3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
23 compat_urllib_request,
31 get_element_by_attribute,
46 class YoutubeBaseInfoExtractor(InfoExtractor):
47 """Provide base functions for Youtube extractors"""
48 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
49 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
50 _NETRC_MACHINE = 'youtube'
51 # If True it will raise an error if no login info is provided
52 _LOGIN_REQUIRED = False
54 def _set_language(self):
56 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
57 # YouTube sets the expire time to about two months
58 expire_time=time.time() + 2 * 30 * 24 * 3600)
60 def _ids_to_results(self, ids):
62 self.url_result(vid_id, 'Youtube', video_id=vid_id)
67 Attempt to log in to YouTube.
68 True is returned if successful or skipped.
69 False is returned if login failed.
71 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
73 (username, password) = self._get_login_info()
74 # No authentication to be performed
76 if self._LOGIN_REQUIRED:
77 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
80 login_page = self._download_webpage(
81 self._LOGIN_URL, None,
82 note='Downloading login page',
83 errnote='unable to fetch login page', fatal=False)
84 if login_page is False:
87 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
88 login_page, 'Login GALX parameter')
92 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
97 'PersistentCookie': 'yes',
99 'bgresponse': 'js_disabled',
100 'checkConnection': '',
101 'checkedDomains': 'youtube',
108 'service': 'youtube',
113 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
115 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
116 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
118 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
119 login_results = self._download_webpage(
121 note='Logging in', errnote='unable to log in', fatal=False)
122 if login_results is False:
125 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
126 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
129 # TODO add SMS and phone call support - these require making a request and then prompting the user
131 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
132 tfa_code = self._get_tfa_info()
135 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
136 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
139 def find_value(element_id):
140 match = re.search(r'id="%s"\s+value="(.+?)">' % element_id, login_results, re.M | re.U)
142 self._downloader.report_warning('Failed to get %s - did the page structure change?' % id)
143 return match.group(1)
145 challengeId = find_value('challengeId')
146 challengeType = find_value('challengeType')
147 gxf = find_value('gxf')
150 'challengeId': challengeId,
151 'challengeType': challengeType, # This doesn't appear to change
152 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
153 'service': 'youtube',
155 'checkedDomains': 'youtube',
161 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
162 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
164 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
165 tfa_results = self._download_webpage(
167 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
169 if tfa_results is False:
172 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
173 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
175 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
176 self._downloader.report_warning('unable to log in - did the page structure change?')
178 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
179 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
182 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
183 self._downloader.report_warning('unable to log in: bad username or password')
187 def _real_initialize(self):
188 if self._downloader is None:
191 if not self._login():
195 class YoutubeIE(YoutubeBaseInfoExtractor):
196 IE_DESC = 'YouTube.com'
197 _VALID_URL = r"""(?x)^
199 (?:https?://|//) # http(s):// or protocol-independent URL
200 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
201 (?:www\.)?deturl\.com/www\.youtube\.com/|
202 (?:www\.)?pwnyoutube\.com/|
203 (?:www\.)?yourepeat\.com/|
204 tube\.majestyc\.net/|
205 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
206 (?:.*?\#/)? # handle anchor (#/) redirect urls
207 (?: # the various things that can precede the ID:
208 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
209 |(?: # or the v= param in all its forms
210 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
211 (?:\?|\#!?) # the params delimiter ? or # or #!
212 (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx)
216 |youtu\.be/ # just youtu.be/xxxx
217 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
219 )? # all until now is optional -> you can pass the naked ID
220 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
221 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
222 (?(1).+)? # if we found the ID, everything can follow
224 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
226 '5': {'ext': 'flv', 'width': 400, 'height': 240},
227 '6': {'ext': 'flv', 'width': 450, 'height': 270},
228 '13': {'ext': '3gp'},
229 '17': {'ext': '3gp', 'width': 176, 'height': 144},
230 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
231 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
232 '34': {'ext': 'flv', 'width': 640, 'height': 360},
233 '35': {'ext': 'flv', 'width': 854, 'height': 480},
234 '36': {'ext': '3gp', 'width': 320, 'height': 240},
235 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
236 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
237 '43': {'ext': 'webm', 'width': 640, 'height': 360},
238 '44': {'ext': 'webm', 'width': 854, 'height': 480},
239 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
240 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
241 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
242 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
246 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
247 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
248 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
249 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
250 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
251 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
252 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
254 # Apple HTTP Live Streaming
255 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
256 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
257 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
258 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
259 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
260 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
261 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
264 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
265 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
266 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
268 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
270 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
273 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
274 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
277 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
278 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
279 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
282 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
283 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
284 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
285 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
286 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
287 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
288 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'},
289 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
299 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
300 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
301 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
302 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
305 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
306 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
308 # Dash webm audio with opus inside
309 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
310 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
311 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
314 '_rtmp': {'protocol': 'rtmp'},
320 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
324 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
325 'uploader': 'Philipp Hagemeister',
326 'uploader_id': 'phihag',
327 'upload_date': '20121002',
328 '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 .',
329 'categories': ['Science & Technology'],
330 'tags': ['youtube-dl'],
332 'dislike_count': int,
338 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
339 'note': 'Test generic use_cipher_signature video (#897)',
343 'upload_date': '20120506',
344 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
345 'description': 'md5:782e8651347686cba06e58f71ab51773',
346 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
347 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
348 'iconic ep', 'iconic', 'love', 'it'],
349 'uploader': 'Icona Pop',
350 'uploader_id': 'IconaPop',
354 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
355 'note': 'Test VEVO video with age protection (#956)',
359 'upload_date': '20130703',
360 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
361 'description': 'md5:64249768eec3bc4276236606ea996373',
362 'uploader': 'justintimberlakeVEVO',
363 'uploader_id': 'justintimberlakeVEVO',
368 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
369 'note': 'Embed-only video (#1746)',
373 'upload_date': '20120608',
374 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
375 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
376 'uploader': 'SET India',
377 'uploader_id': 'setindia'
381 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
382 'note': 'Use the first video ID in the URL',
386 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
387 'uploader': 'Philipp Hagemeister',
388 'uploader_id': 'phihag',
389 'upload_date': '20121002',
390 '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 .',
391 'categories': ['Science & Technology'],
392 'tags': ['youtube-dl'],
394 'dislike_count': int,
397 'skip_download': True,
401 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
402 'note': '256k DASH audio (format 141) via DASH manifest',
406 'upload_date': '20121002',
407 'uploader_id': '8KVIDEO',
409 'uploader': '8KVIDEO',
410 'title': 'UHDTV TEST 8K VIDEO.mp4'
413 'youtube_include_dash_manifest': True,
417 # DASH manifest with encrypted signature
419 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
423 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
424 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
425 'uploader': 'AfrojackVEVO',
426 'uploader_id': 'AfrojackVEVO',
427 'upload_date': '20131011',
430 'youtube_include_dash_manifest': True,
434 # JS player signature function name containing $
436 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
440 'title': 'Taylor Swift - Shake It Off',
441 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
442 'uploader': 'TaylorSwiftVEVO',
443 'uploader_id': 'TaylorSwiftVEVO',
444 'upload_date': '20140818',
447 'youtube_include_dash_manifest': True,
453 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
457 'upload_date': '20100909',
458 'uploader': 'The Amazing Atheist',
459 'uploader_id': 'TheAmazingAtheist',
460 'title': 'Burning Everyone\'s Koran',
461 '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',
464 # Normal age-gate video (No vevo, embed allowed)
466 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
470 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
471 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
472 'uploader': 'The Witcher',
473 'uploader_id': 'WitcherGame',
474 'upload_date': '20140605',
478 # Age-gate video with encrypted signature
480 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
484 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
485 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
486 'uploader': 'LloydVEVO',
487 'uploader_id': 'LloydVEVO',
488 'upload_date': '20110629',
492 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
494 'url': '__2ABJjxzNo',
498 'upload_date': '20100430',
499 'uploader_id': 'deadmau5',
500 'description': 'md5:12c56784b8032162bb936a5f76d55360',
501 'uploader': 'deadmau5',
502 'title': 'Deadmau5 - Some Chords (HD)',
504 'expected_warnings': [
505 'DASH manifest missing',
508 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
510 'url': 'lqQg6PlCWgI',
514 'upload_date': '20120724',
515 'uploader_id': 'olympic',
516 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
517 'uploader': 'Olympics',
518 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
521 'skip_download': 'requires avconv',
526 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
530 'stretched_ratio': 16 / 9.,
531 'upload_date': '20110310',
532 'uploader_id': 'AllenMeow',
533 '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',
535 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
538 # url_encoded_fmt_stream_map is empty string
540 'url': 'qEJwOuvDf7I',
544 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
546 'upload_date': '20150404',
547 'uploader_id': 'spbelect',
548 'uploader': 'Наблюдатели Петербурга',
551 'skip_download': 'requires avconv',
554 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
556 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
560 'title': 'md5:7b81415841e02ecd4313668cde88737a',
561 'description': 'md5:116377fd2963b81ec4ce64b542173306',
562 'upload_date': '20150625',
563 'uploader_id': 'dorappi2000',
564 'uploader': 'dorappi2000',
565 'formats': 'mincount:33',
568 # DASH manifest with segment_list
570 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
571 'md5': '8ce563a1d667b599d21064e982ab9e31',
575 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
576 'uploader': 'Airtek',
577 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
578 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
579 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
582 'youtube_include_dash_manifest': True,
583 'format': '135', # bestvideo
587 # Multifeed videos (multiple cameras), URL is for Main Camera
588 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
591 'title': 'teamPGP: Rocket League Noob Stream',
592 'description': 'md5:dc7872fb300e143831327f1bae3af010',
598 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
599 'description': 'md5:dc7872fb300e143831327f1bae3af010',
600 'upload_date': '20150721',
601 'uploader': 'Beer Games Beer',
602 'uploader_id': 'beergamesbeer',
608 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
609 'description': 'md5:dc7872fb300e143831327f1bae3af010',
610 'upload_date': '20150721',
611 'uploader': 'Beer Games Beer',
612 'uploader_id': 'beergamesbeer',
618 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
619 'description': 'md5:dc7872fb300e143831327f1bae3af010',
620 'upload_date': '20150721',
621 'uploader': 'Beer Games Beer',
622 'uploader_id': 'beergamesbeer',
628 'title': 'teamPGP: Rocket League Noob Stream (zim)',
629 'description': 'md5:dc7872fb300e143831327f1bae3af010',
630 'upload_date': '20150721',
631 'uploader': 'Beer Games Beer',
632 'uploader_id': 'beergamesbeer',
636 'skip_download': True,
641 def __init__(self, *args, **kwargs):
642 super(YoutubeIE, self).__init__(*args, **kwargs)
643 self._player_cache = {}
645 def report_video_info_webpage_download(self, video_id):
646 """Report attempt to download video info webpage."""
647 self.to_screen('%s: Downloading video info webpage' % video_id)
649 def report_information_extraction(self, video_id):
650 """Report attempt to extract video information."""
651 self.to_screen('%s: Extracting video information' % video_id)
653 def report_unavailable_format(self, video_id, format):
654 """Report extracted video URL."""
655 self.to_screen('%s: Format %s not available' % (video_id, format))
657 def report_rtmp_download(self):
658 """Indicate the download will use the RTMP protocol."""
659 self.to_screen('RTMP download detected')
661 def _signature_cache_id(self, example_sig):
662 """ Return a string representation of a signature """
663 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
665 def _extract_signature_function(self, video_id, player_url, example_sig):
667 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
670 raise ExtractorError('Cannot identify player %r' % player_url)
671 player_type = id_m.group('ext')
672 player_id = id_m.group('id')
674 # Read from filesystem cache
675 func_id = '%s_%s_%s' % (
676 player_type, player_id, self._signature_cache_id(example_sig))
677 assert os.path.basename(func_id) == func_id
679 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
680 if cache_spec is not None:
681 return lambda s: ''.join(s[i] for i in cache_spec)
684 'Downloading player %s' % player_url
685 if self._downloader.params.get('verbose') else
686 'Downloading %s player %s' % (player_type, player_id)
688 if player_type == 'js':
689 code = self._download_webpage(
690 player_url, video_id,
692 errnote='Download of %s failed' % player_url)
693 res = self._parse_sig_js(code)
694 elif player_type == 'swf':
695 urlh = self._request_webpage(
696 player_url, video_id,
698 errnote='Download of %s failed' % player_url)
700 res = self._parse_sig_swf(code)
702 assert False, 'Invalid player type %r' % player_type
704 test_string = ''.join(map(compat_chr, range(len(example_sig))))
705 cache_res = res(test_string)
706 cache_spec = [ord(c) for c in cache_res]
708 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
711 def _print_sig_code(self, func, example_sig):
712 def gen_sig_code(idxs):
713 def _genslice(start, end, step):
714 starts = '' if start == 0 else str(start)
715 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
716 steps = '' if step == 1 else (':%d' % step)
717 return 's[%s%s%s]' % (starts, ends, steps)
720 # Quelch pyflakes warnings - start will be set when step is set
721 start = '(Never used)'
722 for i, prev in zip(idxs[1:], idxs[:-1]):
726 yield _genslice(start, prev, step)
729 if i - prev in [-1, 1]:
738 yield _genslice(start, i, step)
740 test_string = ''.join(map(compat_chr, range(len(example_sig))))
741 cache_res = func(test_string)
742 cache_spec = [ord(c) for c in cache_res]
743 expr_code = ' + '.join(gen_sig_code(cache_spec))
744 signature_id_tuple = '(%s)' % (
745 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
746 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
747 ' return %s\n') % (signature_id_tuple, expr_code)
748 self.to_screen('Extracted signature function:\n' + code)
750 def _parse_sig_js(self, jscode):
751 funcname = self._search_regex(
752 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
753 'Initial JS player signature function name')
755 jsi = JSInterpreter(jscode)
756 initial_function = jsi.extract_function(funcname)
757 return lambda s: initial_function([s])
759 def _parse_sig_swf(self, file_contents):
760 swfi = SWFInterpreter(file_contents)
761 TARGET_CLASSNAME = 'SignatureDecipher'
762 searched_class = swfi.extract_class(TARGET_CLASSNAME)
763 initial_function = swfi.extract_function(searched_class, 'decipher')
764 return lambda s: initial_function([s])
766 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
767 """Turn the encrypted s field into a working signature"""
769 if player_url is None:
770 raise ExtractorError('Cannot decrypt signature without player_url')
772 if player_url.startswith('//'):
773 player_url = 'https:' + player_url
775 player_id = (player_url, self._signature_cache_id(s))
776 if player_id not in self._player_cache:
777 func = self._extract_signature_function(
778 video_id, player_url, s
780 self._player_cache[player_id] = func
781 func = self._player_cache[player_id]
782 if self._downloader.params.get('youtube_print_sig_code'):
783 self._print_sig_code(func, s)
785 except Exception as e:
786 tb = traceback.format_exc()
787 raise ExtractorError(
788 'Signature extraction failed: ' + tb, cause=e)
790 def _get_subtitles(self, video_id, webpage):
792 subs_doc = self._download_xml(
793 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
794 video_id, note=False)
795 except ExtractorError as err:
796 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
800 for track in subs_doc.findall('track'):
801 lang = track.attrib['lang_code']
802 if lang in sub_lang_list:
805 for ext in ['sbv', 'vtt', 'srt']:
806 params = compat_urllib_parse.urlencode({
810 'name': track.attrib['name'].encode('utf-8'),
813 'url': 'https://www.youtube.com/api/timedtext?' + params,
816 sub_lang_list[lang] = sub_formats
817 if not sub_lang_list:
818 self._downloader.report_warning('video doesn\'t have subtitles')
822 def _get_automatic_captions(self, video_id, webpage):
823 """We need the webpage for getting the captions url, pass it as an
824 argument to speed up the process."""
825 self.to_screen('%s: Looking for automatic captions' % video_id)
826 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
827 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
829 self._downloader.report_warning(err_msg)
831 player_config = json.loads(mobj.group(1))
833 args = player_config['args']
834 caption_url = args['ttsurl']
835 timestamp = args['timestamp']
836 # We get the available subtitles
837 list_params = compat_urllib_parse.urlencode({
842 list_url = caption_url + '&' + list_params
843 caption_list = self._download_xml(list_url, video_id)
844 original_lang_node = caption_list.find('track')
845 if original_lang_node is None:
846 self._downloader.report_warning('Video doesn\'t have automatic captions')
848 original_lang = original_lang_node.attrib['lang_code']
849 caption_kind = original_lang_node.attrib.get('kind', '')
852 for lang_node in caption_list.findall('target'):
853 sub_lang = lang_node.attrib['lang_code']
855 for ext in ['sbv', 'vtt', 'srt']:
856 params = compat_urllib_parse.urlencode({
857 'lang': original_lang,
861 'kind': caption_kind,
864 'url': caption_url + '&' + params,
867 sub_lang_list[sub_lang] = sub_formats
869 # An extractor error can be raise by the download process if there are
870 # no automatic captions but there are subtitles
871 except (KeyError, ExtractorError):
872 self._downloader.report_warning(err_msg)
876 def extract_id(cls, url):
877 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
879 raise ExtractorError('Invalid URL: %s' % url)
880 video_id = mobj.group(2)
883 def _extract_from_m3u8(self, manifest_url, video_id):
886 def _get_urls(_manifest):
887 lines = _manifest.split('\n')
888 urls = filter(lambda l: l and not l.startswith('#'),
891 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
892 formats_urls = _get_urls(manifest)
893 for format_url in formats_urls:
894 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
895 url_map[itag] = format_url
898 def _extract_annotations(self, video_id):
899 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
900 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
902 def _parse_dash_manifest(
903 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
904 def decrypt_sig(mobj):
906 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
907 return '/signature/%s' % dec_s
908 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
909 dash_doc = self._download_xml(
910 dash_manifest_url, video_id,
911 note='Downloading DASH manifest',
912 errnote='Could not download DASH manifest',
915 if dash_doc is False:
919 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
920 mime_type = a.attrib.get('mimeType')
921 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
922 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
925 if mime_type == 'text/vtt':
926 # TODO implement WebVTT downloading
928 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
929 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
930 format_id = r.attrib['id']
931 video_url = url_el.text
932 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
934 'format_id': format_id,
936 'width': int_or_none(r.attrib.get('width')),
937 'height': int_or_none(r.attrib.get('height')),
938 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
939 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
940 'filesize': filesize,
941 'fps': int_or_none(r.attrib.get('frameRate')),
943 if segment_list is not None:
945 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
946 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
947 'protocol': 'http_dash_segments',
950 existing_format = next(
952 if fo['format_id'] == format_id)
953 except StopIteration:
954 full_info = self._formats.get(format_id, {}).copy()
956 codecs = r.attrib.get('codecs')
958 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
959 full_info['vcodec'] = codecs
960 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
961 full_info['acodec'] = codecs
962 formats.append(full_info)
964 existing_format.update(f)
966 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
969 def _real_extract(self, url):
970 url, smuggled_data = unsmuggle_url(url, {})
973 'http' if self._downloader.params.get('prefer_insecure', False)
978 parsed_url = compat_urllib_parse_urlparse(url)
979 for component in [parsed_url.fragment, parsed_url.query]:
980 query = compat_parse_qs(component)
981 if start_time is None and 't' in query:
982 start_time = parse_duration(query['t'][0])
983 if start_time is None and 'start' in query:
984 start_time = parse_duration(query['start'][0])
985 if end_time is None and 'end' in query:
986 end_time = parse_duration(query['end'][0])
988 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
989 mobj = re.search(self._NEXT_URL_RE, url)
991 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
992 video_id = self.extract_id(url)
995 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
996 video_webpage = self._download_webpage(url, video_id)
998 # Attempt to extract SWF player URL
999 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1000 if mobj is not None:
1001 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1007 def add_dash_mpd(video_info):
1008 dash_mpd = video_info.get('dashmpd')
1009 if dash_mpd and dash_mpd[0] not in dash_mpds:
1010 dash_mpds.append(dash_mpd[0])
1013 embed_webpage = None
1015 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1017 # We simulate the access to the video from www.youtube.com/v/{video_id}
1018 # this can be viewed without login into Youtube
1019 url = proto + '://www.youtube.com/embed/%s' % video_id
1020 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1021 data = compat_urllib_parse.urlencode({
1022 'video_id': video_id,
1023 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1024 'sts': self._search_regex(
1025 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1027 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1028 video_info_webpage = self._download_webpage(
1029 video_info_url, video_id,
1030 note='Refetching age-gated info webpage',
1031 errnote='unable to download video info webpage')
1032 video_info = compat_parse_qs(video_info_webpage)
1033 add_dash_mpd(video_info)
1037 # Try looking directly into the video webpage
1038 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
1040 json_code = uppercase_escape(mobj.group(1))
1041 ytplayer_config = json.loads(json_code)
1042 args = ytplayer_config['args']
1043 if args.get('url_encoded_fmt_stream_map'):
1044 # Convert to the same format returned by compat_parse_qs
1045 video_info = dict((k, [v]) for k, v in args.items())
1046 add_dash_mpd(video_info)
1047 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1049 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1050 # We also try looking in get_video_info since it may contain different dashmpd
1051 # URL that points to a DASH manifest with possibly different itag set (some itags
1052 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1053 # manifest pointed by get_video_info's dashmpd).
1054 # The general idea is to take a union of itags of both DASH manifests (for example
1055 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1056 self.report_video_info_webpage_download(video_id)
1057 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1059 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1060 % (proto, video_id, el_type))
1061 video_info_webpage = self._download_webpage(
1063 video_id, note=False,
1064 errnote='unable to download video info webpage')
1065 get_video_info = compat_parse_qs(video_info_webpage)
1066 if get_video_info.get('use_cipher_signature') != ['True']:
1067 add_dash_mpd(get_video_info)
1069 video_info = get_video_info
1070 if 'token' in get_video_info:
1072 if 'token' not in video_info:
1073 if 'reason' in video_info:
1074 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1075 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1077 raise ExtractorError('YouTube said: This video is available in %s only' % (
1078 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1080 raise ExtractorError(
1081 'YouTube said: %s' % video_info['reason'][0],
1082 expected=True, video_id=video_id)
1084 raise ExtractorError(
1085 '"token" parameter not in video info for unknown reason',
1089 if 'title' in video_info:
1090 video_title = video_info['title'][0]
1092 self._downloader.report_warning('Unable to extract video title')
1096 video_description = get_element_by_id("eow-description", video_webpage)
1097 if video_description:
1098 video_description = re.sub(r'''(?x)
1100 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1102 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1103 class="yt-uix-redirect-link"\s*>
1106 ''', r'\1', video_description)
1107 video_description = clean_html(video_description)
1109 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1111 video_description = unescapeHTML(fd_mobj.group(1))
1113 video_description = ''
1115 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1116 if not self._downloader.params.get('noplaylist'):
1119 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1120 for feed in multifeed_metadata_list.split(','):
1121 feed_data = compat_parse_qs(feed)
1123 '_type': 'url_transparent',
1124 'ie_key': 'Youtube',
1126 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1127 {'force_singlefeed': True}),
1128 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1130 feed_ids.append(feed_data['id'][0])
1132 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1133 % (', '.join(feed_ids), video_id))
1134 return self.playlist_result(entries, video_id, video_title, video_description)
1135 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1137 if 'view_count' in video_info:
1138 view_count = int(video_info['view_count'][0])
1142 # Check for "rental" videos
1143 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1144 raise ExtractorError('"rental" videos not supported')
1146 # Start extracting information
1147 self.report_information_extraction(video_id)
1150 if 'author' not in video_info:
1151 raise ExtractorError('Unable to extract uploader name')
1152 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1155 video_uploader_id = None
1156 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1157 if mobj is not None:
1158 video_uploader_id = mobj.group(1)
1160 self._downloader.report_warning('unable to extract uploader nickname')
1163 # We try first to get a high quality image:
1164 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1165 video_webpage, re.DOTALL)
1166 if m_thumb is not None:
1167 video_thumbnail = m_thumb.group(1)
1168 elif 'thumbnail_url' not in video_info:
1169 self._downloader.report_warning('unable to extract video thumbnail')
1170 video_thumbnail = None
1171 else: # don't panic if we can't find it
1172 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1175 upload_date = self._html_search_meta(
1176 'datePublished', video_webpage, 'upload date', default=None)
1178 upload_date = self._search_regex(
1179 [r'(?s)id="eow-date.*?>(.*?)</span>',
1180 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1181 video_webpage, 'upload date', default=None)
1183 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1184 upload_date = unified_strdate(upload_date)
1186 m_cat_container = self._search_regex(
1187 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1188 video_webpage, 'categories', default=None)
1190 category = self._html_search_regex(
1191 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1193 video_categories = None if category is None else [category]
1195 video_categories = None
1198 unescapeHTML(m.group('content'))
1199 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1201 def _extract_count(count_name):
1202 return str_to_int(self._search_regex(
1203 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1204 % re.escape(count_name),
1205 video_webpage, count_name, default=None))
1207 like_count = _extract_count('like')
1208 dislike_count = _extract_count('dislike')
1211 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1212 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1214 if 'length_seconds' not in video_info:
1215 self._downloader.report_warning('unable to extract video duration')
1216 video_duration = None
1218 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1221 video_annotations = None
1222 if self._downloader.params.get('writeannotations', False):
1223 video_annotations = self._extract_annotations(video_id)
1225 def _map_to_format_list(urlmap):
1227 for itag, video_real_url in urlmap.items():
1230 'url': video_real_url,
1231 'player_url': player_url,
1233 if itag in self._formats:
1234 dct.update(self._formats[itag])
1238 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1239 self.report_rtmp_download()
1241 'format_id': '_rtmp',
1243 'url': video_info['conn'][0],
1244 'player_url': player_url,
1246 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1247 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1248 if 'rtmpe%3Dyes' in encoded_url_map:
1249 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1251 for url_data_str in encoded_url_map.split(','):
1252 url_data = compat_parse_qs(url_data_str)
1253 if 'itag' not in url_data or 'url' not in url_data:
1255 format_id = url_data['itag'][0]
1256 url = url_data['url'][0]
1258 if 'sig' in url_data:
1259 url += '&signature=' + url_data['sig'][0]
1260 elif 's' in url_data:
1261 encrypted_sig = url_data['s'][0]
1262 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1264 jsplayer_url_json = self._search_regex(
1266 embed_webpage if age_gate else video_webpage,
1267 'JS player URL (1)', default=None)
1268 if not jsplayer_url_json and not age_gate:
1269 # We need the embed website after all
1270 if embed_webpage is None:
1271 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1272 embed_webpage = self._download_webpage(
1273 embed_url, video_id, 'Downloading embed webpage')
1274 jsplayer_url_json = self._search_regex(
1275 ASSETS_RE, embed_webpage, 'JS player URL')
1277 player_url = json.loads(jsplayer_url_json)
1278 if player_url is None:
1279 player_url_json = self._search_regex(
1280 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1281 video_webpage, 'age gate player URL')
1282 player_url = json.loads(player_url_json)
1284 if self._downloader.params.get('verbose'):
1285 if player_url is None:
1286 player_version = 'unknown'
1287 player_desc = 'unknown'
1289 if player_url.endswith('swf'):
1290 player_version = self._search_regex(
1291 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1292 'flash player', fatal=False)
1293 player_desc = 'flash player %s' % player_version
1295 player_version = self._search_regex(
1296 r'html5player-([^/]+?)(?:/html5player)?\.js',
1298 'html5 player', fatal=False)
1299 player_desc = 'html5 player %s' % player_version
1301 parts_sizes = self._signature_cache_id(encrypted_sig)
1302 self.to_screen('{%s} signature length %s, %s' %
1303 (format_id, parts_sizes, player_desc))
1305 signature = self._decrypt_signature(
1306 encrypted_sig, video_id, player_url, age_gate)
1307 url += '&signature=' + signature
1308 if 'ratebypass' not in url:
1309 url += '&ratebypass=yes'
1310 url_map[format_id] = url
1311 formats = _map_to_format_list(url_map)
1312 elif video_info.get('hlsvp'):
1313 manifest_url = video_info['hlsvp'][0]
1314 url_map = self._extract_from_m3u8(manifest_url, video_id)
1315 formats = _map_to_format_list(url_map)
1317 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1319 # Look for the DASH manifest
1320 if self._downloader.params.get('youtube_include_dash_manifest', True):
1321 dash_mpd_fatal = True
1322 for dash_manifest_url in dash_mpds:
1325 for df in self._parse_dash_manifest(
1326 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1327 # Do not overwrite DASH format found in some previous DASH manifest
1328 if df['format_id'] not in dash_formats:
1329 dash_formats[df['format_id']] = df
1330 # Additional DASH manifests may end up in HTTP Error 403 therefore
1331 # allow them to fail without bug report message if we already have
1332 # some DASH manifest succeeded. This is temporary workaround to reduce
1333 # burst of bug reports until we figure out the reason and whether it
1334 # can be fixed at all.
1335 dash_mpd_fatal = False
1336 except (ExtractorError, KeyError) as e:
1337 self.report_warning(
1338 'Skipping DASH manifest: %r' % e, video_id)
1340 # Remove the formats we found through non-DASH, they
1341 # contain less info and it can be wrong, because we use
1342 # fixed values (for example the resolution). See
1343 # https://github.com/rg3/youtube-dl/issues/5774 for an
1345 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1346 formats.extend(dash_formats.values())
1348 # Check for malformed aspect ratio
1349 stretched_m = re.search(
1350 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1353 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1355 if f.get('vcodec') != 'none':
1356 f['stretched_ratio'] = ratio
1358 self._sort_formats(formats)
1362 'uploader': video_uploader,
1363 'uploader_id': video_uploader_id,
1364 'upload_date': upload_date,
1365 'title': video_title,
1366 'thumbnail': video_thumbnail,
1367 'description': video_description,
1368 'categories': video_categories,
1370 'subtitles': video_subtitles,
1371 'automatic_captions': automatic_captions,
1372 'duration': video_duration,
1373 'age_limit': 18 if age_gate else 0,
1374 'annotations': video_annotations,
1375 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1376 'view_count': view_count,
1377 'like_count': like_count,
1378 'dislike_count': dislike_count,
1379 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1382 'start_time': start_time,
1383 'end_time': end_time,
1387 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1388 IE_DESC = 'YouTube.com playlists'
1389 _VALID_URL = r"""(?x)(?:
1394 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1395 \? (?:.*?&)*? (?:p|a|list)=
1399 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1400 # Top tracks, they can also include dots
1405 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1407 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1408 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1409 IE_NAME = 'youtube:playlist'
1411 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1413 'title': 'ytdl test PL',
1414 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1416 'playlist_count': 3,
1418 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1420 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1421 'title': 'YDL_Empty_List',
1423 'playlist_count': 0,
1425 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1426 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1428 'title': '29C3: Not my department',
1429 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1431 'playlist_count': 95,
1433 'note': 'issue #673',
1434 'url': 'PLBB231211A4F62143',
1436 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1437 'id': 'PLBB231211A4F62143',
1439 'playlist_mincount': 26,
1441 'note': 'Large playlist',
1442 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1444 'title': 'Uploads from Cauchemar',
1445 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1447 'playlist_mincount': 799,
1449 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1451 'title': 'YDL_safe_search',
1452 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1454 'playlist_count': 2,
1457 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1458 'playlist_count': 4,
1461 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1464 'note': 'Embedded SWF player',
1465 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1466 'playlist_count': 4,
1469 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1472 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1473 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1475 'title': 'Uploads from Interstellar Movie',
1476 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1478 'playlist_mincout': 21,
1481 def _real_initialize(self):
1484 def _extract_mix(self, playlist_id):
1485 # The mixes are generated from a single video
1486 # the id of the playlist is just 'RD' + video_id
1487 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1488 webpage = self._download_webpage(
1489 url, playlist_id, 'Downloading Youtube mix')
1490 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1492 search_title('playlist-title') or
1493 search_title('title long-title') or
1494 search_title('title'))
1495 title = clean_html(title_span)
1496 ids = orderedSet(re.findall(
1497 r'''(?xs)data-video-username=".*?".*?
1498 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1500 url_results = self._ids_to_results(ids)
1502 return self.playlist_result(url_results, playlist_id, title)
1504 def _extract_playlist(self, playlist_id):
1505 url = self._TEMPLATE_URL % playlist_id
1506 page = self._download_webpage(url, playlist_id)
1508 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1509 match = match.strip()
1510 # Check if the playlist exists or is private
1511 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1512 raise ExtractorError(
1513 'The playlist doesn\'t exist or is private, use --username or '
1514 '--netrc to access it.',
1516 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1517 raise ExtractorError(
1518 'Invalid parameters. Maybe URL is incorrect.',
1520 elif re.match(r'[^<]*Choose your language[^<]*', match):
1523 self.report_warning('Youtube gives an alert message: ' + match)
1525 # Extract the video ids from the playlist pages
1527 more_widget_html = content_html = page
1528 for page_num in itertools.count(1):
1529 matches = re.finditer(self._VIDEO_RE, content_html)
1530 # We remove the duplicates and the link with index 0
1531 # (it's not the first video of the playlist)
1532 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1533 for vid_id in new_ids:
1534 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1536 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1540 more = self._download_json(
1541 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1542 'Downloading page #%s' % page_num,
1543 transform_source=uppercase_escape)
1544 content_html = more['content_html']
1545 if not content_html.strip():
1546 # Some webpages show a "Load more" button but they don't
1549 more_widget_html = more['load_more_widget_html']
1551 playlist_title = self._html_search_regex(
1552 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1555 return self.playlist_result(_entries(), playlist_id, playlist_title)
1557 def _real_extract(self, url):
1558 # Extract playlist id
1559 mobj = re.match(self._VALID_URL, url)
1561 raise ExtractorError('Invalid URL: %s' % url)
1562 playlist_id = mobj.group(1) or mobj.group(2)
1564 # Check if it's a video-specific URL
1565 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1566 if 'v' in query_dict:
1567 video_id = query_dict['v'][0]
1568 if self._downloader.params.get('noplaylist'):
1569 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1570 return self.url_result(video_id, 'Youtube', video_id=video_id)
1572 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1574 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1575 # Mixes require a custom extraction process
1576 return self._extract_mix(playlist_id)
1578 return self._extract_playlist(playlist_id)
1581 class YoutubeChannelIE(InfoExtractor):
1582 IE_DESC = 'YouTube.com channels'
1583 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1584 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1585 IE_NAME = 'youtube:channel'
1587 'note': 'paginated channel',
1588 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1589 'playlist_mincount': 91,
1591 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1596 def extract_videos_from_page(page):
1599 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1600 video_id = mobj.group('id')
1601 video_title = unescapeHTML(mobj.group('title'))
1603 idx = ids_in_page.index(video_id)
1604 if video_title and not titles_in_page[idx]:
1605 titles_in_page[idx] = video_title
1607 ids_in_page.append(video_id)
1608 titles_in_page.append(video_title)
1609 return zip(ids_in_page, titles_in_page)
1611 def _real_extract(self, url):
1612 channel_id = self._match_id(url)
1614 url = self._TEMPLATE_URL % channel_id
1616 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1617 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1618 # otherwise fallback on channel by page extraction
1619 channel_page = self._download_webpage(
1620 url + '?view=57', channel_id,
1621 'Downloading channel page', fatal=False)
1622 channel_playlist_id = self._html_search_meta(
1623 'channelId', channel_page, 'channel id', default=None)
1624 if not channel_playlist_id:
1625 channel_playlist_id = self._search_regex(
1626 r'data-channel-external-id="([^"]+)"',
1627 channel_page, 'channel id', default=None)
1628 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1629 playlist_id = 'UU' + channel_playlist_id[2:]
1630 return self.url_result(
1631 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1633 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1634 autogenerated = re.search(r'''(?x)
1636 channel-header-autogenerated-label|
1637 yt-channel-title-autogenerated
1638 )[^"]*"''', channel_page) is not None
1641 # The videos are contained in a single page
1642 # the ajax pages can't be used, they are empty
1645 video_id, 'Youtube', video_id=video_id,
1646 video_title=video_title)
1647 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1648 return self.playlist_result(entries, channel_id)
1651 more_widget_html = content_html = channel_page
1652 for pagenum in itertools.count(1):
1654 for video_id, video_title in self.extract_videos_from_page(content_html):
1655 yield self.url_result(
1656 video_id, 'Youtube', video_id=video_id,
1657 video_title=video_title)
1660 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1665 more = self._download_json(
1666 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1667 'Downloading page #%s' % (pagenum + 1),
1668 transform_source=uppercase_escape)
1669 content_html = more['content_html']
1670 more_widget_html = more['load_more_widget_html']
1672 return self.playlist_result(_entries(), channel_id)
1675 class YoutubeUserIE(YoutubeChannelIE):
1676 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1677 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1678 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1679 IE_NAME = 'youtube:user'
1682 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1683 'playlist_mincount': 320,
1685 'title': 'TheLinuxFoundation',
1688 'url': 'ytuser:phihag',
1689 'only_matching': True,
1693 def suitable(cls, url):
1694 # Don't return True if the url can be extracted with other youtube
1695 # extractor, the regex would is too permissive and it would match.
1696 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1697 if any(ie.suitable(url) for ie in other_ies):
1700 return super(YoutubeUserIE, cls).suitable(url)
1703 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1704 IE_DESC = 'YouTube.com searches'
1705 # there doesn't appear to be a real limit, for example if you search for
1706 # 'python' you get more than 8.000.000 results
1707 _MAX_RESULTS = float('inf')
1708 IE_NAME = 'youtube:search'
1709 _SEARCH_KEY = 'ytsearch'
1710 _EXTRA_QUERY_ARGS = {}
1713 def _get_n_results(self, query, n):
1714 """Get a specified number of results for a query"""
1719 for pagenum in itertools.count(1):
1721 'search_query': query.encode('utf-8'),
1725 url_query.update(self._EXTRA_QUERY_ARGS)
1726 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1727 data = self._download_json(
1728 result_url, video_id='query "%s"' % query,
1729 note='Downloading page %s' % pagenum,
1730 errnote='Unable to download API page')
1731 html_content = data[1]['body']['content']
1733 if 'class="search-message' in html_content:
1734 raise ExtractorError(
1735 '[youtube] No video results', expected=True)
1737 new_videos = self._ids_to_results(orderedSet(re.findall(
1738 r'href="/watch\?v=(.{11})', html_content)))
1739 videos += new_videos
1740 if not new_videos or len(videos) > limit:
1745 return self.playlist_result(videos, query)
1748 class YoutubeSearchDateIE(YoutubeSearchIE):
1749 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1750 _SEARCH_KEY = 'ytsearchdate'
1751 IE_DESC = 'YouTube.com searches, newest videos first'
1752 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1755 class YoutubeSearchURLIE(InfoExtractor):
1756 IE_DESC = 'YouTube.com search URLs'
1757 IE_NAME = 'youtube:search_url'
1758 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1760 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1761 'playlist_mincount': 5,
1763 'title': 'youtube-dl test video',
1767 def _real_extract(self, url):
1768 mobj = re.match(self._VALID_URL, url)
1769 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1771 webpage = self._download_webpage(url, query)
1772 result_code = self._search_regex(
1773 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1775 part_codes = re.findall(
1776 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1778 for part_code in part_codes:
1779 part_title = self._html_search_regex(
1780 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1781 part_url_snippet = self._html_search_regex(
1782 r'(?s)href="([^"]+)"', part_code, 'item URL')
1783 part_url = compat_urlparse.urljoin(
1784 'https://www.youtube.com/', part_url_snippet)
1788 'title': part_title,
1792 '_type': 'playlist',
1798 class YoutubeShowIE(InfoExtractor):
1799 IE_DESC = 'YouTube.com (multi-season) shows'
1800 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1801 IE_NAME = 'youtube:show'
1803 'url': 'http://www.youtube.com/show/airdisasters',
1804 'playlist_mincount': 3,
1806 'id': 'airdisasters',
1807 'title': 'Air Disasters',
1811 def _real_extract(self, url):
1812 mobj = re.match(self._VALID_URL, url)
1813 playlist_id = mobj.group('id')
1814 webpage = self._download_webpage(
1815 url, playlist_id, 'Downloading show webpage')
1816 # There's one playlist for each season of the show
1817 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1818 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1821 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1822 for season in m_seasons
1824 title = self._og_search_title(webpage, fatal=False)
1827 '_type': 'playlist',
1834 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1836 Base class for feed extractors
1837 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1839 _LOGIN_REQUIRED = True
1843 return 'youtube:%s' % self._FEED_NAME
1845 def _real_initialize(self):
1848 def _real_extract(self, url):
1849 page = self._download_webpage(
1850 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1852 # The extraction process is the same as for playlists, but the regex
1853 # for the video ids doesn't contain an index
1855 more_widget_html = content_html = page
1856 for page_num in itertools.count(1):
1857 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1859 # 'recommended' feed has infinite 'load more' and each new portion spins
1860 # the same videos in (sometimes) slightly different order, so we'll check
1861 # for unicity and break when portion has no new videos
1862 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1868 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1872 more = self._download_json(
1873 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1874 'Downloading page #%s' % page_num,
1875 transform_source=uppercase_escape)
1876 content_html = more['content_html']
1877 more_widget_html = more['load_more_widget_html']
1879 return self.playlist_result(
1880 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1883 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1884 IE_NAME = 'youtube:watchlater'
1885 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1886 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1888 _TESTS = [] # override PlaylistIE tests
1890 def _real_extract(self, url):
1891 return self._extract_playlist('WL')
1894 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1895 IE_NAME = 'youtube:favorites'
1896 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1897 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1898 _LOGIN_REQUIRED = True
1900 def _real_extract(self, url):
1901 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1902 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1903 return self.url_result(playlist_id, 'YoutubePlaylist')
1906 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1907 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1908 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1909 _FEED_NAME = 'recommended'
1910 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1913 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1914 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1915 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1916 _FEED_NAME = 'subscriptions'
1917 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1920 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1921 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1922 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1923 _FEED_NAME = 'history'
1924 _PLAYLIST_TITLE = 'Youtube History'
1927 class YoutubeTruncatedURLIE(InfoExtractor):
1928 IE_NAME = 'youtube:truncated_url'
1929 IE_DESC = False # Do not list
1930 _VALID_URL = r'''(?x)
1932 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1935 annotation_id=annotation_[^&]+|
1940 attribution_link\?a=[^&]+
1946 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1947 'only_matching': True,
1949 'url': 'http://www.youtube.com/watch?',
1950 'only_matching': True,
1952 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1953 'only_matching': True,
1955 'url': 'https://www.youtube.com/watch?feature=foo',
1956 'only_matching': True,
1958 'url': 'https://www.youtube.com/watch?hl=en-GB',
1959 'only_matching': True,
1962 def _real_extract(self, url):
1963 raise ExtractorError(
1964 'Did you forget to quote the URL? Remember that & is a meta '
1965 'character in most shells, so you want to put the URL in quotes, '
1967 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1968 ' or simply youtube-dl BaW_jenozKc .',
1972 class YoutubeTruncatedIDIE(InfoExtractor):
1973 IE_NAME = 'youtube:truncated_id'
1974 IE_DESC = False # Do not list
1975 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1978 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1979 'only_matching': True,
1982 def _real_extract(self, url):
1983 video_id = self._match_id(url)
1984 raise ExtractorError(
1985 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),