2282 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2283 'only_matching': True,
2284 }]
2285
2286 @classmethod
2287 def suitable(cls, url):
2288 # Don't return True if the url can be extracted with other youtube
2289 # extractor, the regex would is too permissive and it would match.
2290 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2291 if any(ie.suitable(url) for ie in other_yt_ies):
2292 return False
2293 else:
2294 return super(YoutubeUserIE, cls).suitable(url)
2295
2296 def _build_template_url(self, url, channel_id):
2297 mobj = re.match(self._VALID_URL, url)
2298 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2299
2300
2301 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2302 IE_DESC = 'YouTube.com live streams'
2303 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2304 IE_NAME = 'youtube:live'
2305
2306 _TESTS = [{
2307 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2308 'info_dict': {
2309 'id': 'a48o2S1cPoo',
2310 'ext': 'mp4',
2311 'title': 'The Young Turks - Live Main Show',
2312 'uploader': 'The Young Turks',
2313 'uploader_id': 'TheYoungTurks',
2314 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2315 'upload_date': '20150715',
2316 'license': 'Standard YouTube License',
2317 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2318 'categories': ['News & Politics'],
2319 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2320 'like_count': int,
2321 'dislike_count': int,
2322 },
2323 'params': {
2324 'skip_download': True,
2325 },
2326 }, {
2327 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2328 'only_matching': True,
2329 }, {
2330 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2331 'only_matching': True,
2332 }, {
2333 'url': 'https://www.youtube.com/TheYoungTurks/live',
2334 'only_matching': True,
2335 }]
2336
2337 def _real_extract(self, url):
2338 mobj = re.match(self._VALID_URL, url)
2339 channel_id = mobj.group('id')
2340 base_url = mobj.group('base_url')
2341 webpage = self._download_webpage(url, channel_id, fatal=False)
2342 if webpage:
2343 page_type = self._og_search_property(
2344 'type', webpage, 'page type', default=None)
2345 video_id = self._html_search_meta(
2346 'videoId', webpage, 'video id', default=None)
2347 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2348 return self.url_result(video_id, YoutubeIE.ie_key())
2349 return self.url_result(base_url)
2350
2351
2352 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2353 IE_DESC = 'YouTube.com user/channel playlists'
2354 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2355 IE_NAME = 'youtube:playlists'
2356
2357 _TESTS = [{
2358 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2359 'playlist_mincount': 4,
2360 'info_dict': {
2361 'id': 'ThirstForScience',
2362 'title': 'Thirst for Science',
2363 },
2364 }, {
2365 # with "Load more" button
2366 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2367 'playlist_mincount': 70,
2368 'info_dict': {
2369 'id': 'igorkle1',
2370 'title': 'Игорь Клейнер',
2371 },
2372 }, {
2373 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2374 'playlist_mincount': 17,
2375 'info_dict': {
2376 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2377 'title': 'Chem Player',
2378 },
2379 }]
2380
2381
2382 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2383 IE_DESC = 'YouTube.com searches'
2384 # there doesn't appear to be a real limit, for example if you search for
2385 # 'python' you get more than 8.000.000 results
2386 _MAX_RESULTS = float('inf')
2387 IE_NAME = 'youtube:search'
2388 _SEARCH_KEY = 'ytsearch'
2389 _EXTRA_QUERY_ARGS = {}
2390 _TESTS = []
2391
2392 def _get_n_results(self, query, n):
2393 """Get a specified number of results for a query"""
2394
2395 videos = []
2396 limit = n
2397
2398 url_query = {
2399 'search_query': query.encode('utf-8'),
2400 }
2401 url_query.update(self._EXTRA_QUERY_ARGS)
2402 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2403
2404 for pagenum in itertools.count(1):
2405 data = self._download_json(
2406 result_url, video_id='query "%s"' % query,
2407 note='Downloading page %s' % pagenum,
2408 errnote='Unable to download API page',
2409 query={'spf': 'navigate'})
2410 html_content = data[1]['body']['content']
2411
2412 if 'class="search-message' in html_content:
2413 raise ExtractorError(
2414 '[youtube] No video results', expected=True)
2415
2416 new_videos = self._ids_to_results(orderedSet(re.findall(
2417 r'href="/watch\?v=(.{11})', html_content)))
2418 videos += new_videos
2419 if not new_videos or len(videos) > limit:
2420 break
2421 next_link = self._html_search_regex(
2422 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2423 html_content, 'next link', default=None)
2424 if next_link is None:
2425 break
2426 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2427
2428 if len(videos) > n:
2429 videos = videos[:n]
2430 return self.playlist_result(videos, query)
2431
2432
2433 class YoutubeSearchDateIE(YoutubeSearchIE):
2434 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2435 _SEARCH_KEY = 'ytsearchdate'
2436 IE_DESC = 'YouTube.com searches, newest videos first'
2437 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2438
2439
2440 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2441 IE_DESC = 'YouTube.com search URLs'
2442 IE_NAME = 'youtube:search_url'
2443 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2444 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2445 _TESTS = [{
2446 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2447 'playlist_mincount': 5,
2448 'info_dict': {
2449 'title': 'youtube-dl test video',
2450 }
2451 }, {
2452 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2453 'only_matching': True,
2454 }]
2455
2456 def _real_extract(self, url):
2457 mobj = re.match(self._VALID_URL, url)
2458 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2459 webpage = self._download_webpage(url, query)
2460 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2461
2462
2463 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2464 IE_DESC = 'YouTube.com (multi-season) shows'
2465 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2466 IE_NAME = 'youtube:show'
2467 _TESTS = [{
2468 'url': 'https://www.youtube.com/show/airdisasters',
2469 'playlist_mincount': 5,
2470 'info_dict': {
2471 'id': 'airdisasters',
2472 'title': 'Air Disasters',
2473 }
2474 }]
2475
2476 def _real_extract(self, url):
2477 playlist_id = self._match_id(url)
2478 return super(YoutubeShowIE, self)._real_extract(
2479 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2480
2481
2482 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2483 """
2484 Base class for feed extractors
2485 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2486 """
2487 _LOGIN_REQUIRED = True
2488
2489 @property
2490 def IE_NAME(self):
2491 return 'youtube:%s' % self._FEED_NAME
2492
2493 def _real_initialize(self):
2494 self._login()
2495
2496 def _real_extract(self, url):
2497 page = self._download_webpage(
2498 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2499
2500 # The extraction process is the same as for playlists, but the regex
2501 # for the video ids doesn't contain an index
2502 ids = []
2503 more_widget_html = content_html = page
2504 for page_num in itertools.count(1):
2505 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2506
2507 # 'recommended' feed has infinite 'load more' and each new portion spins
2508 # the same videos in (sometimes) slightly different order, so we'll check
2509 # for unicity and break when portion has no new videos
2510 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2511 if not new_ids:
2512 break
2513
2514 ids.extend(new_ids)
2515
2516 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2517 if not mobj:
2518 break
2519
2520 more = self._download_json(
2521 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2522 'Downloading page #%s' % page_num,
2523 transform_source=uppercase_escape)
2524 content_html = more['content_html']
2525 more_widget_html = more['load_more_widget_html']
2526
2527 return self.playlist_result(
2528 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2529
2530
2531 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2532 IE_NAME = 'youtube:watchlater'
2533 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2534 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2535
2536 _TESTS = [{
2537 'url': 'https://www.youtube.com/playlist?list=WL',
2538 'only_matching': True,
2539 }, {
2540 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2541 'only_matching': True,
2542 }]
2543
2544 def _real_extract(self, url):
2545 _, video = self._check_download_just_video(url, 'WL')
2546 if video:
2547 return video
2548 _, playlist = self._extract_playlist('WL')
2549 return playlist
2550
2551
2552 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2553 IE_NAME = 'youtube:favorites'
2554 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2555 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2556 _LOGIN_REQUIRED = True
2557
2558 def _real_extract(self, url):
2559 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2560 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2561 return self.url_result(playlist_id, 'YoutubePlaylist')
2562
2563
2564 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2565 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2566 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2567 _FEED_NAME = 'recommended'
2568 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2569
2570
2571 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2572 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2573 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2574 _FEED_NAME = 'subscriptions'
2575 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2576
2577
2578 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2579 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2580 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2581 _FEED_NAME = 'history'
2582 _PLAYLIST_TITLE = 'Youtube History'
2583
2584
2585 class YoutubeTruncatedURLIE(InfoExtractor):
2586 IE_NAME = 'youtube:truncated_url'
2587 IE_DESC = False # Do not list
2588 _VALID_URL = r'''(?x)
2589 (?:https?://)?
2590 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2591 (?:watch\?(?:
2592 feature=[a-z_]+|
2593 annotation_id=annotation_[^&]+|
2594 x-yt-cl=[0-9]+|
2595 hl=[^&]*|
2596 t=[0-9]+
2597 )?
2598 |
2599 attribution_link\?a=[^&]+
2600 )
2601 $
2602 '''
2603
2604 _TESTS = [{
2605 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2606 'only_matching': True,
2607 }, {
2608 'url': 'https://www.youtube.com/watch?',
2609 'only_matching': True,
2610 }, {
2611 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2612 'only_matching': True,
2613 }, {
2614 'url': 'https://www.youtube.com/watch?feature=foo',
2615 'only_matching': True,
2616 }, {
2617 'url': 'https://www.youtube.com/watch?hl=en-GB',
2618 'only_matching': True,
2619 }, {
2620 'url': 'https://www.youtube.com/watch?t=2372',
2621 'only_matching': True,
2622 }]
2623
2624 def _real_extract(self, url):
2625 raise ExtractorError(
2626 'Did you forget to quote the URL? Remember that & is a meta '
2627 'character in most shells, so you want to put the URL in quotes, '
2628 'like youtube-dl '
2629 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2630 ' or simply youtube-dl BaW_jenozKc .',
2631 expected=True)
2632
2633
2634 class YoutubeTruncatedIDIE(InfoExtractor):
2635 IE_NAME = 'youtube:truncated_id'
2636 IE_DESC = False # Do not list
2637 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2638
2639 _TESTS = [{
2640 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2641 'only_matching': True,
2642 }]
2643
2644 def _real_extract(self, url):
2645 video_id = self._match_id(url)
2646 raise ExtractorError(
2647 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2648 expected=True)