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/SecondFactor'
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="gaia_secondfactorform"', 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 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
141 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
143 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
144 secTok = match.group(1)
145 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
147 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
148 timeStmp = match.group(1)
151 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
153 'smsUserPin': tfa_code,
154 'smsVerifyPin': 'Verify',
156 'PersistentCookie': 'yes',
157 'checkConnection': '',
158 'checkedDomains': 'youtube',
161 'timeStmp': timeStmp,
162 'service': 'youtube',
165 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
166 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
168 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
169 tfa_results = self._download_webpage(
171 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
173 if tfa_results is False:
176 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
177 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
179 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
180 self._downloader.report_warning('unable to log in - did the page structure change?')
182 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
183 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
186 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
187 self._downloader.report_warning('unable to log in: bad username or password')
191 def _real_initialize(self):
192 if self._downloader is None:
195 if not self._login():
199 class YoutubeIE(YoutubeBaseInfoExtractor):
200 IE_DESC = 'YouTube.com'
201 _VALID_URL = r"""(?x)^
203 (?:https?://|//) # http(s):// or protocol-independent URL
204 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
205 (?:www\.)?deturl\.com/www\.youtube\.com/|
206 (?:www\.)?pwnyoutube\.com/|
207 (?:www\.)?yourepeat\.com/|
208 tube\.majestyc\.net/|
209 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
210 (?:.*?\#/)? # handle anchor (#/) redirect urls
211 (?: # the various things that can precede the ID:
212 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
213 |(?: # or the v= param in all its forms
214 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
215 (?:\?|\#!?) # the params delimiter ? or # or #!
216 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
220 |youtu\.be/ # just youtu.be/xxxx
221 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
223 )? # all until now is optional -> you can pass the naked ID
224 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
225 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
226 (?(1).+)? # if we found the ID, everything can follow
228 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
230 '5': {'ext': 'flv', 'width': 400, 'height': 240},
231 '6': {'ext': 'flv', 'width': 450, 'height': 270},
232 '13': {'ext': '3gp'},
233 '17': {'ext': '3gp', 'width': 176, 'height': 144},
234 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
235 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
236 '34': {'ext': 'flv', 'width': 640, 'height': 360},
237 '35': {'ext': 'flv', 'width': 854, 'height': 480},
238 '36': {'ext': '3gp', 'width': 320, 'height': 240},
239 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
240 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
241 '43': {'ext': 'webm', 'width': 640, 'height': 360},
242 '44': {'ext': 'webm', 'width': 854, 'height': 480},
243 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
244 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
245 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
246 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
250 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
251 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
252 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
253 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
254 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
255 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
256 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
258 # Apple HTTP Live Streaming
259 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
260 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
261 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
262 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
263 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
264 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
265 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
268 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
274 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
276 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
277 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
278 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
281 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
282 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
283 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
286 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
291 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
292 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
293 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
301 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
302 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
303 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
304 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
305 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
306 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
309 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
310 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
312 # Dash webm audio with opus inside
313 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
314 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
315 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
318 '_rtmp': {'protocol': 'rtmp'},
324 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
328 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
329 'uploader': 'Philipp Hagemeister',
330 'uploader_id': 'phihag',
331 'upload_date': '20121002',
332 '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 .',
333 'categories': ['Science & Technology'],
335 'dislike_count': int,
341 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
342 'note': 'Test generic use_cipher_signature video (#897)',
346 'upload_date': '20120506',
347 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
348 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
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',
367 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
368 'note': 'Embed-only video (#1746)',
372 'upload_date': '20120608',
373 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
374 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
375 'uploader': 'SET India',
376 'uploader_id': 'setindia'
380 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
381 'note': '256k DASH audio (format 141) via DASH manifest',
385 'upload_date': '20121002',
386 'uploader_id': '8KVIDEO',
388 'uploader': '8KVIDEO',
389 'title': 'UHDTV TEST 8K VIDEO.mp4'
392 'youtube_include_dash_manifest': True,
396 # DASH manifest with encrypted signature
398 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
402 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
403 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
404 'uploader': 'AfrojackVEVO',
405 'uploader_id': 'AfrojackVEVO',
406 'upload_date': '20131011',
409 'youtube_include_dash_manifest': True,
413 # JS player signature function name containing $
415 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
419 'title': 'Taylor Swift - Shake It Off',
420 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
421 'uploader': 'TaylorSwiftVEVO',
422 'uploader_id': 'TaylorSwiftVEVO',
423 'upload_date': '20140818',
426 'youtube_include_dash_manifest': True,
432 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
436 'upload_date': '20100909',
437 'uploader': 'The Amazing Atheist',
438 'uploader_id': 'TheAmazingAtheist',
439 'title': 'Burning Everyone\'s Koran',
440 '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',
443 # Normal age-gate video (No vevo, embed allowed)
445 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
449 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
450 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
451 'uploader': 'The Witcher',
452 'uploader_id': 'WitcherGame',
453 'upload_date': '20140605',
456 # Age-gate video with encrypted signature
458 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
462 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
463 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
464 'uploader': 'LloydVEVO',
465 'uploader_id': 'LloydVEVO',
466 'upload_date': '20110629',
469 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
471 'url': '__2ABJjxzNo',
475 'upload_date': '20100430',
476 'uploader_id': 'deadmau5',
477 'description': 'md5:12c56784b8032162bb936a5f76d55360',
478 'uploader': 'deadmau5',
479 'title': 'Deadmau5 - Some Chords (HD)',
481 'expected_warnings': [
482 'DASH manifest missing',
485 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
487 'url': 'lqQg6PlCWgI',
491 'upload_date': '20120731',
492 'uploader_id': 'olympic',
493 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
494 'uploader': 'Olympics',
495 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
498 'skip_download': 'requires avconv',
503 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
507 'stretched_ratio': 16 / 9.,
508 'upload_date': '20110310',
509 'uploader_id': 'AllenMeow',
510 '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',
512 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
515 # url_encoded_fmt_stream_map is empty string
517 'url': 'qEJwOuvDf7I',
521 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
523 'upload_date': '20150404',
524 'uploader_id': 'spbelect',
525 'uploader': 'Наблюдатели Петербурга',
528 'skip_download': 'requires avconv',
531 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
533 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
537 'title': 'md5:7b81415841e02ecd4313668cde88737a',
538 'description': 'md5:116377fd2963b81ec4ce64b542173306',
539 'upload_date': '20150625',
540 'uploader_id': 'dorappi2000',
541 'uploader': 'dorappi2000',
542 'formats': 'mincount:33',
545 # DASH manifest with segment_list
547 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
548 'md5': '8ce563a1d667b599d21064e982ab9e31',
552 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
553 'uploader': 'Airtek',
554 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
555 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
556 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
559 'youtube_include_dash_manifest': True,
560 'format': '135', # bestvideo
564 # Multifeed videos (multiple cameras), URL is for Main Camera
565 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
568 'title': 'teamPGP: Rocket League Noob Stream',
569 'description': 'md5:dc7872fb300e143831327f1bae3af010',
575 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
576 'description': 'md5:dc7872fb300e143831327f1bae3af010',
577 'upload_date': '20150721',
578 'uploader': 'Beer Games Beer',
579 'uploader_id': 'beergamesbeer',
585 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
586 'description': 'md5:dc7872fb300e143831327f1bae3af010',
587 'upload_date': '20150721',
588 'uploader': 'Beer Games Beer',
589 'uploader_id': 'beergamesbeer',
595 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
596 'description': 'md5:dc7872fb300e143831327f1bae3af010',
597 'upload_date': '20150721',
598 'uploader': 'Beer Games Beer',
599 'uploader_id': 'beergamesbeer',
605 'title': 'teamPGP: Rocket League Noob Stream (zim)',
606 'description': 'md5:dc7872fb300e143831327f1bae3af010',
607 'upload_date': '20150721',
608 'uploader': 'Beer Games Beer',
609 'uploader_id': 'beergamesbeer',
613 'skip_download': True,
618 def __init__(self, *args, **kwargs):
619 super(YoutubeIE, self).__init__(*args, **kwargs)
620 self._player_cache = {}
622 def report_video_info_webpage_download(self, video_id):
623 """Report attempt to download video info webpage."""
624 self.to_screen('%s: Downloading video info webpage' % video_id)
626 def report_information_extraction(self, video_id):
627 """Report attempt to extract video information."""
628 self.to_screen('%s: Extracting video information' % video_id)
630 def report_unavailable_format(self, video_id, format):
631 """Report extracted video URL."""
632 self.to_screen('%s: Format %s not available' % (video_id, format))
634 def report_rtmp_download(self):
635 """Indicate the download will use the RTMP protocol."""
636 self.to_screen('RTMP download detected')
638 def _signature_cache_id(self, example_sig):
639 """ Return a string representation of a signature """
640 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
642 def _extract_signature_function(self, video_id, player_url, example_sig):
644 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
647 raise ExtractorError('Cannot identify player %r' % player_url)
648 player_type = id_m.group('ext')
649 player_id = id_m.group('id')
651 # Read from filesystem cache
652 func_id = '%s_%s_%s' % (
653 player_type, player_id, self._signature_cache_id(example_sig))
654 assert os.path.basename(func_id) == func_id
656 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
657 if cache_spec is not None:
658 return lambda s: ''.join(s[i] for i in cache_spec)
661 'Downloading player %s' % player_url
662 if self._downloader.params.get('verbose') else
663 'Downloading %s player %s' % (player_type, player_id)
665 if player_type == 'js':
666 code = self._download_webpage(
667 player_url, video_id,
669 errnote='Download of %s failed' % player_url)
670 res = self._parse_sig_js(code)
671 elif player_type == 'swf':
672 urlh = self._request_webpage(
673 player_url, video_id,
675 errnote='Download of %s failed' % player_url)
677 res = self._parse_sig_swf(code)
679 assert False, 'Invalid player type %r' % player_type
681 test_string = ''.join(map(compat_chr, range(len(example_sig))))
682 cache_res = res(test_string)
683 cache_spec = [ord(c) for c in cache_res]
685 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
688 def _print_sig_code(self, func, example_sig):
689 def gen_sig_code(idxs):
690 def _genslice(start, end, step):
691 starts = '' if start == 0 else str(start)
692 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
693 steps = '' if step == 1 else (':%d' % step)
694 return 's[%s%s%s]' % (starts, ends, steps)
697 # Quelch pyflakes warnings - start will be set when step is set
698 start = '(Never used)'
699 for i, prev in zip(idxs[1:], idxs[:-1]):
703 yield _genslice(start, prev, step)
706 if i - prev in [-1, 1]:
715 yield _genslice(start, i, step)
717 test_string = ''.join(map(compat_chr, range(len(example_sig))))
718 cache_res = func(test_string)
719 cache_spec = [ord(c) for c in cache_res]
720 expr_code = ' + '.join(gen_sig_code(cache_spec))
721 signature_id_tuple = '(%s)' % (
722 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
723 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
724 ' return %s\n') % (signature_id_tuple, expr_code)
725 self.to_screen('Extracted signature function:\n' + code)
727 def _parse_sig_js(self, jscode):
728 funcname = self._search_regex(
729 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
730 'Initial JS player signature function name')
732 jsi = JSInterpreter(jscode)
733 initial_function = jsi.extract_function(funcname)
734 return lambda s: initial_function([s])
736 def _parse_sig_swf(self, file_contents):
737 swfi = SWFInterpreter(file_contents)
738 TARGET_CLASSNAME = 'SignatureDecipher'
739 searched_class = swfi.extract_class(TARGET_CLASSNAME)
740 initial_function = swfi.extract_function(searched_class, 'decipher')
741 return lambda s: initial_function([s])
743 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
744 """Turn the encrypted s field into a working signature"""
746 if player_url is None:
747 raise ExtractorError('Cannot decrypt signature without player_url')
749 if player_url.startswith('//'):
750 player_url = 'https:' + player_url
752 player_id = (player_url, self._signature_cache_id(s))
753 if player_id not in self._player_cache:
754 func = self._extract_signature_function(
755 video_id, player_url, s
757 self._player_cache[player_id] = func
758 func = self._player_cache[player_id]
759 if self._downloader.params.get('youtube_print_sig_code'):
760 self._print_sig_code(func, s)
762 except Exception as e:
763 tb = traceback.format_exc()
764 raise ExtractorError(
765 'Signature extraction failed: ' + tb, cause=e)
767 def _get_subtitles(self, video_id, webpage):
769 subs_doc = self._download_xml(
770 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
771 video_id, note=False)
772 except ExtractorError as err:
773 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
777 for track in subs_doc.findall('track'):
778 lang = track.attrib['lang_code']
779 if lang in sub_lang_list:
782 for ext in ['sbv', 'vtt', 'srt']:
783 params = compat_urllib_parse.urlencode({
787 'name': track.attrib['name'].encode('utf-8'),
790 'url': 'https://www.youtube.com/api/timedtext?' + params,
793 sub_lang_list[lang] = sub_formats
794 if not sub_lang_list:
795 self._downloader.report_warning('video doesn\'t have subtitles')
799 def _get_automatic_captions(self, video_id, webpage):
800 """We need the webpage for getting the captions url, pass it as an
801 argument to speed up the process."""
802 self.to_screen('%s: Looking for automatic captions' % video_id)
803 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
804 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
806 self._downloader.report_warning(err_msg)
808 player_config = json.loads(mobj.group(1))
810 args = player_config['args']
811 caption_url = args['ttsurl']
812 timestamp = args['timestamp']
813 # We get the available subtitles
814 list_params = compat_urllib_parse.urlencode({
819 list_url = caption_url + '&' + list_params
820 caption_list = self._download_xml(list_url, video_id)
821 original_lang_node = caption_list.find('track')
822 if original_lang_node is None:
823 self._downloader.report_warning('Video doesn\'t have automatic captions')
825 original_lang = original_lang_node.attrib['lang_code']
826 caption_kind = original_lang_node.attrib.get('kind', '')
829 for lang_node in caption_list.findall('target'):
830 sub_lang = lang_node.attrib['lang_code']
832 for ext in ['sbv', 'vtt', 'srt']:
833 params = compat_urllib_parse.urlencode({
834 'lang': original_lang,
838 'kind': caption_kind,
841 'url': caption_url + '&' + params,
844 sub_lang_list[sub_lang] = sub_formats
846 # An extractor error can be raise by the download process if there are
847 # no automatic captions but there are subtitles
848 except (KeyError, ExtractorError):
849 self._downloader.report_warning(err_msg)
853 def extract_id(cls, url):
854 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
856 raise ExtractorError('Invalid URL: %s' % url)
857 video_id = mobj.group(2)
860 def _extract_from_m3u8(self, manifest_url, video_id):
863 def _get_urls(_manifest):
864 lines = _manifest.split('\n')
865 urls = filter(lambda l: l and not l.startswith('#'),
868 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
869 formats_urls = _get_urls(manifest)
870 for format_url in formats_urls:
871 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
872 url_map[itag] = format_url
875 def _extract_annotations(self, video_id):
876 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
877 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
879 def _parse_dash_manifest(
880 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
881 def decrypt_sig(mobj):
883 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
884 return '/signature/%s' % dec_s
885 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
886 dash_doc = self._download_xml(
887 dash_manifest_url, video_id,
888 note='Downloading DASH manifest',
889 errnote='Could not download DASH manifest',
892 if dash_doc is False:
896 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
897 mime_type = a.attrib.get('mimeType')
898 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
899 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
902 if mime_type == 'text/vtt':
903 # TODO implement WebVTT downloading
905 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
906 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
907 format_id = r.attrib['id']
908 video_url = url_el.text
909 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
911 'format_id': format_id,
913 'width': int_or_none(r.attrib.get('width')),
914 'height': int_or_none(r.attrib.get('height')),
915 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
916 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
917 'filesize': filesize,
918 'fps': int_or_none(r.attrib.get('frameRate')),
920 if segment_list is not None:
922 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
923 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
924 'protocol': 'http_dash_segments',
927 existing_format = next(
929 if fo['format_id'] == format_id)
930 except StopIteration:
931 full_info = self._formats.get(format_id, {}).copy()
933 codecs = r.attrib.get('codecs')
935 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
936 full_info['vcodec'] = codecs
937 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
938 full_info['acodec'] = codecs
939 formats.append(full_info)
941 existing_format.update(f)
943 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
946 def _real_extract(self, url):
947 url, smuggled_data = unsmuggle_url(url, {})
950 'http' if self._downloader.params.get('prefer_insecure', False)
955 parsed_url = compat_urllib_parse_urlparse(url)
956 for component in [parsed_url.fragment, parsed_url.query]:
957 query = compat_parse_qs(component)
958 if start_time is None and 't' in query:
959 start_time = parse_duration(query['t'][0])
960 if start_time is None and 'start' in query:
961 start_time = parse_duration(query['start'][0])
962 if end_time is None and 'end' in query:
963 end_time = parse_duration(query['end'][0])
965 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
966 mobj = re.search(self._NEXT_URL_RE, url)
968 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
969 video_id = self.extract_id(url)
972 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
973 video_webpage = self._download_webpage(url, video_id)
975 # Attempt to extract SWF player URL
976 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
978 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
984 def add_dash_mpd(video_info):
985 dash_mpd = video_info.get('dashmpd')
986 if dash_mpd and dash_mpd[0] not in dash_mpds:
987 dash_mpds.append(dash_mpd[0])
992 if re.search(r'player-age-gate-content">', video_webpage) is not None:
994 # We simulate the access to the video from www.youtube.com/v/{video_id}
995 # this can be viewed without login into Youtube
996 url = proto + '://www.youtube.com/embed/%s' % video_id
997 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
998 data = compat_urllib_parse.urlencode({
999 'video_id': video_id,
1000 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1001 'sts': self._search_regex(
1002 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1004 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1005 video_info_webpage = self._download_webpage(
1006 video_info_url, video_id,
1007 note='Refetching age-gated info webpage',
1008 errnote='unable to download video info webpage')
1009 video_info = compat_parse_qs(video_info_webpage)
1010 add_dash_mpd(video_info)
1014 # Try looking directly into the video webpage
1015 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
1017 json_code = uppercase_escape(mobj.group(1))
1018 ytplayer_config = json.loads(json_code)
1019 args = ytplayer_config['args']
1020 if args.get('url_encoded_fmt_stream_map'):
1021 # Convert to the same format returned by compat_parse_qs
1022 video_info = dict((k, [v]) for k, v in args.items())
1023 add_dash_mpd(video_info)
1024 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1026 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1027 # We also try looking in get_video_info since it may contain different dashmpd
1028 # URL that points to a DASH manifest with possibly different itag set (some itags
1029 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1030 # manifest pointed by get_video_info's dashmpd).
1031 # The general idea is to take a union of itags of both DASH manifests (for example
1032 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1033 self.report_video_info_webpage_download(video_id)
1034 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1036 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1037 % (proto, video_id, el_type))
1038 video_info_webpage = self._download_webpage(
1040 video_id, note=False,
1041 errnote='unable to download video info webpage')
1042 get_video_info = compat_parse_qs(video_info_webpage)
1043 if get_video_info.get('use_cipher_signature') != ['True']:
1044 add_dash_mpd(get_video_info)
1046 video_info = get_video_info
1047 if 'token' in get_video_info:
1049 if 'token' not in video_info:
1050 if 'reason' in video_info:
1051 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1052 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1054 raise ExtractorError('YouTube said: This video is available in %s only' % (
1055 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1057 raise ExtractorError(
1058 'YouTube said: %s' % video_info['reason'][0],
1059 expected=True, video_id=video_id)
1061 raise ExtractorError(
1062 '"token" parameter not in video info for unknown reason',
1066 if 'title' in video_info:
1067 video_title = video_info['title'][0]
1069 self._downloader.report_warning('Unable to extract video title')
1073 video_description = get_element_by_id("eow-description", video_webpage)
1074 if video_description:
1075 video_description = re.sub(r'''(?x)
1077 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1079 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1080 class="yt-uix-redirect-link"\s*>
1083 ''', r'\1', video_description)
1084 video_description = clean_html(video_description)
1086 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1088 video_description = unescapeHTML(fd_mobj.group(1))
1090 video_description = ''
1092 if (not self._downloader.params.get('noplaylist') and
1093 not smuggled_data.get('force_singlefeed', False) and
1094 'multifeed_metadata_list' in video_info):
1096 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1097 for feed in multifeed_metadata_list.split(','):
1098 feed_data = compat_parse_qs(feed)
1100 '_type': 'url_transparent',
1101 'ie_key': 'Youtube',
1103 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1104 {'force_singlefeed': True}),
1105 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1107 return self.playlist_result(entries, video_id, video_title, video_description)
1109 if 'view_count' in video_info:
1110 view_count = int(video_info['view_count'][0])
1114 # Check for "rental" videos
1115 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1116 raise ExtractorError('"rental" videos not supported')
1118 # Start extracting information
1119 self.report_information_extraction(video_id)
1122 if 'author' not in video_info:
1123 raise ExtractorError('Unable to extract uploader name')
1124 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1127 video_uploader_id = None
1128 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1129 if mobj is not None:
1130 video_uploader_id = mobj.group(1)
1132 self._downloader.report_warning('unable to extract uploader nickname')
1135 # We try first to get a high quality image:
1136 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1137 video_webpage, re.DOTALL)
1138 if m_thumb is not None:
1139 video_thumbnail = m_thumb.group(1)
1140 elif 'thumbnail_url' not in video_info:
1141 self._downloader.report_warning('unable to extract video thumbnail')
1142 video_thumbnail = None
1143 else: # don't panic if we can't find it
1144 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1147 upload_date = self._html_search_meta(
1148 'datePublished', video_webpage, 'upload date', default=None)
1150 upload_date = self._search_regex(
1151 [r'(?s)id="eow-date.*?>(.*?)</span>',
1152 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1153 video_webpage, 'upload date', default=None)
1155 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1156 upload_date = unified_strdate(upload_date)
1158 m_cat_container = self._search_regex(
1159 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1160 video_webpage, 'categories', default=None)
1162 category = self._html_search_regex(
1163 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1165 video_categories = None if category is None else [category]
1167 video_categories = None
1169 def _extract_count(count_name):
1170 return str_to_int(self._search_regex(
1171 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1172 % re.escape(count_name),
1173 video_webpage, count_name, default=None))
1175 like_count = _extract_count('like')
1176 dislike_count = _extract_count('dislike')
1179 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1180 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1182 if 'length_seconds' not in video_info:
1183 self._downloader.report_warning('unable to extract video duration')
1184 video_duration = None
1186 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1189 video_annotations = None
1190 if self._downloader.params.get('writeannotations', False):
1191 video_annotations = self._extract_annotations(video_id)
1193 def _map_to_format_list(urlmap):
1195 for itag, video_real_url in urlmap.items():
1198 'url': video_real_url,
1199 'player_url': player_url,
1201 if itag in self._formats:
1202 dct.update(self._formats[itag])
1206 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1207 self.report_rtmp_download()
1209 'format_id': '_rtmp',
1211 'url': video_info['conn'][0],
1212 'player_url': player_url,
1214 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1215 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1216 if 'rtmpe%3Dyes' in encoded_url_map:
1217 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1219 for url_data_str in encoded_url_map.split(','):
1220 url_data = compat_parse_qs(url_data_str)
1221 if 'itag' not in url_data or 'url' not in url_data:
1223 format_id = url_data['itag'][0]
1224 url = url_data['url'][0]
1226 if 'sig' in url_data:
1227 url += '&signature=' + url_data['sig'][0]
1228 elif 's' in url_data:
1229 encrypted_sig = url_data['s'][0]
1230 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1232 jsplayer_url_json = self._search_regex(
1234 embed_webpage if age_gate else video_webpage,
1235 'JS player URL (1)', default=None)
1236 if not jsplayer_url_json and not age_gate:
1237 # We need the embed website after all
1238 if embed_webpage is None:
1239 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1240 embed_webpage = self._download_webpage(
1241 embed_url, video_id, 'Downloading embed webpage')
1242 jsplayer_url_json = self._search_regex(
1243 ASSETS_RE, embed_webpage, 'JS player URL')
1245 player_url = json.loads(jsplayer_url_json)
1246 if player_url is None:
1247 player_url_json = self._search_regex(
1248 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1249 video_webpage, 'age gate player URL')
1250 player_url = json.loads(player_url_json)
1252 if self._downloader.params.get('verbose'):
1253 if player_url is None:
1254 player_version = 'unknown'
1255 player_desc = 'unknown'
1257 if player_url.endswith('swf'):
1258 player_version = self._search_regex(
1259 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1260 'flash player', fatal=False)
1261 player_desc = 'flash player %s' % player_version
1263 player_version = self._search_regex(
1264 r'html5player-([^/]+?)(?:/html5player)?\.js',
1266 'html5 player', fatal=False)
1267 player_desc = 'html5 player %s' % player_version
1269 parts_sizes = self._signature_cache_id(encrypted_sig)
1270 self.to_screen('{%s} signature length %s, %s' %
1271 (format_id, parts_sizes, player_desc))
1273 signature = self._decrypt_signature(
1274 encrypted_sig, video_id, player_url, age_gate)
1275 url += '&signature=' + signature
1276 if 'ratebypass' not in url:
1277 url += '&ratebypass=yes'
1278 url_map[format_id] = url
1279 formats = _map_to_format_list(url_map)
1280 elif video_info.get('hlsvp'):
1281 manifest_url = video_info['hlsvp'][0]
1282 url_map = self._extract_from_m3u8(manifest_url, video_id)
1283 formats = _map_to_format_list(url_map)
1285 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1287 # Look for the DASH manifest
1288 if self._downloader.params.get('youtube_include_dash_manifest', True):
1289 dash_mpd_fatal = True
1290 for dash_manifest_url in dash_mpds:
1293 for df in self._parse_dash_manifest(
1294 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1295 # Do not overwrite DASH format found in some previous DASH manifest
1296 if df['format_id'] not in dash_formats:
1297 dash_formats[df['format_id']] = df
1298 # Additional DASH manifests may end up in HTTP Error 403 therefore
1299 # allow them to fail without bug report message if we already have
1300 # some DASH manifest succeeded. This is temporary workaround to reduce
1301 # burst of bug reports until we figure out the reason and whether it
1302 # can be fixed at all.
1303 dash_mpd_fatal = False
1304 except (ExtractorError, KeyError) as e:
1305 self.report_warning(
1306 'Skipping DASH manifest: %r' % e, video_id)
1308 # Remove the formats we found through non-DASH, they
1309 # contain less info and it can be wrong, because we use
1310 # fixed values (for example the resolution). See
1311 # https://github.com/rg3/youtube-dl/issues/5774 for an
1313 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1314 formats.extend(dash_formats.values())
1316 # Check for malformed aspect ratio
1317 stretched_m = re.search(
1318 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1321 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1323 if f.get('vcodec') != 'none':
1324 f['stretched_ratio'] = ratio
1326 self._sort_formats(formats)
1330 'uploader': video_uploader,
1331 'uploader_id': video_uploader_id,
1332 'upload_date': upload_date,
1333 'title': video_title,
1334 'thumbnail': video_thumbnail,
1335 'description': video_description,
1336 'categories': video_categories,
1337 'subtitles': video_subtitles,
1338 'automatic_captions': automatic_captions,
1339 'duration': video_duration,
1340 'age_limit': 18 if age_gate else 0,
1341 'annotations': video_annotations,
1342 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1343 'view_count': view_count,
1344 'like_count': like_count,
1345 'dislike_count': dislike_count,
1346 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1349 'start_time': start_time,
1350 'end_time': end_time,
1354 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1355 IE_DESC = 'YouTube.com playlists'
1356 _VALID_URL = r"""(?x)(?:
1361 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1362 \? (?:.*?&)*? (?:p|a|list)=
1366 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1367 # Top tracks, they can also include dots
1372 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1374 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1375 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1376 IE_NAME = 'youtube:playlist'
1378 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1380 'title': 'ytdl test PL',
1381 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1383 'playlist_count': 3,
1385 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1387 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1388 'title': 'YDL_Empty_List',
1390 'playlist_count': 0,
1392 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1393 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1395 'title': '29C3: Not my department',
1396 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1398 'playlist_count': 95,
1400 'note': 'issue #673',
1401 'url': 'PLBB231211A4F62143',
1403 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1404 'id': 'PLBB231211A4F62143',
1406 'playlist_mincount': 26,
1408 'note': 'Large playlist',
1409 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1411 'title': 'Uploads from Cauchemar',
1412 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1414 'playlist_mincount': 799,
1416 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1418 'title': 'YDL_safe_search',
1419 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1421 'playlist_count': 2,
1424 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1425 'playlist_count': 4,
1428 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1431 'note': 'Embedded SWF player',
1432 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1433 'playlist_count': 4,
1436 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1439 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1440 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1442 'title': 'Uploads from Interstellar Movie',
1443 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1445 'playlist_mincout': 21,
1448 def _real_initialize(self):
1451 def _extract_mix(self, playlist_id):
1452 # The mixes are generated from a single video
1453 # the id of the playlist is just 'RD' + video_id
1454 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1455 webpage = self._download_webpage(
1456 url, playlist_id, 'Downloading Youtube mix')
1457 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1459 search_title('playlist-title') or
1460 search_title('title long-title') or
1461 search_title('title'))
1462 title = clean_html(title_span)
1463 ids = orderedSet(re.findall(
1464 r'''(?xs)data-video-username=".*?".*?
1465 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1467 url_results = self._ids_to_results(ids)
1469 return self.playlist_result(url_results, playlist_id, title)
1471 def _extract_playlist(self, playlist_id):
1472 url = self._TEMPLATE_URL % playlist_id
1473 page = self._download_webpage(url, playlist_id)
1475 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1476 match = match.strip()
1477 # Check if the playlist exists or is private
1478 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1479 raise ExtractorError(
1480 'The playlist doesn\'t exist or is private, use --username or '
1481 '--netrc to access it.',
1483 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1484 raise ExtractorError(
1485 'Invalid parameters. Maybe URL is incorrect.',
1487 elif re.match(r'[^<]*Choose your language[^<]*', match):
1490 self.report_warning('Youtube gives an alert message: ' + match)
1492 # Extract the video ids from the playlist pages
1494 more_widget_html = content_html = page
1495 for page_num in itertools.count(1):
1496 matches = re.finditer(self._VIDEO_RE, content_html)
1497 # We remove the duplicates and the link with index 0
1498 # (it's not the first video of the playlist)
1499 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1500 for vid_id in new_ids:
1501 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1503 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1507 more = self._download_json(
1508 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1509 'Downloading page #%s' % page_num,
1510 transform_source=uppercase_escape)
1511 content_html = more['content_html']
1512 if not content_html.strip():
1513 # Some webpages show a "Load more" button but they don't
1516 more_widget_html = more['load_more_widget_html']
1518 playlist_title = self._html_search_regex(
1519 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1522 return self.playlist_result(_entries(), playlist_id, playlist_title)
1524 def _real_extract(self, url):
1525 # Extract playlist id
1526 mobj = re.match(self._VALID_URL, url)
1528 raise ExtractorError('Invalid URL: %s' % url)
1529 playlist_id = mobj.group(1) or mobj.group(2)
1531 # Check if it's a video-specific URL
1532 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1533 if 'v' in query_dict:
1534 video_id = query_dict['v'][0]
1535 if self._downloader.params.get('noplaylist'):
1536 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1537 return self.url_result(video_id, 'Youtube', video_id=video_id)
1539 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1541 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1542 # Mixes require a custom extraction process
1543 return self._extract_mix(playlist_id)
1545 return self._extract_playlist(playlist_id)
1548 class YoutubeChannelIE(InfoExtractor):
1549 IE_DESC = 'YouTube.com channels'
1550 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1551 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1552 IE_NAME = 'youtube:channel'
1554 'note': 'paginated channel',
1555 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1556 'playlist_mincount': 91,
1558 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1563 def extract_videos_from_page(page):
1566 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1567 video_id = mobj.group('id')
1568 video_title = unescapeHTML(mobj.group('title'))
1570 idx = ids_in_page.index(video_id)
1571 if video_title and not titles_in_page[idx]:
1572 titles_in_page[idx] = video_title
1574 ids_in_page.append(video_id)
1575 titles_in_page.append(video_title)
1576 return zip(ids_in_page, titles_in_page)
1578 def _real_extract(self, url):
1579 channel_id = self._match_id(url)
1581 url = self._TEMPLATE_URL % channel_id
1583 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1584 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1585 # otherwise fallback on channel by page extraction
1586 channel_page = self._download_webpage(
1587 url + '?view=57', channel_id,
1588 'Downloading channel page', fatal=False)
1589 channel_playlist_id = self._html_search_meta(
1590 'channelId', channel_page, 'channel id', default=None)
1591 if not channel_playlist_id:
1592 channel_playlist_id = self._search_regex(
1593 r'data-channel-external-id="([^"]+)"',
1594 channel_page, 'channel id', default=None)
1595 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1596 playlist_id = 'UU' + channel_playlist_id[2:]
1597 return self.url_result(
1598 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1600 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1601 autogenerated = re.search(r'''(?x)
1603 channel-header-autogenerated-label|
1604 yt-channel-title-autogenerated
1605 )[^"]*"''', channel_page) is not None
1608 # The videos are contained in a single page
1609 # the ajax pages can't be used, they are empty
1612 video_id, 'Youtube', video_id=video_id,
1613 video_title=video_title)
1614 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1615 return self.playlist_result(entries, channel_id)
1618 more_widget_html = content_html = channel_page
1619 for pagenum in itertools.count(1):
1621 for video_id, video_title in self.extract_videos_from_page(content_html):
1622 yield self.url_result(
1623 video_id, 'Youtube', video_id=video_id,
1624 video_title=video_title)
1627 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1632 more = self._download_json(
1633 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1634 'Downloading page #%s' % (pagenum + 1),
1635 transform_source=uppercase_escape)
1636 content_html = more['content_html']
1637 more_widget_html = more['load_more_widget_html']
1639 return self.playlist_result(_entries(), channel_id)
1642 class YoutubeUserIE(YoutubeChannelIE):
1643 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1644 _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_-]+)'
1645 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1646 IE_NAME = 'youtube:user'
1649 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1650 'playlist_mincount': 320,
1652 'title': 'TheLinuxFoundation',
1655 'url': 'ytuser:phihag',
1656 'only_matching': True,
1660 def suitable(cls, url):
1661 # Don't return True if the url can be extracted with other youtube
1662 # extractor, the regex would is too permissive and it would match.
1663 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1664 if any(ie.suitable(url) for ie in other_ies):
1667 return super(YoutubeUserIE, cls).suitable(url)
1670 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1671 IE_DESC = 'YouTube.com searches'
1672 # there doesn't appear to be a real limit, for example if you search for
1673 # 'python' you get more than 8.000.000 results
1674 _MAX_RESULTS = float('inf')
1675 IE_NAME = 'youtube:search'
1676 _SEARCH_KEY = 'ytsearch'
1677 _EXTRA_QUERY_ARGS = {}
1680 def _get_n_results(self, query, n):
1681 """Get a specified number of results for a query"""
1686 for pagenum in itertools.count(1):
1688 'search_query': query.encode('utf-8'),
1692 url_query.update(self._EXTRA_QUERY_ARGS)
1693 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1694 data = self._download_json(
1695 result_url, video_id='query "%s"' % query,
1696 note='Downloading page %s' % pagenum,
1697 errnote='Unable to download API page')
1698 html_content = data[1]['body']['content']
1700 if 'class="search-message' in html_content:
1701 raise ExtractorError(
1702 '[youtube] No video results', expected=True)
1704 new_videos = self._ids_to_results(orderedSet(re.findall(
1705 r'href="/watch\?v=(.{11})', html_content)))
1706 videos += new_videos
1707 if not new_videos or len(videos) > limit:
1712 return self.playlist_result(videos, query)
1715 class YoutubeSearchDateIE(YoutubeSearchIE):
1716 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1717 _SEARCH_KEY = 'ytsearchdate'
1718 IE_DESC = 'YouTube.com searches, newest videos first'
1719 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1722 class YoutubeSearchURLIE(InfoExtractor):
1723 IE_DESC = 'YouTube.com search URLs'
1724 IE_NAME = 'youtube:search_url'
1725 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1727 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1728 'playlist_mincount': 5,
1730 'title': 'youtube-dl test video',
1734 def _real_extract(self, url):
1735 mobj = re.match(self._VALID_URL, url)
1736 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1738 webpage = self._download_webpage(url, query)
1739 result_code = self._search_regex(
1740 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1742 part_codes = re.findall(
1743 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1745 for part_code in part_codes:
1746 part_title = self._html_search_regex(
1747 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1748 part_url_snippet = self._html_search_regex(
1749 r'(?s)href="([^"]+)"', part_code, 'item URL')
1750 part_url = compat_urlparse.urljoin(
1751 'https://www.youtube.com/', part_url_snippet)
1755 'title': part_title,
1759 '_type': 'playlist',
1765 class YoutubeShowIE(InfoExtractor):
1766 IE_DESC = 'YouTube.com (multi-season) shows'
1767 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1768 IE_NAME = 'youtube:show'
1770 'url': 'http://www.youtube.com/show/airdisasters',
1771 'playlist_mincount': 3,
1773 'id': 'airdisasters',
1774 'title': 'Air Disasters',
1778 def _real_extract(self, url):
1779 mobj = re.match(self._VALID_URL, url)
1780 playlist_id = mobj.group('id')
1781 webpage = self._download_webpage(
1782 url, playlist_id, 'Downloading show webpage')
1783 # There's one playlist for each season of the show
1784 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1785 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1788 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1789 for season in m_seasons
1791 title = self._og_search_title(webpage, fatal=False)
1794 '_type': 'playlist',
1801 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1803 Base class for feed extractors
1804 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1806 _LOGIN_REQUIRED = True
1810 return 'youtube:%s' % self._FEED_NAME
1812 def _real_initialize(self):
1815 def _real_extract(self, url):
1816 page = self._download_webpage(
1817 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1819 # The extraction process is the same as for playlists, but the regex
1820 # for the video ids doesn't contain an index
1822 more_widget_html = content_html = page
1823 for page_num in itertools.count(1):
1824 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1826 # 'recommended' feed has infinite 'load more' and each new portion spins
1827 # the same videos in (sometimes) slightly different order, so we'll check
1828 # for unicity and break when portion has no new videos
1829 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1835 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1839 more = self._download_json(
1840 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1841 'Downloading page #%s' % page_num,
1842 transform_source=uppercase_escape)
1843 content_html = more['content_html']
1844 more_widget_html = more['load_more_widget_html']
1846 return self.playlist_result(
1847 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1850 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1851 IE_NAME = 'youtube:watchlater'
1852 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1853 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1855 _TESTS = [] # override PlaylistIE tests
1857 def _real_extract(self, url):
1858 return self._extract_playlist('WL')
1861 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1862 IE_NAME = 'youtube:favorites'
1863 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1864 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1865 _LOGIN_REQUIRED = True
1867 def _real_extract(self, url):
1868 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1869 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1870 return self.url_result(playlist_id, 'YoutubePlaylist')
1873 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1874 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1875 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1876 _FEED_NAME = 'recommended'
1877 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1880 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1881 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1882 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1883 _FEED_NAME = 'subscriptions'
1884 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1887 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1888 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1889 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1890 _FEED_NAME = 'history'
1891 _PLAYLIST_TITLE = 'Youtube History'
1894 class YoutubeTruncatedURLIE(InfoExtractor):
1895 IE_NAME = 'youtube:truncated_url'
1896 IE_DESC = False # Do not list
1897 _VALID_URL = r'''(?x)
1899 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1902 annotation_id=annotation_[^&]+|
1907 attribution_link\?a=[^&]+
1913 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1914 'only_matching': True,
1916 'url': 'http://www.youtube.com/watch?',
1917 'only_matching': True,
1919 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1920 'only_matching': True,
1922 'url': 'https://www.youtube.com/watch?feature=foo',
1923 'only_matching': True,
1925 'url': 'https://www.youtube.com/watch?hl=en-GB',
1926 'only_matching': True,
1929 def _real_extract(self, url):
1930 raise ExtractorError(
1931 'Did you forget to quote the URL? Remember that & is a meta '
1932 'character in most shells, so you want to put the URL in quotes, '
1934 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1935 ' or simply youtube-dl BaW_jenozKc .',
1939 class YoutubeTruncatedIDIE(InfoExtractor):
1940 IE_NAME = 'youtube:truncated_id'
1941 IE_DESC = False # Do not list
1942 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1945 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1946 'only_matching': True,
1949 def _real_extract(self, url):
1950 video_id = self._match_id(url)
1951 raise ExtractorError(
1952 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),