]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/youtube.py
[utils] Rename try_multipart_encode to _multipart_encode_impl
[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 sanitized_Request,
42 smuggle_url,
43 str_to_int,
44 try_get,
45 unescapeHTML,
46 unified_strdate,
47 unsmuggle_url,
48 uppercase_escape,
49 urlencode_postdata,
50 )
51
52
53 class YoutubeBaseInfoExtractor(InfoExtractor):
54 """Provide base functions for Youtube extractors"""
55 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
56 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
57 _PASSWORD_CHALLENGE_URL = 'https://accounts.google.com/signin/challenge/sl/password'
58 _NETRC_MACHINE = 'youtube'
59 # If True it will raise an error if no login info is provided
60 _LOGIN_REQUIRED = False
61
62 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
63
64 def _set_language(self):
65 self._set_cookie(
66 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
67 # YouTube sets the expire time to about two months
68 expire_time=time.time() + 2 * 30 * 24 * 3600)
69
70 def _ids_to_results(self, ids):
71 return [
72 self.url_result(vid_id, 'Youtube', video_id=vid_id)
73 for vid_id in ids]
74
75 def _login(self):
76 """
77 Attempt to log in to YouTube.
78 True is returned if successful or skipped.
79 False is returned if login failed.
80
81 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
82 """
83 (username, password) = self._get_login_info()
84 # No authentication to be performed
85 if username is None:
86 if self._LOGIN_REQUIRED:
87 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
88 return True
89
90 login_page = self._download_webpage(
91 self._LOGIN_URL, None,
92 note='Downloading login page',
93 errnote='unable to fetch login page', fatal=False)
94 if login_page is False:
95 return
96
97 login_form = self._hidden_inputs(login_page)
98
99 login_form.update({
100 'checkConnection': 'youtube',
101 'Email': username,
102 'Passwd': password,
103 })
104
105 login_results = self._download_webpage(
106 self._PASSWORD_CHALLENGE_URL, None,
107 note='Logging in', errnote='unable to log in', fatal=False,
108 data=urlencode_postdata(login_form))
109 if login_results is False:
110 return False
111
112 error_msg = self._html_search_regex(
113 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
114 login_results, 'error message', default=None)
115 if error_msg:
116 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
117
118 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
119 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
120
121 # Two-Factor
122 # TODO add SMS and phone call support - these require making a request and then prompting the user
123
124 if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
125 tfa_code = self._get_tfa_info('2-step verification code')
126
127 if not tfa_code:
128 self._downloader.report_warning(
129 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
130 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
131 return False
132
133 tfa_code = remove_start(tfa_code, 'G-')
134
135 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
136
137 tfa_form_strs.update({
138 'Pin': tfa_code,
139 'TrustDevice': 'on',
140 })
141
142 tfa_data = urlencode_postdata(tfa_form_strs)
143
144 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
145 tfa_results = self._download_webpage(
146 tfa_req, None,
147 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
148
149 if tfa_results is False:
150 return False
151
152 if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
153 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
154 return False
155 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
156 self._downloader.report_warning('unable to log in - did the page structure change?')
157 return False
158 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
159 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
160 return False
161
162 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
163 self._downloader.report_warning('unable to log in: bad username or password')
164 return False
165 return True
166
167 def _real_initialize(self):
168 if self._downloader is None:
169 return
170 self._set_language()
171 if not self._login():
172 return
173
174
175 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
176 # Extract entries from page with "Load more" button
177 def _entries(self, page, playlist_id):
178 more_widget_html = content_html = page
179 for page_num in itertools.count(1):
180 for entry in self._process_page(content_html):
181 yield entry
182
183 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
184 if not mobj:
185 break
186
187 more = self._download_json(
188 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
189 'Downloading page #%s' % page_num,
190 transform_source=uppercase_escape)
191 content_html = more['content_html']
192 if not content_html.strip():
193 # Some webpages show a "Load more" button but they don't
194 # have more videos
195 break
196 more_widget_html = more['load_more_widget_html']
197
198
199 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
200 def _process_page(self, content):
201 for video_id, video_title in self.extract_videos_from_page(content):
202 yield self.url_result(video_id, 'Youtube', video_id, video_title)
203
204 def extract_videos_from_page(self, page):
205 ids_in_page = []
206 titles_in_page = []
207 for mobj in re.finditer(self._VIDEO_RE, page):
208 # The link with index 0 is not the first video of the playlist (not sure if still actual)
209 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
210 continue
211 video_id = mobj.group('id')
212 video_title = unescapeHTML(mobj.group('title'))
213 if video_title:
214 video_title = video_title.strip()
215 try:
216 idx = ids_in_page.index(video_id)
217 if video_title and not titles_in_page[idx]:
218 titles_in_page[idx] = video_title
219 except ValueError:
220 ids_in_page.append(video_id)
221 titles_in_page.append(video_title)
222 return zip(ids_in_page, titles_in_page)
223
224
225 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
226 def _process_page(self, content):
227 for playlist_id in orderedSet(re.findall(
228 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
229 content)):
230 yield self.url_result(
231 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
232
233 def _real_extract(self, url):
234 playlist_id = self._match_id(url)
235 webpage = self._download_webpage(url, playlist_id)
236 title = self._og_search_title(webpage, fatal=False)
237 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
238
239
240 class YoutubeIE(YoutubeBaseInfoExtractor):
241 IE_DESC = 'YouTube.com'
242 _VALID_URL = r"""(?x)^
243 (
244 (?:https?://|//) # http(s):// or protocol-independent URL
245 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
246 (?:www\.)?deturl\.com/www\.youtube\.com/|
247 (?:www\.)?pwnyoutube\.com/|
248 (?:www\.)?yourepeat\.com/|
249 tube\.majestyc\.net/|
250 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
251 (?:.*?\#/)? # handle anchor (#/) redirect urls
252 (?: # the various things that can precede the ID:
253 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
254 |(?: # or the v= param in all its forms
255 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
256 (?:\?|\#!?) # the params delimiter ? or # or #!
257 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
258 v=
259 )
260 ))
261 |(?:
262 youtu\.be| # just youtu.be/xxxx
263 vid\.plus| # or vid.plus/xxxx
264 zwearz\.com/watch| # or zwearz.com/watch/xxxx
265 )/
266 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
267 )
268 )? # all until now is optional -> you can pass the naked ID
269 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
270 (?!.*?\blist=
271 (?:
272 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
273 WL # WL are handled by the watch later IE
274 )
275 )
276 (?(1).+)? # if we found the ID, everything can follow
277 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
278 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
279 _formats = {
280 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
281 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
282 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
283 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
284 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
285 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
286 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
287 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
288 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
289 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
290 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
291 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
292 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
293 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
294 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
295 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
296 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
297 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
298
299
300 # 3D videos
301 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
302 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
303 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
304 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
305 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
306 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
307 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
308
309 # Apple HTTP Live Streaming
310 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
311 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
312 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
313 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
314 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
315 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
316 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
317 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
318
319 # DASH mp4 video
320 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
321 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
322 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
323 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
324 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
325 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
326 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
327 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
328 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
329 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
330 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
331 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
332
333 # Dash mp4 audio
334 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
335 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
336 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
337 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
338 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
339 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
340 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
341
342 # Dash webm
343 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
344 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
345 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
346 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
347 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
348 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
349 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
350 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
351 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
352 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
353 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
354 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
355 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
356 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
357 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
358 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
359 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
360 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
361 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
362 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
363 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
364 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
365
366 # Dash webm audio
367 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
368 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
369
370 # Dash webm audio with opus inside
371 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
372 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
373 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
374
375 # RTMP (unnamed)
376 '_rtmp': {'protocol': 'rtmp'},
377 }
378 _SUBTITLE_FORMATS = ('ttml', 'vtt')
379
380 _GEO_BYPASS = False
381
382 IE_NAME = 'youtube'
383 _TESTS = [
384 {
385 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
386 'info_dict': {
387 'id': 'BaW_jenozKc',
388 'ext': 'mp4',
389 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
390 'uploader': 'Philipp Hagemeister',
391 'uploader_id': 'phihag',
392 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
393 'upload_date': '20121002',
394 'license': 'Standard YouTube License',
395 '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 .',
396 'categories': ['Science & Technology'],
397 'tags': ['youtube-dl'],
398 'duration': 10,
399 'like_count': int,
400 'dislike_count': int,
401 'start_time': 1,
402 'end_time': 9,
403 }
404 },
405 {
406 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
407 'note': 'Test generic use_cipher_signature video (#897)',
408 'info_dict': {
409 'id': 'UxxajLWwzqY',
410 'ext': 'mp4',
411 'upload_date': '20120506',
412 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
413 'alt_title': 'I Love It (feat. Charli XCX)',
414 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
415 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
416 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
417 'iconic ep', 'iconic', 'love', 'it'],
418 'duration': 180,
419 'uploader': 'Icona Pop',
420 'uploader_id': 'IconaPop',
421 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
422 'license': 'Standard YouTube License',
423 'creator': 'Icona Pop',
424 }
425 },
426 {
427 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
428 'note': 'Test VEVO video with age protection (#956)',
429 'info_dict': {
430 'id': '07FYdnEawAQ',
431 'ext': 'mp4',
432 'upload_date': '20130703',
433 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
434 'alt_title': 'Tunnel Vision',
435 'description': 'md5:64249768eec3bc4276236606ea996373',
436 'duration': 419,
437 'uploader': 'justintimberlakeVEVO',
438 'uploader_id': 'justintimberlakeVEVO',
439 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
440 'license': 'Standard YouTube License',
441 'creator': 'Justin Timberlake',
442 'age_limit': 18,
443 }
444 },
445 {
446 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
447 'note': 'Embed-only video (#1746)',
448 'info_dict': {
449 'id': 'yZIXLfi8CZQ',
450 'ext': 'mp4',
451 'upload_date': '20120608',
452 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
453 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
454 'uploader': 'SET India',
455 'uploader_id': 'setindia',
456 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
457 'license': 'Standard YouTube License',
458 'age_limit': 18,
459 }
460 },
461 {
462 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
463 'note': 'Use the first video ID in the URL',
464 'info_dict': {
465 'id': 'BaW_jenozKc',
466 'ext': 'mp4',
467 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
468 'uploader': 'Philipp Hagemeister',
469 'uploader_id': 'phihag',
470 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
471 'upload_date': '20121002',
472 'license': 'Standard YouTube License',
473 '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 .',
474 'categories': ['Science & Technology'],
475 'tags': ['youtube-dl'],
476 'duration': 10,
477 'like_count': int,
478 'dislike_count': int,
479 },
480 'params': {
481 'skip_download': True,
482 },
483 },
484 {
485 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
486 'note': '256k DASH audio (format 141) via DASH manifest',
487 'info_dict': {
488 'id': 'a9LDPn-MO4I',
489 'ext': 'm4a',
490 'upload_date': '20121002',
491 'uploader_id': '8KVIDEO',
492 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
493 'description': '',
494 'uploader': '8KVIDEO',
495 'license': 'Standard YouTube License',
496 'title': 'UHDTV TEST 8K VIDEO.mp4'
497 },
498 'params': {
499 'youtube_include_dash_manifest': True,
500 'format': '141',
501 },
502 'skip': 'format 141 not served anymore',
503 },
504 # DASH manifest with encrypted signature
505 {
506 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
507 'info_dict': {
508 'id': 'IB3lcPjvWLA',
509 'ext': 'm4a',
510 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
511 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
512 'duration': 244,
513 'uploader': 'AfrojackVEVO',
514 'uploader_id': 'AfrojackVEVO',
515 'upload_date': '20131011',
516 'license': 'Standard YouTube License',
517 },
518 'params': {
519 'youtube_include_dash_manifest': True,
520 'format': '141/bestaudio[ext=m4a]',
521 },
522 },
523 # JS player signature function name containing $
524 {
525 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
526 'info_dict': {
527 'id': 'nfWlot6h_JM',
528 'ext': 'm4a',
529 'title': 'Taylor Swift - Shake It Off',
530 'alt_title': 'Shake It Off',
531 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
532 'duration': 242,
533 'uploader': 'TaylorSwiftVEVO',
534 'uploader_id': 'TaylorSwiftVEVO',
535 'upload_date': '20140818',
536 'license': 'Standard YouTube License',
537 'creator': 'Taylor Swift',
538 },
539 'params': {
540 'youtube_include_dash_manifest': True,
541 'format': '141/bestaudio[ext=m4a]',
542 },
543 },
544 # Controversy video
545 {
546 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
547 'info_dict': {
548 'id': 'T4XJQO3qol8',
549 'ext': 'mp4',
550 'duration': 219,
551 'upload_date': '20100909',
552 'uploader': 'The Amazing Atheist',
553 'uploader_id': 'TheAmazingAtheist',
554 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
555 'license': 'Standard YouTube License',
556 'title': 'Burning Everyone\'s Koran',
557 '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',
558 }
559 },
560 # Normal age-gate video (No vevo, embed allowed)
561 {
562 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
563 'info_dict': {
564 'id': 'HtVdAasjOgU',
565 'ext': 'mp4',
566 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
567 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
568 'duration': 142,
569 'uploader': 'The Witcher',
570 'uploader_id': 'WitcherGame',
571 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
572 'upload_date': '20140605',
573 'license': 'Standard YouTube License',
574 'age_limit': 18,
575 },
576 },
577 # Age-gate video with encrypted signature
578 {
579 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
580 'info_dict': {
581 'id': '6kLq3WMV1nU',
582 'ext': 'mp4',
583 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
584 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
585 'duration': 247,
586 'uploader': 'LloydVEVO',
587 'uploader_id': 'LloydVEVO',
588 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
589 'upload_date': '20110629',
590 'license': 'Standard YouTube License',
591 'age_limit': 18,
592 },
593 },
594 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
595 {
596 'url': '__2ABJjxzNo',
597 'info_dict': {
598 'id': '__2ABJjxzNo',
599 'ext': 'mp4',
600 'duration': 266,
601 'upload_date': '20100430',
602 'uploader_id': 'deadmau5',
603 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
604 'creator': 'deadmau5',
605 'description': 'md5:12c56784b8032162bb936a5f76d55360',
606 'uploader': 'deadmau5',
607 'license': 'Standard YouTube License',
608 'title': 'Deadmau5 - Some Chords (HD)',
609 'alt_title': 'Some Chords',
610 },
611 'expected_warnings': [
612 'DASH manifest missing',
613 ]
614 },
615 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
616 {
617 'url': 'lqQg6PlCWgI',
618 'info_dict': {
619 'id': 'lqQg6PlCWgI',
620 'ext': 'mp4',
621 'duration': 6085,
622 'upload_date': '20150827',
623 'uploader_id': 'olympic',
624 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
625 'license': 'Standard YouTube License',
626 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
627 'uploader': 'Olympic',
628 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
629 },
630 'params': {
631 'skip_download': 'requires avconv',
632 }
633 },
634 # Non-square pixels
635 {
636 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
637 'info_dict': {
638 'id': '_b-2C3KPAM0',
639 'ext': 'mp4',
640 'stretched_ratio': 16 / 9.,
641 'duration': 85,
642 'upload_date': '20110310',
643 'uploader_id': 'AllenMeow',
644 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
645 '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',
646 'uploader': '孫艾倫',
647 'license': 'Standard YouTube License',
648 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
649 },
650 },
651 # url_encoded_fmt_stream_map is empty string
652 {
653 'url': 'qEJwOuvDf7I',
654 'info_dict': {
655 'id': 'qEJwOuvDf7I',
656 'ext': 'webm',
657 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
658 'description': '',
659 'upload_date': '20150404',
660 'uploader_id': 'spbelect',
661 'uploader': 'Наблюдатели Петербурга',
662 },
663 'params': {
664 'skip_download': 'requires avconv',
665 },
666 'skip': 'This live event has ended.',
667 },
668 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
669 {
670 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
671 'info_dict': {
672 'id': 'FIl7x6_3R5Y',
673 'ext': 'mp4',
674 'title': 'md5:7b81415841e02ecd4313668cde88737a',
675 'description': 'md5:116377fd2963b81ec4ce64b542173306',
676 'duration': 220,
677 'upload_date': '20150625',
678 'uploader_id': 'dorappi2000',
679 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
680 'uploader': 'dorappi2000',
681 'license': 'Standard YouTube License',
682 'formats': 'mincount:32',
683 },
684 },
685 # DASH manifest with segment_list
686 {
687 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
688 'md5': '8ce563a1d667b599d21064e982ab9e31',
689 'info_dict': {
690 'id': 'CsmdDsKjzN8',
691 'ext': 'mp4',
692 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
693 'uploader': 'Airtek',
694 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
695 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
696 'license': 'Standard YouTube License',
697 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
698 },
699 'params': {
700 'youtube_include_dash_manifest': True,
701 'format': '135', # bestvideo
702 },
703 'skip': 'This live event has ended.',
704 },
705 {
706 # Multifeed videos (multiple cameras), URL is for Main Camera
707 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
708 'info_dict': {
709 'id': 'jqWvoWXjCVs',
710 'title': 'teamPGP: Rocket League Noob Stream',
711 'description': 'md5:dc7872fb300e143831327f1bae3af010',
712 },
713 'playlist': [{
714 'info_dict': {
715 'id': 'jqWvoWXjCVs',
716 'ext': 'mp4',
717 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
718 'description': 'md5:dc7872fb300e143831327f1bae3af010',
719 'duration': 7335,
720 'upload_date': '20150721',
721 'uploader': 'Beer Games Beer',
722 'uploader_id': 'beergamesbeer',
723 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
724 'license': 'Standard YouTube License',
725 },
726 }, {
727 'info_dict': {
728 'id': '6h8e8xoXJzg',
729 'ext': 'mp4',
730 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
731 'description': 'md5:dc7872fb300e143831327f1bae3af010',
732 'duration': 7337,
733 'upload_date': '20150721',
734 'uploader': 'Beer Games Beer',
735 'uploader_id': 'beergamesbeer',
736 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
737 'license': 'Standard YouTube License',
738 },
739 }, {
740 'info_dict': {
741 'id': 'PUOgX5z9xZw',
742 'ext': 'mp4',
743 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
744 'description': 'md5:dc7872fb300e143831327f1bae3af010',
745 'duration': 7337,
746 'upload_date': '20150721',
747 'uploader': 'Beer Games Beer',
748 'uploader_id': 'beergamesbeer',
749 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
750 'license': 'Standard YouTube License',
751 },
752 }, {
753 'info_dict': {
754 'id': 'teuwxikvS5k',
755 'ext': 'mp4',
756 'title': 'teamPGP: Rocket League Noob Stream (zim)',
757 'description': 'md5:dc7872fb300e143831327f1bae3af010',
758 'duration': 7334,
759 'upload_date': '20150721',
760 'uploader': 'Beer Games Beer',
761 'uploader_id': 'beergamesbeer',
762 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
763 'license': 'Standard YouTube License',
764 },
765 }],
766 'params': {
767 'skip_download': True,
768 },
769 },
770 {
771 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
772 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
773 'info_dict': {
774 'id': 'gVfLd0zydlo',
775 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
776 },
777 'playlist_count': 2,
778 'skip': 'Not multifeed anymore',
779 },
780 {
781 'url': 'https://vid.plus/FlRa-iH7PGw',
782 'only_matching': True,
783 },
784 {
785 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
786 'only_matching': True,
787 },
788 {
789 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
790 # Also tests cut-off URL expansion in video description (see
791 # https://github.com/rg3/youtube-dl/issues/1892,
792 # https://github.com/rg3/youtube-dl/issues/8164)
793 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
794 'info_dict': {
795 'id': 'lsguqyKfVQg',
796 'ext': 'mp4',
797 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
798 'alt_title': 'Dark Walk',
799 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
800 'duration': 133,
801 'upload_date': '20151119',
802 'uploader_id': 'IronSoulElf',
803 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
804 'uploader': 'IronSoulElf',
805 'license': 'Standard YouTube License',
806 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
807 },
808 'params': {
809 'skip_download': True,
810 },
811 },
812 {
813 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
814 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
815 'only_matching': True,
816 },
817 {
818 # Video with yt:stretch=17:0
819 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
820 'info_dict': {
821 'id': 'Q39EVAstoRM',
822 'ext': 'mp4',
823 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
824 'description': 'md5:ee18a25c350637c8faff806845bddee9',
825 'upload_date': '20151107',
826 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
827 'uploader': 'CH GAMER DROID',
828 },
829 'params': {
830 'skip_download': True,
831 },
832 'skip': 'This video does not exist.',
833 },
834 {
835 # Video licensed under Creative Commons
836 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
837 'info_dict': {
838 'id': 'M4gD1WSo5mA',
839 'ext': 'mp4',
840 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
841 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
842 'duration': 721,
843 'upload_date': '20150127',
844 'uploader_id': 'BerkmanCenter',
845 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
846 'uploader': 'The Berkman Klein Center for Internet & Society',
847 'license': 'Creative Commons Attribution license (reuse allowed)',
848 },
849 'params': {
850 'skip_download': True,
851 },
852 },
853 {
854 # Channel-like uploader_url
855 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
856 'info_dict': {
857 'id': 'eQcmzGIKrzg',
858 'ext': 'mp4',
859 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
860 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
861 'duration': 4060,
862 'upload_date': '20151119',
863 'uploader': 'Bernie 2016',
864 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
865 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
866 'license': 'Creative Commons Attribution license (reuse allowed)',
867 },
868 'params': {
869 'skip_download': True,
870 },
871 },
872 {
873 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
874 'only_matching': True,
875 },
876 {
877 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
878 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
879 'only_matching': True,
880 },
881 {
882 # Rental video preview
883 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
884 'info_dict': {
885 'id': 'uGpuVWrhIzE',
886 'ext': 'mp4',
887 'title': 'Piku - Trailer',
888 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
889 'upload_date': '20150811',
890 'uploader': 'FlixMatrix',
891 'uploader_id': 'FlixMatrixKaravan',
892 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
893 'license': 'Standard YouTube License',
894 },
895 'params': {
896 'skip_download': True,
897 },
898 },
899 {
900 # YouTube Red video with episode data
901 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
902 'info_dict': {
903 'id': 'iqKdEhx-dD4',
904 'ext': 'mp4',
905 'title': 'Isolation - Mind Field (Ep 1)',
906 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
907 'duration': 2085,
908 'upload_date': '20170118',
909 'uploader': 'Vsauce',
910 'uploader_id': 'Vsauce',
911 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
912 'license': 'Standard YouTube License',
913 'series': 'Mind Field',
914 'season_number': 1,
915 'episode_number': 1,
916 },
917 'params': {
918 'skip_download': True,
919 },
920 'expected_warnings': [
921 'Skipping DASH manifest',
922 ],
923 },
924 {
925 # itag 212
926 'url': '1t24XAntNCY',
927 'only_matching': True,
928 },
929 {
930 # geo restricted to JP
931 'url': 'sJL6WA-aGkQ',
932 'only_matching': True,
933 },
934 {
935 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
936 'only_matching': True,
937 },
938 ]
939
940 def __init__(self, *args, **kwargs):
941 super(YoutubeIE, self).__init__(*args, **kwargs)
942 self._player_cache = {}
943
944 def report_video_info_webpage_download(self, video_id):
945 """Report attempt to download video info webpage."""
946 self.to_screen('%s: Downloading video info webpage' % video_id)
947
948 def report_information_extraction(self, video_id):
949 """Report attempt to extract video information."""
950 self.to_screen('%s: Extracting video information' % video_id)
951
952 def report_unavailable_format(self, video_id, format):
953 """Report extracted video URL."""
954 self.to_screen('%s: Format %s not available' % (video_id, format))
955
956 def report_rtmp_download(self):
957 """Indicate the download will use the RTMP protocol."""
958 self.to_screen('RTMP download detected')
959
960 def _signature_cache_id(self, example_sig):
961 """ Return a string representation of a signature """
962 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
963
964 def _extract_signature_function(self, video_id, player_url, example_sig):
965 id_m = re.match(
966 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
967 player_url)
968 if not id_m:
969 raise ExtractorError('Cannot identify player %r' % player_url)
970 player_type = id_m.group('ext')
971 player_id = id_m.group('id')
972
973 # Read from filesystem cache
974 func_id = '%s_%s_%s' % (
975 player_type, player_id, self._signature_cache_id(example_sig))
976 assert os.path.basename(func_id) == func_id
977
978 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
979 if cache_spec is not None:
980 return lambda s: ''.join(s[i] for i in cache_spec)
981
982 download_note = (
983 'Downloading player %s' % player_url
984 if self._downloader.params.get('verbose') else
985 'Downloading %s player %s' % (player_type, player_id)
986 )
987 if player_type == 'js':
988 code = self._download_webpage(
989 player_url, video_id,
990 note=download_note,
991 errnote='Download of %s failed' % player_url)
992 res = self._parse_sig_js(code)
993 elif player_type == 'swf':
994 urlh = self._request_webpage(
995 player_url, video_id,
996 note=download_note,
997 errnote='Download of %s failed' % player_url)
998 code = urlh.read()
999 res = self._parse_sig_swf(code)
1000 else:
1001 assert False, 'Invalid player type %r' % player_type
1002
1003 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1004 cache_res = res(test_string)
1005 cache_spec = [ord(c) for c in cache_res]
1006
1007 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1008 return res
1009
1010 def _print_sig_code(self, func, example_sig):
1011 def gen_sig_code(idxs):
1012 def _genslice(start, end, step):
1013 starts = '' if start == 0 else str(start)
1014 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1015 steps = '' if step == 1 else (':%d' % step)
1016 return 's[%s%s%s]' % (starts, ends, steps)
1017
1018 step = None
1019 # Quelch pyflakes warnings - start will be set when step is set
1020 start = '(Never used)'
1021 for i, prev in zip(idxs[1:], idxs[:-1]):
1022 if step is not None:
1023 if i - prev == step:
1024 continue
1025 yield _genslice(start, prev, step)
1026 step = None
1027 continue
1028 if i - prev in [-1, 1]:
1029 step = i - prev
1030 start = prev
1031 continue
1032 else:
1033 yield 's[%d]' % prev
1034 if step is None:
1035 yield 's[%d]' % i
1036 else:
1037 yield _genslice(start, i, step)
1038
1039 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1040 cache_res = func(test_string)
1041 cache_spec = [ord(c) for c in cache_res]
1042 expr_code = ' + '.join(gen_sig_code(cache_spec))
1043 signature_id_tuple = '(%s)' % (
1044 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1045 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1046 ' return %s\n') % (signature_id_tuple, expr_code)
1047 self.to_screen('Extracted signature function:\n' + code)
1048
1049 def _parse_sig_js(self, jscode):
1050 funcname = self._search_regex(
1051 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1052 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1053 jscode, 'Initial JS player signature function name', group='sig')
1054
1055 jsi = JSInterpreter(jscode)
1056 initial_function = jsi.extract_function(funcname)
1057 return lambda s: initial_function([s])
1058
1059 def _parse_sig_swf(self, file_contents):
1060 swfi = SWFInterpreter(file_contents)
1061 TARGET_CLASSNAME = 'SignatureDecipher'
1062 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1063 initial_function = swfi.extract_function(searched_class, 'decipher')
1064 return lambda s: initial_function([s])
1065
1066 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1067 """Turn the encrypted s field into a working signature"""
1068
1069 if player_url is None:
1070 raise ExtractorError('Cannot decrypt signature without player_url')
1071
1072 if player_url.startswith('//'):
1073 player_url = 'https:' + player_url
1074 elif not re.match(r'https?://', player_url):
1075 player_url = compat_urlparse.urljoin(
1076 'https://www.youtube.com', player_url)
1077 try:
1078 player_id = (player_url, self._signature_cache_id(s))
1079 if player_id not in self._player_cache:
1080 func = self._extract_signature_function(
1081 video_id, player_url, s
1082 )
1083 self._player_cache[player_id] = func
1084 func = self._player_cache[player_id]
1085 if self._downloader.params.get('youtube_print_sig_code'):
1086 self._print_sig_code(func, s)
1087 return func(s)
1088 except Exception as e:
1089 tb = traceback.format_exc()
1090 raise ExtractorError(
1091 'Signature extraction failed: ' + tb, cause=e)
1092
1093 def _get_subtitles(self, video_id, webpage):
1094 try:
1095 subs_doc = self._download_xml(
1096 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1097 video_id, note=False)
1098 except ExtractorError as err:
1099 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1100 return {}
1101
1102 sub_lang_list = {}
1103 for track in subs_doc.findall('track'):
1104 lang = track.attrib['lang_code']
1105 if lang in sub_lang_list:
1106 continue
1107 sub_formats = []
1108 for ext in self._SUBTITLE_FORMATS:
1109 params = compat_urllib_parse_urlencode({
1110 'lang': lang,
1111 'v': video_id,
1112 'fmt': ext,
1113 'name': track.attrib['name'].encode('utf-8'),
1114 })
1115 sub_formats.append({
1116 'url': 'https://www.youtube.com/api/timedtext?' + params,
1117 'ext': ext,
1118 })
1119 sub_lang_list[lang] = sub_formats
1120 if not sub_lang_list:
1121 self._downloader.report_warning('video doesn\'t have subtitles')
1122 return {}
1123 return sub_lang_list
1124
1125 def _get_ytplayer_config(self, video_id, webpage):
1126 patterns = (
1127 # User data may contain arbitrary character sequences that may affect
1128 # JSON extraction with regex, e.g. when '};' is contained the second
1129 # regex won't capture the whole JSON. Yet working around by trying more
1130 # concrete regex first keeping in mind proper quoted string handling
1131 # to be implemented in future that will replace this workaround (see
1132 # https://github.com/rg3/youtube-dl/issues/7468,
1133 # https://github.com/rg3/youtube-dl/pull/7599)
1134 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1135 r';ytplayer\.config\s*=\s*({.+?});',
1136 )
1137 config = self._search_regex(
1138 patterns, webpage, 'ytplayer.config', default=None)
1139 if config:
1140 return self._parse_json(
1141 uppercase_escape(config), video_id, fatal=False)
1142
1143 def _get_automatic_captions(self, video_id, webpage):
1144 """We need the webpage for getting the captions url, pass it as an
1145 argument to speed up the process."""
1146 self.to_screen('%s: Looking for automatic captions' % video_id)
1147 player_config = self._get_ytplayer_config(video_id, webpage)
1148 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1149 if not player_config:
1150 self._downloader.report_warning(err_msg)
1151 return {}
1152 try:
1153 args = player_config['args']
1154 caption_url = args.get('ttsurl')
1155 if caption_url:
1156 timestamp = args['timestamp']
1157 # We get the available subtitles
1158 list_params = compat_urllib_parse_urlencode({
1159 'type': 'list',
1160 'tlangs': 1,
1161 'asrs': 1,
1162 })
1163 list_url = caption_url + '&' + list_params
1164 caption_list = self._download_xml(list_url, video_id)
1165 original_lang_node = caption_list.find('track')
1166 if original_lang_node is None:
1167 self._downloader.report_warning('Video doesn\'t have automatic captions')
1168 return {}
1169 original_lang = original_lang_node.attrib['lang_code']
1170 caption_kind = original_lang_node.attrib.get('kind', '')
1171
1172 sub_lang_list = {}
1173 for lang_node in caption_list.findall('target'):
1174 sub_lang = lang_node.attrib['lang_code']
1175 sub_formats = []
1176 for ext in self._SUBTITLE_FORMATS:
1177 params = compat_urllib_parse_urlencode({
1178 'lang': original_lang,
1179 'tlang': sub_lang,
1180 'fmt': ext,
1181 'ts': timestamp,
1182 'kind': caption_kind,
1183 })
1184 sub_formats.append({
1185 'url': caption_url + '&' + params,
1186 'ext': ext,
1187 })
1188 sub_lang_list[sub_lang] = sub_formats
1189 return sub_lang_list
1190
1191 # Some videos don't provide ttsurl but rather caption_tracks and
1192 # caption_translation_languages (e.g. 20LmZk1hakA)
1193 caption_tracks = args['caption_tracks']
1194 caption_translation_languages = args['caption_translation_languages']
1195 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1196 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1197 caption_qs = compat_parse_qs(parsed_caption_url.query)
1198
1199 sub_lang_list = {}
1200 for lang in caption_translation_languages.split(','):
1201 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1202 sub_lang = lang_qs.get('lc', [None])[0]
1203 if not sub_lang:
1204 continue
1205 sub_formats = []
1206 for ext in self._SUBTITLE_FORMATS:
1207 caption_qs.update({
1208 'tlang': [sub_lang],
1209 'fmt': [ext],
1210 })
1211 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1212 query=compat_urllib_parse_urlencode(caption_qs, True)))
1213 sub_formats.append({
1214 'url': sub_url,
1215 'ext': ext,
1216 })
1217 sub_lang_list[sub_lang] = sub_formats
1218 return sub_lang_list
1219 # An extractor error can be raise by the download process if there are
1220 # no automatic captions but there are subtitles
1221 except (KeyError, ExtractorError):
1222 self._downloader.report_warning(err_msg)
1223 return {}
1224
1225 def _mark_watched(self, video_id, video_info):
1226 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1227 if not playback_url:
1228 return
1229 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1230 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1231
1232 # cpn generation algorithm is reverse engineered from base.js.
1233 # In fact it works even with dummy cpn.
1234 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1235 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1236
1237 qs.update({
1238 'ver': ['2'],
1239 'cpn': [cpn],
1240 })
1241 playback_url = compat_urlparse.urlunparse(
1242 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1243
1244 self._download_webpage(
1245 playback_url, video_id, 'Marking watched',
1246 'Unable to mark watched', fatal=False)
1247
1248 @classmethod
1249 def extract_id(cls, url):
1250 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1251 if mobj is None:
1252 raise ExtractorError('Invalid URL: %s' % url)
1253 video_id = mobj.group(2)
1254 return video_id
1255
1256 def _extract_annotations(self, video_id):
1257 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1258 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1259
1260 @staticmethod
1261 def _extract_chapters(description, duration):
1262 if not description:
1263 return None
1264 chapter_lines = re.findall(
1265 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1266 description)
1267 if not chapter_lines:
1268 return None
1269 chapters = []
1270 for next_num, (chapter_line, time_point) in enumerate(
1271 chapter_lines, start=1):
1272 start_time = parse_duration(time_point)
1273 if start_time is None:
1274 continue
1275 end_time = (duration if next_num == len(chapter_lines)
1276 else parse_duration(chapter_lines[next_num][1]))
1277 if end_time is None:
1278 continue
1279 chapter_title = re.sub(
1280 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1281 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1282 chapters.append({
1283 'start_time': start_time,
1284 'end_time': end_time,
1285 'title': chapter_title,
1286 })
1287 return chapters
1288
1289 def _real_extract(self, url):
1290 url, smuggled_data = unsmuggle_url(url, {})
1291
1292 proto = (
1293 'http' if self._downloader.params.get('prefer_insecure', False)
1294 else 'https')
1295
1296 start_time = None
1297 end_time = None
1298 parsed_url = compat_urllib_parse_urlparse(url)
1299 for component in [parsed_url.fragment, parsed_url.query]:
1300 query = compat_parse_qs(component)
1301 if start_time is None and 't' in query:
1302 start_time = parse_duration(query['t'][0])
1303 if start_time is None and 'start' in query:
1304 start_time = parse_duration(query['start'][0])
1305 if end_time is None and 'end' in query:
1306 end_time = parse_duration(query['end'][0])
1307
1308 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1309 mobj = re.search(self._NEXT_URL_RE, url)
1310 if mobj:
1311 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1312 video_id = self.extract_id(url)
1313
1314 # Get video webpage
1315 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1316 video_webpage = self._download_webpage(url, video_id)
1317
1318 # Attempt to extract SWF player URL
1319 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1320 if mobj is not None:
1321 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1322 else:
1323 player_url = None
1324
1325 dash_mpds = []
1326
1327 def add_dash_mpd(video_info):
1328 dash_mpd = video_info.get('dashmpd')
1329 if dash_mpd and dash_mpd[0] not in dash_mpds:
1330 dash_mpds.append(dash_mpd[0])
1331
1332 # Get video info
1333 embed_webpage = None
1334 is_live = None
1335 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1336 age_gate = True
1337 # We simulate the access to the video from www.youtube.com/v/{video_id}
1338 # this can be viewed without login into Youtube
1339 url = proto + '://www.youtube.com/embed/%s' % video_id
1340 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1341 data = compat_urllib_parse_urlencode({
1342 'video_id': video_id,
1343 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1344 'sts': self._search_regex(
1345 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1346 })
1347 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1348 video_info_webpage = self._download_webpage(
1349 video_info_url, video_id,
1350 note='Refetching age-gated info webpage',
1351 errnote='unable to download video info webpage')
1352 video_info = compat_parse_qs(video_info_webpage)
1353 add_dash_mpd(video_info)
1354 else:
1355 age_gate = False
1356 video_info = None
1357 # Try looking directly into the video webpage
1358 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1359 if ytplayer_config:
1360 args = ytplayer_config['args']
1361 if args.get('url_encoded_fmt_stream_map'):
1362 # Convert to the same format returned by compat_parse_qs
1363 video_info = dict((k, [v]) for k, v in args.items())
1364 add_dash_mpd(video_info)
1365 # Rental video is not rented but preview is available (e.g.
1366 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1367 # https://github.com/rg3/youtube-dl/issues/10532)
1368 if not video_info and args.get('ypc_vid'):
1369 return self.url_result(
1370 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1371 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1372 is_live = True
1373 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1374 # We also try looking in get_video_info since it may contain different dashmpd
1375 # URL that points to a DASH manifest with possibly different itag set (some itags
1376 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1377 # manifest pointed by get_video_info's dashmpd).
1378 # The general idea is to take a union of itags of both DASH manifests (for example
1379 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1380 self.report_video_info_webpage_download(video_id)
1381 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1382 video_info_url = (
1383 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1384 % (proto, video_id, el_type))
1385 video_info_webpage = self._download_webpage(
1386 video_info_url,
1387 video_id, note=False,
1388 errnote='unable to download video info webpage')
1389 get_video_info = compat_parse_qs(video_info_webpage)
1390 if get_video_info.get('use_cipher_signature') != ['True']:
1391 add_dash_mpd(get_video_info)
1392 if not video_info:
1393 video_info = get_video_info
1394 if 'token' in get_video_info:
1395 # Different get_video_info requests may report different results, e.g.
1396 # some may report video unavailability, but some may serve it without
1397 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1398 # the original webpage as well as el=info and el=embedded get_video_info
1399 # requests report video unavailability due to geo restriction while
1400 # el=detailpage succeeds and returns valid data). This is probably
1401 # due to YouTube measures against IP ranges of hosting providers.
1402 # Working around by preferring the first succeeded video_info containing
1403 # the token if no such video_info yet was found.
1404 if 'token' not in video_info:
1405 video_info = get_video_info
1406 break
1407 if 'token' not in video_info:
1408 if 'reason' in video_info:
1409 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1410 regions_allowed = self._html_search_meta(
1411 'regionsAllowed', video_webpage, default=None)
1412 countries = regions_allowed.split(',') if regions_allowed else None
1413 self.raise_geo_restricted(
1414 msg=video_info['reason'][0], countries=countries)
1415 raise ExtractorError(
1416 'YouTube said: %s' % video_info['reason'][0],
1417 expected=True, video_id=video_id)
1418 else:
1419 raise ExtractorError(
1420 '"token" parameter not in video info for unknown reason',
1421 video_id=video_id)
1422
1423 # title
1424 if 'title' in video_info:
1425 video_title = video_info['title'][0]
1426 else:
1427 self._downloader.report_warning('Unable to extract video title')
1428 video_title = '_'
1429
1430 # description
1431 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1432 if video_description:
1433 description_original = video_description = re.sub(r'''(?x)
1434 <a\s+
1435 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1436 (?:title|href)="([^"]+)"\s+
1437 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1438 class="[^"]*"[^>]*>
1439 [^<]+\.{3}\s*
1440 </a>
1441 ''', r'\1', video_description)
1442 video_description = clean_html(video_description)
1443 else:
1444 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1445 if fd_mobj:
1446 video_description = unescapeHTML(fd_mobj.group(1))
1447 else:
1448 video_description = ''
1449
1450 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1451 if not self._downloader.params.get('noplaylist'):
1452 entries = []
1453 feed_ids = []
1454 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1455 for feed in multifeed_metadata_list.split(','):
1456 # Unquote should take place before split on comma (,) since textual
1457 # fields may contain comma as well (see
1458 # https://github.com/rg3/youtube-dl/issues/8536)
1459 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1460 entries.append({
1461 '_type': 'url_transparent',
1462 'ie_key': 'Youtube',
1463 'url': smuggle_url(
1464 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1465 {'force_singlefeed': True}),
1466 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1467 })
1468 feed_ids.append(feed_data['id'][0])
1469 self.to_screen(
1470 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1471 % (', '.join(feed_ids), video_id))
1472 return self.playlist_result(entries, video_id, video_title, video_description)
1473 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1474
1475 if 'view_count' in video_info:
1476 view_count = int(video_info['view_count'][0])
1477 else:
1478 view_count = None
1479
1480 # Check for "rental" videos
1481 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1482 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1483
1484 # Start extracting information
1485 self.report_information_extraction(video_id)
1486
1487 # uploader
1488 if 'author' not in video_info:
1489 raise ExtractorError('Unable to extract uploader name')
1490 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1491
1492 # uploader_id
1493 video_uploader_id = None
1494 video_uploader_url = None
1495 mobj = re.search(
1496 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1497 video_webpage)
1498 if mobj is not None:
1499 video_uploader_id = mobj.group('uploader_id')
1500 video_uploader_url = mobj.group('uploader_url')
1501 else:
1502 self._downloader.report_warning('unable to extract uploader nickname')
1503
1504 # thumbnail image
1505 # We try first to get a high quality image:
1506 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1507 video_webpage, re.DOTALL)
1508 if m_thumb is not None:
1509 video_thumbnail = m_thumb.group(1)
1510 elif 'thumbnail_url' not in video_info:
1511 self._downloader.report_warning('unable to extract video thumbnail')
1512 video_thumbnail = None
1513 else: # don't panic if we can't find it
1514 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1515
1516 # upload date
1517 upload_date = self._html_search_meta(
1518 'datePublished', video_webpage, 'upload date', default=None)
1519 if not upload_date:
1520 upload_date = self._search_regex(
1521 [r'(?s)id="eow-date.*?>(.*?)</span>',
1522 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1523 video_webpage, 'upload date', default=None)
1524 if upload_date:
1525 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1526 upload_date = unified_strdate(upload_date)
1527
1528 video_license = self._html_search_regex(
1529 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1530 video_webpage, 'license', default=None)
1531
1532 m_music = re.search(
1533 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1534 video_webpage)
1535 if m_music:
1536 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1537 video_creator = clean_html(m_music.group('creator'))
1538 else:
1539 video_alt_title = video_creator = None
1540
1541 m_episode = re.search(
1542 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1543 video_webpage)
1544 if m_episode:
1545 series = m_episode.group('series')
1546 season_number = int(m_episode.group('season'))
1547 episode_number = int(m_episode.group('episode'))
1548 else:
1549 series = season_number = episode_number = None
1550
1551 m_cat_container = self._search_regex(
1552 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1553 video_webpage, 'categories', default=None)
1554 if m_cat_container:
1555 category = self._html_search_regex(
1556 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1557 default=None)
1558 video_categories = None if category is None else [category]
1559 else:
1560 video_categories = None
1561
1562 video_tags = [
1563 unescapeHTML(m.group('content'))
1564 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1565
1566 def _extract_count(count_name):
1567 return str_to_int(self._search_regex(
1568 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1569 % re.escape(count_name),
1570 video_webpage, count_name, default=None))
1571
1572 like_count = _extract_count('like')
1573 dislike_count = _extract_count('dislike')
1574
1575 # subtitles
1576 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1577 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1578
1579 video_duration = try_get(
1580 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1581 if not video_duration:
1582 video_duration = parse_duration(self._html_search_meta(
1583 'duration', video_webpage, 'video duration'))
1584
1585 # annotations
1586 video_annotations = None
1587 if self._downloader.params.get('writeannotations', False):
1588 video_annotations = self._extract_annotations(video_id)
1589
1590 chapters = self._extract_chapters(description_original, video_duration)
1591
1592 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1593 self.report_rtmp_download()
1594 formats = [{
1595 'format_id': '_rtmp',
1596 'protocol': 'rtmp',
1597 'url': video_info['conn'][0],
1598 'player_url': player_url,
1599 }]
1600 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1601 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1602 if 'rtmpe%3Dyes' in encoded_url_map:
1603 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1604 formats_spec = {}
1605 fmt_list = video_info.get('fmt_list', [''])[0]
1606 if fmt_list:
1607 for fmt in fmt_list.split(','):
1608 spec = fmt.split('/')
1609 if len(spec) > 1:
1610 width_height = spec[1].split('x')
1611 if len(width_height) == 2:
1612 formats_spec[spec[0]] = {
1613 'resolution': spec[1],
1614 'width': int_or_none(width_height[0]),
1615 'height': int_or_none(width_height[1]),
1616 }
1617 formats = []
1618 for url_data_str in encoded_url_map.split(','):
1619 url_data = compat_parse_qs(url_data_str)
1620 if 'itag' not in url_data or 'url' not in url_data:
1621 continue
1622 format_id = url_data['itag'][0]
1623 url = url_data['url'][0]
1624
1625 if 'sig' in url_data:
1626 url += '&signature=' + url_data['sig'][0]
1627 elif 's' in url_data:
1628 encrypted_sig = url_data['s'][0]
1629 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1630
1631 jsplayer_url_json = self._search_regex(
1632 ASSETS_RE,
1633 embed_webpage if age_gate else video_webpage,
1634 'JS player URL (1)', default=None)
1635 if not jsplayer_url_json and not age_gate:
1636 # We need the embed website after all
1637 if embed_webpage is None:
1638 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1639 embed_webpage = self._download_webpage(
1640 embed_url, video_id, 'Downloading embed webpage')
1641 jsplayer_url_json = self._search_regex(
1642 ASSETS_RE, embed_webpage, 'JS player URL')
1643
1644 player_url = json.loads(jsplayer_url_json)
1645 if player_url is None:
1646 player_url_json = self._search_regex(
1647 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1648 video_webpage, 'age gate player URL')
1649 player_url = json.loads(player_url_json)
1650
1651 if self._downloader.params.get('verbose'):
1652 if player_url is None:
1653 player_version = 'unknown'
1654 player_desc = 'unknown'
1655 else:
1656 if player_url.endswith('swf'):
1657 player_version = self._search_regex(
1658 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1659 'flash player', fatal=False)
1660 player_desc = 'flash player %s' % player_version
1661 else:
1662 player_version = self._search_regex(
1663 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1664 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1665 player_url,
1666 'html5 player', fatal=False)
1667 player_desc = 'html5 player %s' % player_version
1668
1669 parts_sizes = self._signature_cache_id(encrypted_sig)
1670 self.to_screen('{%s} signature length %s, %s' %
1671 (format_id, parts_sizes, player_desc))
1672
1673 signature = self._decrypt_signature(
1674 encrypted_sig, video_id, player_url, age_gate)
1675 url += '&signature=' + signature
1676 if 'ratebypass' not in url:
1677 url += '&ratebypass=yes'
1678
1679 dct = {
1680 'format_id': format_id,
1681 'url': url,
1682 'player_url': player_url,
1683 }
1684 if format_id in self._formats:
1685 dct.update(self._formats[format_id])
1686 if format_id in formats_spec:
1687 dct.update(formats_spec[format_id])
1688
1689 # Some itags are not included in DASH manifest thus corresponding formats will
1690 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1691 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1692 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1693 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1694
1695 more_fields = {
1696 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1697 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1698 'width': width,
1699 'height': height,
1700 'fps': int_or_none(url_data.get('fps', [None])[0]),
1701 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1702 }
1703 for key, value in more_fields.items():
1704 if value:
1705 dct[key] = value
1706 type_ = url_data.get('type', [None])[0]
1707 if type_:
1708 type_split = type_.split(';')
1709 kind_ext = type_split[0].split('/')
1710 if len(kind_ext) == 2:
1711 kind, _ = kind_ext
1712 dct['ext'] = mimetype2ext(type_split[0])
1713 if kind in ('audio', 'video'):
1714 codecs = None
1715 for mobj in re.finditer(
1716 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1717 if mobj.group('key') == 'codecs':
1718 codecs = mobj.group('val')
1719 break
1720 if codecs:
1721 dct.update(parse_codecs(codecs))
1722 formats.append(dct)
1723 elif video_info.get('hlsvp'):
1724 manifest_url = video_info['hlsvp'][0]
1725 formats = []
1726 m3u8_formats = self._extract_m3u8_formats(
1727 manifest_url, video_id, 'mp4', fatal=False)
1728 for a_format in m3u8_formats:
1729 itag = self._search_regex(
1730 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1731 if itag:
1732 a_format['format_id'] = itag
1733 if itag in self._formats:
1734 dct = self._formats[itag].copy()
1735 dct.update(a_format)
1736 a_format = dct
1737 a_format['player_url'] = player_url
1738 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1739 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1740 formats.append(a_format)
1741 else:
1742 unavailable_message = self._html_search_regex(
1743 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1744 video_webpage, 'unavailable message', default=None)
1745 if unavailable_message:
1746 raise ExtractorError(unavailable_message, expected=True)
1747 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1748
1749 # Look for the DASH manifest
1750 if self._downloader.params.get('youtube_include_dash_manifest', True):
1751 dash_mpd_fatal = True
1752 for mpd_url in dash_mpds:
1753 dash_formats = {}
1754 try:
1755 def decrypt_sig(mobj):
1756 s = mobj.group(1)
1757 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1758 return '/signature/%s' % dec_s
1759
1760 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1761
1762 for df in self._extract_mpd_formats(
1763 mpd_url, video_id, fatal=dash_mpd_fatal,
1764 formats_dict=self._formats):
1765 # Do not overwrite DASH format found in some previous DASH manifest
1766 if df['format_id'] not in dash_formats:
1767 dash_formats[df['format_id']] = df
1768 # Additional DASH manifests may end up in HTTP Error 403 therefore
1769 # allow them to fail without bug report message if we already have
1770 # some DASH manifest succeeded. This is temporary workaround to reduce
1771 # burst of bug reports until we figure out the reason and whether it
1772 # can be fixed at all.
1773 dash_mpd_fatal = False
1774 except (ExtractorError, KeyError) as e:
1775 self.report_warning(
1776 'Skipping DASH manifest: %r' % e, video_id)
1777 if dash_formats:
1778 # Remove the formats we found through non-DASH, they
1779 # contain less info and it can be wrong, because we use
1780 # fixed values (for example the resolution). See
1781 # https://github.com/rg3/youtube-dl/issues/5774 for an
1782 # example.
1783 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1784 formats.extend(dash_formats.values())
1785
1786 # Check for malformed aspect ratio
1787 stretched_m = re.search(
1788 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1789 video_webpage)
1790 if stretched_m:
1791 w = float(stretched_m.group('w'))
1792 h = float(stretched_m.group('h'))
1793 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1794 # We will only process correct ratios.
1795 if w > 0 and h > 0:
1796 ratio = w / h
1797 for f in formats:
1798 if f.get('vcodec') != 'none':
1799 f['stretched_ratio'] = ratio
1800
1801 self._sort_formats(formats)
1802
1803 self.mark_watched(video_id, video_info)
1804
1805 return {
1806 'id': video_id,
1807 'uploader': video_uploader,
1808 'uploader_id': video_uploader_id,
1809 'uploader_url': video_uploader_url,
1810 'upload_date': upload_date,
1811 'license': video_license,
1812 'creator': video_creator,
1813 'title': video_title,
1814 'alt_title': video_alt_title,
1815 'thumbnail': video_thumbnail,
1816 'description': video_description,
1817 'categories': video_categories,
1818 'tags': video_tags,
1819 'subtitles': video_subtitles,
1820 'automatic_captions': automatic_captions,
1821 'duration': video_duration,
1822 'age_limit': 18 if age_gate else 0,
1823 'annotations': video_annotations,
1824 'chapters': chapters,
1825 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1826 'view_count': view_count,
1827 'like_count': like_count,
1828 'dislike_count': dislike_count,
1829 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1830 'formats': formats,
1831 'is_live': is_live,
1832 'start_time': start_time,
1833 'end_time': end_time,
1834 'series': series,
1835 'season_number': season_number,
1836 'episode_number': episode_number,
1837 }
1838
1839
1840 class YoutubeSharedVideoIE(InfoExtractor):
1841 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1842 IE_NAME = 'youtube:shared'
1843
1844 _TEST = {
1845 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1846 'info_dict': {
1847 'id': 'uPDB5I9wfp8',
1848 'ext': 'webm',
1849 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1850 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1851 'upload_date': '20160219',
1852 'uploader': 'Pocoyo - Português (BR)',
1853 'uploader_id': 'PocoyoBrazil',
1854 },
1855 'add_ie': ['Youtube'],
1856 'params': {
1857 # There are already too many Youtube downloads
1858 'skip_download': True,
1859 },
1860 }
1861
1862 def _real_extract(self, url):
1863 video_id = self._match_id(url)
1864
1865 webpage = self._download_webpage(url, video_id)
1866
1867 real_video_id = self._html_search_meta(
1868 'videoId', webpage, 'YouTube video id', fatal=True)
1869
1870 return self.url_result(real_video_id, YoutubeIE.ie_key())
1871
1872
1873 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1874 IE_DESC = 'YouTube.com playlists'
1875 _VALID_URL = r"""(?x)(?:
1876 (?:https?://)?
1877 (?:\w+\.)?
1878 (?:
1879 youtube\.com/
1880 (?:
1881 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
1882 \? (?:.*?[&;])*? (?:p|a|list)=
1883 | p/
1884 )|
1885 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
1886 )
1887 (
1888 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
1889 # Top tracks, they can also include dots
1890 |(?:MC)[\w\.]*
1891 )
1892 .*
1893 |
1894 (%(playlist_id)s)
1895 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
1896 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
1897 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1898 IE_NAME = 'youtube:playlist'
1899 _TESTS = [{
1900 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1901 'info_dict': {
1902 'title': 'ytdl test PL',
1903 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1904 },
1905 'playlist_count': 3,
1906 }, {
1907 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1908 'info_dict': {
1909 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1910 'title': 'YDL_Empty_List',
1911 },
1912 'playlist_count': 0,
1913 'skip': 'This playlist is private',
1914 }, {
1915 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1916 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1917 'info_dict': {
1918 'title': '29C3: Not my department',
1919 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1920 },
1921 'playlist_count': 95,
1922 }, {
1923 'note': 'issue #673',
1924 'url': 'PLBB231211A4F62143',
1925 'info_dict': {
1926 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1927 'id': 'PLBB231211A4F62143',
1928 },
1929 'playlist_mincount': 26,
1930 }, {
1931 'note': 'Large playlist',
1932 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1933 'info_dict': {
1934 'title': 'Uploads from Cauchemar',
1935 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1936 },
1937 'playlist_mincount': 799,
1938 }, {
1939 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1940 'info_dict': {
1941 'title': 'YDL_safe_search',
1942 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1943 },
1944 'playlist_count': 2,
1945 'skip': 'This playlist is private',
1946 }, {
1947 'note': 'embedded',
1948 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1949 'playlist_count': 4,
1950 'info_dict': {
1951 'title': 'JODA15',
1952 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1953 }
1954 }, {
1955 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
1956 'playlist_mincount': 485,
1957 'info_dict': {
1958 'title': '2017 華語最新單曲 (2/24更新)',
1959 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
1960 }
1961 }, {
1962 'note': 'Embedded SWF player',
1963 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1964 'playlist_count': 4,
1965 'info_dict': {
1966 'title': 'JODA7',
1967 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1968 }
1969 }, {
1970 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1971 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1972 'info_dict': {
1973 'title': 'Uploads from Interstellar Movie',
1974 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1975 },
1976 'playlist_mincount': 21,
1977 }, {
1978 # Playlist URL that does not actually serve a playlist
1979 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
1980 'info_dict': {
1981 'id': 'FqZTN594JQw',
1982 'ext': 'webm',
1983 'title': "Smiley's People 01 detective, Adventure Series, Action",
1984 'uploader': 'STREEM',
1985 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
1986 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
1987 'upload_date': '20150526',
1988 'license': 'Standard YouTube License',
1989 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
1990 'categories': ['People & Blogs'],
1991 'tags': list,
1992 'like_count': int,
1993 'dislike_count': int,
1994 },
1995 'params': {
1996 'skip_download': True,
1997 },
1998 'add_ie': [YoutubeIE.ie_key()],
1999 }, {
2000 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2001 'info_dict': {
2002 'id': 'yeWKywCrFtk',
2003 'ext': 'mp4',
2004 'title': 'Small Scale Baler and Braiding Rugs',
2005 'uploader': 'Backus-Page House Museum',
2006 'uploader_id': 'backuspagemuseum',
2007 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2008 'upload_date': '20161008',
2009 'license': 'Standard YouTube License',
2010 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2011 'categories': ['Nonprofits & Activism'],
2012 'tags': list,
2013 'like_count': int,
2014 'dislike_count': int,
2015 },
2016 'params': {
2017 'noplaylist': True,
2018 'skip_download': True,
2019 },
2020 }, {
2021 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2022 'only_matching': True,
2023 }, {
2024 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2025 'only_matching': True,
2026 }]
2027
2028 def _real_initialize(self):
2029 self._login()
2030
2031 def _extract_mix(self, playlist_id):
2032 # The mixes are generated from a single video
2033 # the id of the playlist is just 'RD' + video_id
2034 ids = []
2035 last_id = playlist_id[-11:]
2036 for n in itertools.count(1):
2037 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2038 webpage = self._download_webpage(
2039 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2040 new_ids = orderedSet(re.findall(
2041 r'''(?xs)data-video-username=".*?".*?
2042 href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2043 webpage))
2044 # Fetch new pages until all the videos are repeated, it seems that
2045 # there are always 51 unique videos.
2046 new_ids = [_id for _id in new_ids if _id not in ids]
2047 if not new_ids:
2048 break
2049 ids.extend(new_ids)
2050 last_id = ids[-1]
2051
2052 url_results = self._ids_to_results(ids)
2053
2054 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2055 title_span = (
2056 search_title('playlist-title') or
2057 search_title('title long-title') or
2058 search_title('title'))
2059 title = clean_html(title_span)
2060
2061 return self.playlist_result(url_results, playlist_id, title)
2062
2063 def _extract_playlist(self, playlist_id):
2064 url = self._TEMPLATE_URL % playlist_id
2065 page = self._download_webpage(url, playlist_id)
2066
2067 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2068 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2069 match = match.strip()
2070 # Check if the playlist exists or is private
2071 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2072 if mobj:
2073 reason = mobj.group('reason')
2074 message = 'This playlist %s' % reason
2075 if 'private' in reason:
2076 message += ', use --username or --netrc to access it'
2077 message += '.'
2078 raise ExtractorError(message, expected=True)
2079 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2080 raise ExtractorError(
2081 'Invalid parameters. Maybe URL is incorrect.',
2082 expected=True)
2083 elif re.match(r'[^<]*Choose your language[^<]*', match):
2084 continue
2085 else:
2086 self.report_warning('Youtube gives an alert message: ' + match)
2087
2088 playlist_title = self._html_search_regex(
2089 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2090 page, 'title', default=None)
2091
2092 has_videos = True
2093
2094 if not playlist_title:
2095 try:
2096 # Some playlist URLs don't actually serve a playlist (e.g.
2097 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2098 next(self._entries(page, playlist_id))
2099 except StopIteration:
2100 has_videos = False
2101
2102 return has_videos, self.playlist_result(
2103 self._entries(page, playlist_id), playlist_id, playlist_title)
2104
2105 def _check_download_just_video(self, url, playlist_id):
2106 # Check if it's a video-specific URL
2107 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2108 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2109 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2110 'video id', default=None)
2111 if video_id:
2112 if self._downloader.params.get('noplaylist'):
2113 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2114 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2115 else:
2116 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2117 return video_id, None
2118 return None, None
2119
2120 def _real_extract(self, url):
2121 # Extract playlist id
2122 mobj = re.match(self._VALID_URL, url)
2123 if mobj is None:
2124 raise ExtractorError('Invalid URL: %s' % url)
2125 playlist_id = mobj.group(1) or mobj.group(2)
2126
2127 video_id, video = self._check_download_just_video(url, playlist_id)
2128 if video:
2129 return video
2130
2131 if playlist_id.startswith(('RD', 'UL', 'PU')):
2132 # Mixes require a custom extraction process
2133 return self._extract_mix(playlist_id)
2134
2135 has_videos, playlist = self._extract_playlist(playlist_id)
2136 if has_videos or not video_id:
2137 return playlist
2138
2139 # Some playlist URLs don't actually serve a playlist (see
2140 # https://github.com/rg3/youtube-dl/issues/10537).
2141 # Fallback to plain video extraction if there is a video id
2142 # along with playlist id.
2143 return self.url_result(video_id, 'Youtube', video_id=video_id)
2144
2145
2146 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2147 IE_DESC = 'YouTube.com channels'
2148 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2149 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2150 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2151 IE_NAME = 'youtube:channel'
2152 _TESTS = [{
2153 'note': 'paginated channel',
2154 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2155 'playlist_mincount': 91,
2156 'info_dict': {
2157 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2158 'title': 'Uploads from lex will',
2159 }
2160 }, {
2161 'note': 'Age restricted channel',
2162 # from https://www.youtube.com/user/DeusExOfficial
2163 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2164 'playlist_mincount': 64,
2165 'info_dict': {
2166 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2167 'title': 'Uploads from Deus Ex',
2168 },
2169 }]
2170
2171 @classmethod
2172 def suitable(cls, url):
2173 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2174 else super(YoutubeChannelIE, cls).suitable(url))
2175
2176 def _build_template_url(self, url, channel_id):
2177 return self._TEMPLATE_URL % channel_id
2178
2179 def _real_extract(self, url):
2180 channel_id = self._match_id(url)
2181
2182 url = self._build_template_url(url, channel_id)
2183
2184 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2185 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2186 # otherwise fallback on channel by page extraction
2187 channel_page = self._download_webpage(
2188 url + '?view=57', channel_id,
2189 'Downloading channel page', fatal=False)
2190 if channel_page is False:
2191 channel_playlist_id = False
2192 else:
2193 channel_playlist_id = self._html_search_meta(
2194 'channelId', channel_page, 'channel id', default=None)
2195 if not channel_playlist_id:
2196 channel_url = self._html_search_meta(
2197 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2198 channel_page, 'channel url', default=None)
2199 if channel_url:
2200 channel_playlist_id = self._search_regex(
2201 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2202 channel_url, 'channel id', default=None)
2203 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2204 playlist_id = 'UU' + channel_playlist_id[2:]
2205 return self.url_result(
2206 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2207
2208 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2209 autogenerated = re.search(r'''(?x)
2210 class="[^"]*?(?:
2211 channel-header-autogenerated-label|
2212 yt-channel-title-autogenerated
2213 )[^"]*"''', channel_page) is not None
2214
2215 if autogenerated:
2216 # The videos are contained in a single page
2217 # the ajax pages can't be used, they are empty
2218 entries = [
2219 self.url_result(
2220 video_id, 'Youtube', video_id=video_id,
2221 video_title=video_title)
2222 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2223 return self.playlist_result(entries, channel_id)
2224
2225 try:
2226 next(self._entries(channel_page, channel_id))
2227 except StopIteration:
2228 alert_message = self._html_search_regex(
2229 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2230 channel_page, 'alert', default=None, group='alert')
2231 if alert_message:
2232 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2233
2234 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2235
2236
2237 class YoutubeUserIE(YoutubeChannelIE):
2238 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2239 _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_-]+)'
2240 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2241 IE_NAME = 'youtube:user'
2242
2243 _TESTS = [{
2244 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2245 'playlist_mincount': 320,
2246 'info_dict': {
2247 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2248 'title': 'Uploads from The Linux Foundation',
2249 }
2250 }, {
2251 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2252 # but not https://www.youtube.com/user/12minuteathlete/videos
2253 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2254 'playlist_mincount': 249,
2255 'info_dict': {
2256 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2257 'title': 'Uploads from 12 Minute Athlete',
2258 }
2259 }, {
2260 'url': 'ytuser:phihag',
2261 'only_matching': True,
2262 }, {
2263 'url': 'https://www.youtube.com/c/gametrailers',
2264 'only_matching': True,
2265 }, {
2266 'url': 'https://www.youtube.com/gametrailers',
2267 'only_matching': True,
2268 }, {
2269 # This channel is not available, geo restricted to JP
2270 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2271 'only_matching': True,
2272 }]
2273
2274 @classmethod
2275 def suitable(cls, url):
2276 # Don't return True if the url can be extracted with other youtube
2277 # extractor, the regex would is too permissive and it would match.
2278 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2279 if any(ie.suitable(url) for ie in other_yt_ies):
2280 return False
2281 else:
2282 return super(YoutubeUserIE, cls).suitable(url)
2283
2284 def _build_template_url(self, url, channel_id):
2285 mobj = re.match(self._VALID_URL, url)
2286 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2287
2288
2289 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2290 IE_DESC = 'YouTube.com live streams'
2291 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2292 IE_NAME = 'youtube:live'
2293
2294 _TESTS = [{
2295 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2296 'info_dict': {
2297 'id': 'a48o2S1cPoo',
2298 'ext': 'mp4',
2299 'title': 'The Young Turks - Live Main Show',
2300 'uploader': 'The Young Turks',
2301 'uploader_id': 'TheYoungTurks',
2302 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2303 'upload_date': '20150715',
2304 'license': 'Standard YouTube License',
2305 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2306 'categories': ['News & Politics'],
2307 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2308 'like_count': int,
2309 'dislike_count': int,
2310 },
2311 'params': {
2312 'skip_download': True,
2313 },
2314 }, {
2315 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2316 'only_matching': True,
2317 }, {
2318 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2319 'only_matching': True,
2320 }, {
2321 'url': 'https://www.youtube.com/TheYoungTurks/live',
2322 'only_matching': True,
2323 }]
2324
2325 def _real_extract(self, url):
2326 mobj = re.match(self._VALID_URL, url)
2327 channel_id = mobj.group('id')
2328 base_url = mobj.group('base_url')
2329 webpage = self._download_webpage(url, channel_id, fatal=False)
2330 if webpage:
2331 page_type = self._og_search_property(
2332 'type', webpage, 'page type', default=None)
2333 video_id = self._html_search_meta(
2334 'videoId', webpage, 'video id', default=None)
2335 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2336 return self.url_result(video_id, YoutubeIE.ie_key())
2337 return self.url_result(base_url)
2338
2339
2340 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2341 IE_DESC = 'YouTube.com user/channel playlists'
2342 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2343 IE_NAME = 'youtube:playlists'
2344
2345 _TESTS = [{
2346 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2347 'playlist_mincount': 4,
2348 'info_dict': {
2349 'id': 'ThirstForScience',
2350 'title': 'Thirst for Science',
2351 },
2352 }, {
2353 # with "Load more" button
2354 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2355 'playlist_mincount': 70,
2356 'info_dict': {
2357 'id': 'igorkle1',
2358 'title': 'Игорь Клейнер',
2359 },
2360 }, {
2361 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2362 'playlist_mincount': 17,
2363 'info_dict': {
2364 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2365 'title': 'Chem Player',
2366 },
2367 }]
2368
2369
2370 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2371 IE_DESC = 'YouTube.com searches'
2372 # there doesn't appear to be a real limit, for example if you search for
2373 # 'python' you get more than 8.000.000 results
2374 _MAX_RESULTS = float('inf')
2375 IE_NAME = 'youtube:search'
2376 _SEARCH_KEY = 'ytsearch'
2377 _EXTRA_QUERY_ARGS = {}
2378 _TESTS = []
2379
2380 def _get_n_results(self, query, n):
2381 """Get a specified number of results for a query"""
2382
2383 videos = []
2384 limit = n
2385
2386 url_query = {
2387 'search_query': query.encode('utf-8'),
2388 }
2389 url_query.update(self._EXTRA_QUERY_ARGS)
2390 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2391
2392 for pagenum in itertools.count(1):
2393 data = self._download_json(
2394 result_url, video_id='query "%s"' % query,
2395 note='Downloading page %s' % pagenum,
2396 errnote='Unable to download API page',
2397 query={'spf': 'navigate'})
2398 html_content = data[1]['body']['content']
2399
2400 if 'class="search-message' in html_content:
2401 raise ExtractorError(
2402 '[youtube] No video results', expected=True)
2403
2404 new_videos = self._ids_to_results(orderedSet(re.findall(
2405 r'href="/watch\?v=(.{11})', html_content)))
2406 videos += new_videos
2407 if not new_videos or len(videos) > limit:
2408 break
2409 next_link = self._html_search_regex(
2410 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2411 html_content, 'next link', default=None)
2412 if next_link is None:
2413 break
2414 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2415
2416 if len(videos) > n:
2417 videos = videos[:n]
2418 return self.playlist_result(videos, query)
2419
2420
2421 class YoutubeSearchDateIE(YoutubeSearchIE):
2422 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2423 _SEARCH_KEY = 'ytsearchdate'
2424 IE_DESC = 'YouTube.com searches, newest videos first'
2425 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2426
2427
2428 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2429 IE_DESC = 'YouTube.com search URLs'
2430 IE_NAME = 'youtube:search_url'
2431 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2432 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2433 _TESTS = [{
2434 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2435 'playlist_mincount': 5,
2436 'info_dict': {
2437 'title': 'youtube-dl test video',
2438 }
2439 }, {
2440 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2441 'only_matching': True,
2442 }]
2443
2444 def _real_extract(self, url):
2445 mobj = re.match(self._VALID_URL, url)
2446 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2447 webpage = self._download_webpage(url, query)
2448 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2449
2450
2451 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2452 IE_DESC = 'YouTube.com (multi-season) shows'
2453 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2454 IE_NAME = 'youtube:show'
2455 _TESTS = [{
2456 'url': 'https://www.youtube.com/show/airdisasters',
2457 'playlist_mincount': 5,
2458 'info_dict': {
2459 'id': 'airdisasters',
2460 'title': 'Air Disasters',
2461 }
2462 }]
2463
2464 def _real_extract(self, url):
2465 playlist_id = self._match_id(url)
2466 return super(YoutubeShowIE, self)._real_extract(
2467 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2468
2469
2470 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2471 """
2472 Base class for feed extractors
2473 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2474 """
2475 _LOGIN_REQUIRED = True
2476
2477 @property
2478 def IE_NAME(self):
2479 return 'youtube:%s' % self._FEED_NAME
2480
2481 def _real_initialize(self):
2482 self._login()
2483
2484 def _real_extract(self, url):
2485 page = self._download_webpage(
2486 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2487
2488 # The extraction process is the same as for playlists, but the regex
2489 # for the video ids doesn't contain an index
2490 ids = []
2491 more_widget_html = content_html = page
2492 for page_num in itertools.count(1):
2493 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2494
2495 # 'recommended' feed has infinite 'load more' and each new portion spins
2496 # the same videos in (sometimes) slightly different order, so we'll check
2497 # for unicity and break when portion has no new videos
2498 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2499 if not new_ids:
2500 break
2501
2502 ids.extend(new_ids)
2503
2504 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2505 if not mobj:
2506 break
2507
2508 more = self._download_json(
2509 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2510 'Downloading page #%s' % page_num,
2511 transform_source=uppercase_escape)
2512 content_html = more['content_html']
2513 more_widget_html = more['load_more_widget_html']
2514
2515 return self.playlist_result(
2516 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2517
2518
2519 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2520 IE_NAME = 'youtube:watchlater'
2521 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2522 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2523
2524 _TESTS = [{
2525 'url': 'https://www.youtube.com/playlist?list=WL',
2526 'only_matching': True,
2527 }, {
2528 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2529 'only_matching': True,
2530 }]
2531
2532 def _real_extract(self, url):
2533 _, video = self._check_download_just_video(url, 'WL')
2534 if video:
2535 return video
2536 _, playlist = self._extract_playlist('WL')
2537 return playlist
2538
2539
2540 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2541 IE_NAME = 'youtube:favorites'
2542 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2543 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2544 _LOGIN_REQUIRED = True
2545
2546 def _real_extract(self, url):
2547 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2548 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2549 return self.url_result(playlist_id, 'YoutubePlaylist')
2550
2551
2552 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2553 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2554 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2555 _FEED_NAME = 'recommended'
2556 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2557
2558
2559 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2560 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2561 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2562 _FEED_NAME = 'subscriptions'
2563 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2564
2565
2566 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2567 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2568 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2569 _FEED_NAME = 'history'
2570 _PLAYLIST_TITLE = 'Youtube History'
2571
2572
2573 class YoutubeTruncatedURLIE(InfoExtractor):
2574 IE_NAME = 'youtube:truncated_url'
2575 IE_DESC = False # Do not list
2576 _VALID_URL = r'''(?x)
2577 (?:https?://)?
2578 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2579 (?:watch\?(?:
2580 feature=[a-z_]+|
2581 annotation_id=annotation_[^&]+|
2582 x-yt-cl=[0-9]+|
2583 hl=[^&]*|
2584 t=[0-9]+
2585 )?
2586 |
2587 attribution_link\?a=[^&]+
2588 )
2589 $
2590 '''
2591
2592 _TESTS = [{
2593 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2594 'only_matching': True,
2595 }, {
2596 'url': 'https://www.youtube.com/watch?',
2597 'only_matching': True,
2598 }, {
2599 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2600 'only_matching': True,
2601 }, {
2602 'url': 'https://www.youtube.com/watch?feature=foo',
2603 'only_matching': True,
2604 }, {
2605 'url': 'https://www.youtube.com/watch?hl=en-GB',
2606 'only_matching': True,
2607 }, {
2608 'url': 'https://www.youtube.com/watch?t=2372',
2609 'only_matching': True,
2610 }]
2611
2612 def _real_extract(self, url):
2613 raise ExtractorError(
2614 'Did you forget to quote the URL? Remember that & is a meta '
2615 'character in most shells, so you want to put the URL in quotes, '
2616 'like youtube-dl '
2617 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2618 ' or simply youtube-dl BaW_jenozKc .',
2619 expected=True)
2620
2621
2622 class YoutubeTruncatedIDIE(InfoExtractor):
2623 IE_NAME = 'youtube:truncated_id'
2624 IE_DESC = False # Do not list
2625 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2626
2627 _TESTS = [{
2628 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2629 'only_matching': True,
2630 }]
2631
2632 def _real_extract(self, url):
2633 video_id = self._match_id(url)
2634 raise ExtractorError(
2635 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
2636 expected=True)