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_request,
28 get_element_by_attribute,
40 class YoutubeBaseInfoExtractor(InfoExtractor):
41 """Provide base functions for Youtube extractors"""
42 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
43 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
44 _NETRC_MACHINE = 'youtube'
45 # If True it will raise an error if no login info is provided
46 _LOGIN_REQUIRED = False
48 def _set_language(self):
50 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
51 # YouTube sets the expire time to about two months
52 expire_time=time.time() + 2 * 30 * 24 * 3600)
54 def _ids_to_results(self, ids):
56 self.url_result(vid_id, 'Youtube', video_id=vid_id)
61 Attempt to log in to YouTube.
62 True is returned if successful or skipped.
63 False is returned if login failed.
65 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
67 (username, password) = self._get_login_info()
68 # No authentication to be performed
70 if self._LOGIN_REQUIRED:
71 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
74 login_page = self._download_webpage(
75 self._LOGIN_URL, None,
76 note='Downloading login page',
77 errnote='unable to fetch login page', fatal=False)
78 if login_page is False:
81 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
82 login_page, 'Login GALX parameter')
86 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
91 'PersistentCookie': 'yes',
93 'bgresponse': 'js_disabled',
94 'checkConnection': '',
95 'checkedDomains': 'youtube',
102 'service': 'youtube',
107 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
109 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
110 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
112 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
113 login_results = self._download_webpage(
115 note='Logging in', errnote='unable to log in', fatal=False)
116 if login_results is False:
119 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
120 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
123 # TODO add SMS and phone call support - these require making a request and then prompting the user
125 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
126 tfa_code = self._get_tfa_info()
129 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
130 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
133 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
135 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
137 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
138 secTok = match.group(1)
139 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
141 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
142 timeStmp = match.group(1)
145 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
147 'smsUserPin': tfa_code,
148 'smsVerifyPin': 'Verify',
150 'PersistentCookie': 'yes',
151 'checkConnection': '',
152 'checkedDomains': 'youtube',
155 'timeStmp': timeStmp,
156 'service': 'youtube',
159 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
160 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
162 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
163 tfa_results = self._download_webpage(
165 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
167 if tfa_results is False:
170 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
171 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
173 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
174 self._downloader.report_warning('unable to log in - did the page structure change?')
176 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
177 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
180 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
181 self._downloader.report_warning('unable to log in: bad username or password')
185 def _real_initialize(self):
186 if self._downloader is None:
189 if not self._login():
193 class YoutubeIE(YoutubeBaseInfoExtractor):
194 IE_DESC = 'YouTube.com'
195 _VALID_URL = r"""(?x)^
197 (?:https?://|//) # http(s):// or protocol-independent URL
198 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
199 (?:www\.)?deturl\.com/www\.youtube\.com/|
200 (?:www\.)?pwnyoutube\.com/|
201 (?:www\.)?yourepeat\.com/|
202 tube\.majestyc\.net/|
203 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
204 (?:.*?\#/)? # handle anchor (#/) redirect urls
205 (?: # the various things that can precede the ID:
206 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
207 |(?: # or the v= param in all its forms
208 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
209 (?:\?|\#!?) # the params delimiter ? or # or #!
210 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
214 |youtu\.be/ # just youtu.be/xxxx
215 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
217 )? # all until now is optional -> you can pass the naked ID
218 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
219 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
220 (?(1).+)? # if we found the ID, everything can follow
222 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
224 '5': {'ext': 'flv', 'width': 400, 'height': 240},
225 '6': {'ext': 'flv', 'width': 450, 'height': 270},
226 '13': {'ext': '3gp'},
227 '17': {'ext': '3gp', 'width': 176, 'height': 144},
228 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
229 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
230 '34': {'ext': 'flv', 'width': 640, 'height': 360},
231 '35': {'ext': 'flv', 'width': 854, 'height': 480},
232 '36': {'ext': '3gp', 'width': 320, 'height': 240},
233 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
234 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
235 '43': {'ext': 'webm', 'width': 640, 'height': 360},
236 '44': {'ext': 'webm', 'width': 854, 'height': 480},
237 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
238 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
239 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
240 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
244 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
245 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
246 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
247 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
248 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
249 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
250 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
252 # Apple HTTP Live Streaming
253 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
254 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
255 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
256 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
257 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
258 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
259 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
262 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
263 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
264 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
265 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
266 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
268 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
271 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
272 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
275 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
276 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
277 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
280 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
281 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
282 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
283 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
284 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
287 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
289 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
297 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
298 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
299 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
300 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
303 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
304 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
306 # Dash webm audio with opus inside
307 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
308 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
309 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
312 '_rtmp': {'protocol': 'rtmp'},
318 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
322 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
323 'uploader': 'Philipp Hagemeister',
324 'uploader_id': 'phihag',
325 'upload_date': '20121002',
326 '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 .',
327 'categories': ['Science & Technology'],
329 'dislike_count': int,
333 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
334 'note': 'Test generic use_cipher_signature video (#897)',
338 'upload_date': '20120506',
339 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
340 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
341 'uploader': 'Icona Pop',
342 'uploader_id': 'IconaPop',
346 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
347 'note': 'Test VEVO video with age protection (#956)',
351 'upload_date': '20130703',
352 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
353 'description': 'md5:64249768eec3bc4276236606ea996373',
354 'uploader': 'justintimberlakeVEVO',
355 'uploader_id': 'justintimberlakeVEVO',
359 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
360 'note': 'Embed-only video (#1746)',
364 'upload_date': '20120608',
365 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
366 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
367 'uploader': 'SET India',
368 'uploader_id': 'setindia'
372 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
373 'note': '256k DASH audio (format 141) via DASH manifest',
377 'upload_date': '20121002',
378 'uploader_id': '8KVIDEO',
380 'uploader': '8KVIDEO',
381 'title': 'UHDTV TEST 8K VIDEO.mp4'
384 'youtube_include_dash_manifest': True,
388 # DASH manifest with encrypted signature
390 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
394 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
395 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
396 'uploader': 'AfrojackVEVO',
397 'uploader_id': 'AfrojackVEVO',
398 'upload_date': '20131011',
401 'youtube_include_dash_manifest': True,
405 # JS player signature function name containing $
407 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
411 'title': 'Taylor Swift - Shake It Off',
412 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
413 'uploader': 'TaylorSwiftVEVO',
414 'uploader_id': 'TaylorSwiftVEVO',
415 'upload_date': '20140818',
418 'youtube_include_dash_manifest': True,
424 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
428 'upload_date': '20100909',
429 'uploader': 'The Amazing Atheist',
430 'uploader_id': 'TheAmazingAtheist',
431 'title': 'Burning Everyone\'s Koran',
432 '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',
435 # Normal age-gate video (No vevo, embed allowed)
437 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
441 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
442 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
443 'uploader': 'The Witcher',
444 'uploader_id': 'WitcherGame',
445 'upload_date': '20140605',
448 # Age-gate video with encrypted signature
450 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
454 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
455 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
456 'uploader': 'LloydVEVO',
457 'uploader_id': 'LloydVEVO',
458 'upload_date': '20110629',
461 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
463 'url': '__2ABJjxzNo',
467 'upload_date': '20100430',
468 'uploader_id': 'deadmau5',
469 'description': 'md5:12c56784b8032162bb936a5f76d55360',
470 'uploader': 'deadmau5',
471 'title': 'Deadmau5 - Some Chords (HD)',
473 'expected_warnings': [
474 'DASH manifest missing',
477 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
479 'url': 'lqQg6PlCWgI',
483 'upload_date': '20120731',
484 'uploader_id': 'olympic',
485 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
486 'uploader': 'Olympics',
487 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
490 'skip_download': 'requires avconv',
495 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
499 'stretched_ratio': 16 / 9.,
500 'upload_date': '20110310',
501 'uploader_id': 'AllenMeow',
502 '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',
504 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
507 # url_encoded_fmt_stream_map is empty string
509 'url': 'qEJwOuvDf7I',
513 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
515 'upload_date': '20150404',
516 'uploader_id': 'spbelect',
517 'uploader': 'Наблюдатели Петербурга',
520 'skip_download': 'requires avconv',
523 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
525 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
529 'title': 'md5:7b81415841e02ecd4313668cde88737a',
530 'description': 'md5:116377fd2963b81ec4ce64b542173306',
531 'upload_date': '20150625',
532 'uploader_id': 'dorappi2000',
533 'uploader': 'dorappi2000',
534 'formats': 'mincount:33',
539 def __init__(self, *args, **kwargs):
540 super(YoutubeIE, self).__init__(*args, **kwargs)
541 self._player_cache = {}
543 def report_video_info_webpage_download(self, video_id):
544 """Report attempt to download video info webpage."""
545 self.to_screen('%s: Downloading video info webpage' % video_id)
547 def report_information_extraction(self, video_id):
548 """Report attempt to extract video information."""
549 self.to_screen('%s: Extracting video information' % video_id)
551 def report_unavailable_format(self, video_id, format):
552 """Report extracted video URL."""
553 self.to_screen('%s: Format %s not available' % (video_id, format))
555 def report_rtmp_download(self):
556 """Indicate the download will use the RTMP protocol."""
557 self.to_screen('RTMP download detected')
559 def _signature_cache_id(self, example_sig):
560 """ Return a string representation of a signature """
561 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
563 def _extract_signature_function(self, video_id, player_url, example_sig):
565 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
568 raise ExtractorError('Cannot identify player %r' % player_url)
569 player_type = id_m.group('ext')
570 player_id = id_m.group('id')
572 # Read from filesystem cache
573 func_id = '%s_%s_%s' % (
574 player_type, player_id, self._signature_cache_id(example_sig))
575 assert os.path.basename(func_id) == func_id
577 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
578 if cache_spec is not None:
579 return lambda s: ''.join(s[i] for i in cache_spec)
582 'Downloading player %s' % player_url
583 if self._downloader.params.get('verbose') else
584 'Downloading %s player %s' % (player_type, player_id)
586 if player_type == 'js':
587 code = self._download_webpage(
588 player_url, video_id,
590 errnote='Download of %s failed' % player_url)
591 res = self._parse_sig_js(code)
592 elif player_type == 'swf':
593 urlh = self._request_webpage(
594 player_url, video_id,
596 errnote='Download of %s failed' % player_url)
598 res = self._parse_sig_swf(code)
600 assert False, 'Invalid player type %r' % player_type
602 test_string = ''.join(map(compat_chr, range(len(example_sig))))
603 cache_res = res(test_string)
604 cache_spec = [ord(c) for c in cache_res]
606 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
609 def _print_sig_code(self, func, example_sig):
610 def gen_sig_code(idxs):
611 def _genslice(start, end, step):
612 starts = '' if start == 0 else str(start)
613 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
614 steps = '' if step == 1 else (':%d' % step)
615 return 's[%s%s%s]' % (starts, ends, steps)
618 # Quelch pyflakes warnings - start will be set when step is set
619 start = '(Never used)'
620 for i, prev in zip(idxs[1:], idxs[:-1]):
624 yield _genslice(start, prev, step)
627 if i - prev in [-1, 1]:
636 yield _genslice(start, i, step)
638 test_string = ''.join(map(compat_chr, range(len(example_sig))))
639 cache_res = func(test_string)
640 cache_spec = [ord(c) for c in cache_res]
641 expr_code = ' + '.join(gen_sig_code(cache_spec))
642 signature_id_tuple = '(%s)' % (
643 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
644 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
645 ' return %s\n') % (signature_id_tuple, expr_code)
646 self.to_screen('Extracted signature function:\n' + code)
648 def _parse_sig_js(self, jscode):
649 funcname = self._search_regex(
650 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
651 'Initial JS player signature function name')
653 jsi = JSInterpreter(jscode)
654 initial_function = jsi.extract_function(funcname)
655 return lambda s: initial_function([s])
657 def _parse_sig_swf(self, file_contents):
658 swfi = SWFInterpreter(file_contents)
659 TARGET_CLASSNAME = 'SignatureDecipher'
660 searched_class = swfi.extract_class(TARGET_CLASSNAME)
661 initial_function = swfi.extract_function(searched_class, 'decipher')
662 return lambda s: initial_function([s])
664 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
665 """Turn the encrypted s field into a working signature"""
667 if player_url is None:
668 raise ExtractorError('Cannot decrypt signature without player_url')
670 if player_url.startswith('//'):
671 player_url = 'https:' + player_url
673 player_id = (player_url, self._signature_cache_id(s))
674 if player_id not in self._player_cache:
675 func = self._extract_signature_function(
676 video_id, player_url, s
678 self._player_cache[player_id] = func
679 func = self._player_cache[player_id]
680 if self._downloader.params.get('youtube_print_sig_code'):
681 self._print_sig_code(func, s)
683 except Exception as e:
684 tb = traceback.format_exc()
685 raise ExtractorError(
686 'Signature extraction failed: ' + tb, cause=e)
688 def _get_subtitles(self, video_id, webpage):
690 subs_doc = self._download_xml(
691 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
692 video_id, note=False)
693 except ExtractorError as err:
694 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
698 for track in subs_doc.findall('track'):
699 lang = track.attrib['lang_code']
700 if lang in sub_lang_list:
703 for ext in ['sbv', 'vtt', 'srt']:
704 params = compat_urllib_parse.urlencode({
708 'name': track.attrib['name'].encode('utf-8'),
711 'url': 'https://www.youtube.com/api/timedtext?' + params,
714 sub_lang_list[lang] = sub_formats
715 if not sub_lang_list:
716 self._downloader.report_warning('video doesn\'t have subtitles')
720 def _get_automatic_captions(self, video_id, webpage):
721 """We need the webpage for getting the captions url, pass it as an
722 argument to speed up the process."""
723 self.to_screen('%s: Looking for automatic captions' % video_id)
724 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
725 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
727 self._downloader.report_warning(err_msg)
729 player_config = json.loads(mobj.group(1))
731 args = player_config['args']
732 caption_url = args['ttsurl']
733 timestamp = args['timestamp']
734 # We get the available subtitles
735 list_params = compat_urllib_parse.urlencode({
740 list_url = caption_url + '&' + list_params
741 caption_list = self._download_xml(list_url, video_id)
742 original_lang_node = caption_list.find('track')
743 if original_lang_node is None:
744 self._downloader.report_warning('Video doesn\'t have automatic captions')
746 original_lang = original_lang_node.attrib['lang_code']
747 caption_kind = original_lang_node.attrib.get('kind', '')
750 for lang_node in caption_list.findall('target'):
751 sub_lang = lang_node.attrib['lang_code']
753 for ext in ['sbv', 'vtt', 'srt']:
754 params = compat_urllib_parse.urlencode({
755 'lang': original_lang,
759 'kind': caption_kind,
762 'url': caption_url + '&' + params,
765 sub_lang_list[sub_lang] = sub_formats
767 # An extractor error can be raise by the download process if there are
768 # no automatic captions but there are subtitles
769 except (KeyError, ExtractorError):
770 self._downloader.report_warning(err_msg)
774 def extract_id(cls, url):
775 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
777 raise ExtractorError('Invalid URL: %s' % url)
778 video_id = mobj.group(2)
781 def _extract_from_m3u8(self, manifest_url, video_id):
784 def _get_urls(_manifest):
785 lines = _manifest.split('\n')
786 urls = filter(lambda l: l and not l.startswith('#'),
789 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
790 formats_urls = _get_urls(manifest)
791 for format_url in formats_urls:
792 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
793 url_map[itag] = format_url
796 def _extract_annotations(self, video_id):
797 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
798 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
800 def _parse_dash_manifest(
801 self, video_id, dash_manifest_url, player_url, age_gate):
802 def decrypt_sig(mobj):
804 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
805 return '/signature/%s' % dec_s
806 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
807 dash_doc = self._download_xml(
808 dash_manifest_url, video_id,
809 note='Downloading DASH manifest',
810 errnote='Could not download DASH manifest')
813 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
814 mime_type = a.attrib.get('mimeType')
815 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
816 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
819 if mime_type == 'text/vtt':
820 # TODO implement WebVTT downloading
822 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
823 format_id = r.attrib['id']
824 video_url = url_el.text
825 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
827 'format_id': format_id,
829 'width': int_or_none(r.attrib.get('width')),
830 'height': int_or_none(r.attrib.get('height')),
831 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
832 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
833 'filesize': filesize,
834 'fps': int_or_none(r.attrib.get('frameRate')),
837 existing_format = next(
839 if fo['format_id'] == format_id)
840 except StopIteration:
841 full_info = self._formats.get(format_id, {}).copy()
843 codecs = r.attrib.get('codecs')
845 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
846 full_info['vcodec'] = codecs
847 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
848 full_info['acodec'] = codecs
849 formats.append(full_info)
851 existing_format.update(f)
853 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
856 def _real_extract(self, url):
858 'http' if self._downloader.params.get('prefer_insecure', False)
861 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
862 mobj = re.search(self._NEXT_URL_RE, url)
864 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
865 video_id = self.extract_id(url)
868 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
869 video_webpage = self._download_webpage(url, video_id)
871 # Attempt to extract SWF player URL
872 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
874 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
880 def add_dash_mpd(video_info):
881 dash_mpd = video_info.get('dashmpd')
882 if dash_mpd and dash_mpd[0] not in dash_mpds:
883 dash_mpds.append(dash_mpd[0])
887 if re.search(r'player-age-gate-content">', video_webpage) is not None:
889 # We simulate the access to the video from www.youtube.com/v/{video_id}
890 # this can be viewed without login into Youtube
891 url = proto + '://www.youtube.com/embed/%s' % video_id
892 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
893 data = compat_urllib_parse.urlencode({
894 'video_id': video_id,
895 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
896 'sts': self._search_regex(
897 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
899 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
900 video_info_webpage = self._download_webpage(
901 video_info_url, video_id,
902 note='Refetching age-gated info webpage',
903 errnote='unable to download video info webpage')
904 video_info = compat_parse_qs(video_info_webpage)
905 add_dash_mpd(video_info)
909 # Try looking directly into the video webpage
910 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
912 json_code = uppercase_escape(mobj.group(1))
913 ytplayer_config = json.loads(json_code)
914 args = ytplayer_config['args']
915 if args.get('url_encoded_fmt_stream_map'):
916 # Convert to the same format returned by compat_parse_qs
917 video_info = dict((k, [v]) for k, v in args.items())
918 add_dash_mpd(video_info)
919 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
920 # We also try looking in get_video_info since it may contain different dashmpd
921 # URL that points to a DASH manifest with possibly different itag set (some itags
922 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
923 # manifest pointed by get_video_info's dashmpd).
924 # The general idea is to take a union of itags of both DASH manifests (for example
925 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
926 self.report_video_info_webpage_download(video_id)
927 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
929 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
930 % (proto, video_id, el_type))
931 video_info_webpage = self._download_webpage(
933 video_id, note=False,
934 errnote='unable to download video info webpage')
935 get_video_info = compat_parse_qs(video_info_webpage)
936 add_dash_mpd(get_video_info)
938 video_info = get_video_info
939 if 'token' in get_video_info:
941 if 'token' not in video_info:
942 if 'reason' in video_info:
943 if 'The uploader has not made this video available in your country.' in video_info['reason']:
944 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
945 if regions_allowed is not None:
946 raise ExtractorError('YouTube said: This video is available in %s only' % (
947 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
949 raise ExtractorError(
950 'YouTube said: %s' % video_info['reason'][0],
951 expected=True, video_id=video_id)
953 raise ExtractorError(
954 '"token" parameter not in video info for unknown reason',
957 if 'view_count' in video_info:
958 view_count = int(video_info['view_count'][0])
962 # Check for "rental" videos
963 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
964 raise ExtractorError('"rental" videos not supported')
966 # Start extracting information
967 self.report_information_extraction(video_id)
970 if 'author' not in video_info:
971 raise ExtractorError('Unable to extract uploader name')
972 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
975 video_uploader_id = None
976 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
978 video_uploader_id = mobj.group(1)
980 self._downloader.report_warning('unable to extract uploader nickname')
983 if 'title' in video_info:
984 video_title = video_info['title'][0]
986 self._downloader.report_warning('Unable to extract video title')
990 # We try first to get a high quality image:
991 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
992 video_webpage, re.DOTALL)
993 if m_thumb is not None:
994 video_thumbnail = m_thumb.group(1)
995 elif 'thumbnail_url' not in video_info:
996 self._downloader.report_warning('unable to extract video thumbnail')
997 video_thumbnail = None
998 else: # don't panic if we can't find it
999 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
1003 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
1006 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.*?)</strong>',
1008 if mobj is not None:
1009 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1010 upload_date = unified_strdate(upload_date)
1012 m_cat_container = self._search_regex(
1013 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1014 video_webpage, 'categories', default=None)
1016 category = self._html_search_regex(
1017 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1019 video_categories = None if category is None else [category]
1021 video_categories = None
1024 video_description = get_element_by_id("eow-description", video_webpage)
1025 if video_description:
1026 video_description = re.sub(r'''(?x)
1028 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1030 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1031 class="yt-uix-redirect-link"\s*>
1034 ''', r'\1', video_description)
1035 video_description = clean_html(video_description)
1037 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1039 video_description = unescapeHTML(fd_mobj.group(1))
1041 video_description = ''
1043 def _extract_count(count_name):
1044 return str_to_int(self._search_regex(
1045 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1046 % re.escape(count_name),
1047 video_webpage, count_name, default=None))
1049 like_count = _extract_count('like')
1050 dislike_count = _extract_count('dislike')
1053 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1054 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1056 if 'length_seconds' not in video_info:
1057 self._downloader.report_warning('unable to extract video duration')
1058 video_duration = None
1060 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
1063 video_annotations = None
1064 if self._downloader.params.get('writeannotations', False):
1065 video_annotations = self._extract_annotations(video_id)
1067 def _map_to_format_list(urlmap):
1069 for itag, video_real_url in urlmap.items():
1072 'url': video_real_url,
1073 'player_url': player_url,
1075 if itag in self._formats:
1076 dct.update(self._formats[itag])
1080 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1081 self.report_rtmp_download()
1083 'format_id': '_rtmp',
1085 'url': video_info['conn'][0],
1086 'player_url': player_url,
1088 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1089 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1090 if 'rtmpe%3Dyes' in encoded_url_map:
1091 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1093 for url_data_str in encoded_url_map.split(','):
1094 url_data = compat_parse_qs(url_data_str)
1095 if 'itag' not in url_data or 'url' not in url_data:
1097 format_id = url_data['itag'][0]
1098 url = url_data['url'][0]
1100 if 'sig' in url_data:
1101 url += '&signature=' + url_data['sig'][0]
1102 elif 's' in url_data:
1103 encrypted_sig = url_data['s'][0]
1104 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1106 jsplayer_url_json = self._search_regex(
1108 embed_webpage if age_gate else video_webpage,
1109 'JS player URL (1)', default=None)
1110 if not jsplayer_url_json and not age_gate:
1111 # We need the embed website after all
1112 if embed_webpage is None:
1113 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1114 embed_webpage = self._download_webpage(
1115 embed_url, video_id, 'Downloading embed webpage')
1116 jsplayer_url_json = self._search_regex(
1117 ASSETS_RE, embed_webpage, 'JS player URL')
1119 player_url = json.loads(jsplayer_url_json)
1120 if player_url is None:
1121 player_url_json = self._search_regex(
1122 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1123 video_webpage, 'age gate player URL')
1124 player_url = json.loads(player_url_json)
1126 if self._downloader.params.get('verbose'):
1127 if player_url is None:
1128 player_version = 'unknown'
1129 player_desc = 'unknown'
1131 if player_url.endswith('swf'):
1132 player_version = self._search_regex(
1133 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1134 'flash player', fatal=False)
1135 player_desc = 'flash player %s' % player_version
1137 player_version = self._search_regex(
1138 r'html5player-([^/]+?)(?:/html5player)?\.js',
1140 'html5 player', fatal=False)
1141 player_desc = 'html5 player %s' % player_version
1143 parts_sizes = self._signature_cache_id(encrypted_sig)
1144 self.to_screen('{%s} signature length %s, %s' %
1145 (format_id, parts_sizes, player_desc))
1147 signature = self._decrypt_signature(
1148 encrypted_sig, video_id, player_url, age_gate)
1149 url += '&signature=' + signature
1150 if 'ratebypass' not in url:
1151 url += '&ratebypass=yes'
1152 url_map[format_id] = url
1153 formats = _map_to_format_list(url_map)
1154 elif video_info.get('hlsvp'):
1155 manifest_url = video_info['hlsvp'][0]
1156 url_map = self._extract_from_m3u8(manifest_url, video_id)
1157 formats = _map_to_format_list(url_map)
1159 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1161 # Look for the DASH manifest
1162 if self._downloader.params.get('youtube_include_dash_manifest', True):
1163 for dash_manifest_url in dash_mpds:
1166 for df in self._parse_dash_manifest(
1167 video_id, dash_manifest_url, player_url, age_gate):
1168 # Do not overwrite DASH format found in some previous DASH manifest
1169 if df['format_id'] not in dash_formats:
1170 dash_formats[df['format_id']] = df
1171 except (ExtractorError, KeyError) as e:
1172 self.report_warning(
1173 'Skipping DASH manifest: %r' % e, video_id)
1175 # Remove the formats we found through non-DASH, they
1176 # contain less info and it can be wrong, because we use
1177 # fixed values (for example the resolution). See
1178 # https://github.com/rg3/youtube-dl/issues/5774 for an
1180 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1181 formats.extend(dash_formats.values())
1183 # Check for malformed aspect ratio
1184 stretched_m = re.search(
1185 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1188 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1190 if f.get('vcodec') != 'none':
1191 f['stretched_ratio'] = ratio
1193 self._sort_formats(formats)
1197 'uploader': video_uploader,
1198 'uploader_id': video_uploader_id,
1199 'upload_date': upload_date,
1200 'title': video_title,
1201 'thumbnail': video_thumbnail,
1202 'description': video_description,
1203 'categories': video_categories,
1204 'subtitles': video_subtitles,
1205 'automatic_captions': automatic_captions,
1206 'duration': video_duration,
1207 'age_limit': 18 if age_gate else 0,
1208 'annotations': video_annotations,
1209 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1210 'view_count': view_count,
1211 'like_count': like_count,
1212 'dislike_count': dislike_count,
1213 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1218 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1219 IE_DESC = 'YouTube.com playlists'
1220 _VALID_URL = r"""(?x)(?:
1225 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1226 \? (?:.*?&)*? (?:p|a|list)=
1230 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1231 # Top tracks, they can also include dots
1236 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1238 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1239 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1240 IE_NAME = 'youtube:playlist'
1242 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1244 'title': 'ytdl test PL',
1245 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1247 'playlist_count': 3,
1249 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1251 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1252 'title': 'YDL_Empty_List',
1254 'playlist_count': 0,
1256 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1257 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1259 'title': '29C3: Not my department',
1260 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1262 'playlist_count': 95,
1264 'note': 'issue #673',
1265 'url': 'PLBB231211A4F62143',
1267 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1268 'id': 'PLBB231211A4F62143',
1270 'playlist_mincount': 26,
1272 'note': 'Large playlist',
1273 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1275 'title': 'Uploads from Cauchemar',
1276 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1278 'playlist_mincount': 799,
1280 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1282 'title': 'YDL_safe_search',
1283 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1285 'playlist_count': 2,
1288 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1289 'playlist_count': 4,
1292 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1295 'note': 'Embedded SWF player',
1296 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1297 'playlist_count': 4,
1300 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1303 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1304 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1306 'title': 'Uploads from Interstellar Movie',
1307 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1309 'playlist_mincout': 21,
1312 def _real_initialize(self):
1315 def _extract_mix(self, playlist_id):
1316 # The mixes are generated from a single video
1317 # the id of the playlist is just 'RD' + video_id
1318 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1319 webpage = self._download_webpage(
1320 url, playlist_id, 'Downloading Youtube mix')
1321 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1323 search_title('playlist-title') or
1324 search_title('title long-title') or
1325 search_title('title'))
1326 title = clean_html(title_span)
1327 ids = orderedSet(re.findall(
1328 r'''(?xs)data-video-username=".*?".*?
1329 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1331 url_results = self._ids_to_results(ids)
1333 return self.playlist_result(url_results, playlist_id, title)
1335 def _extract_playlist(self, playlist_id):
1336 url = self._TEMPLATE_URL % playlist_id
1337 page = self._download_webpage(url, playlist_id)
1339 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1340 match = match.strip()
1341 # Check if the playlist exists or is private
1342 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1343 raise ExtractorError(
1344 'The playlist doesn\'t exist or is private, use --username or '
1345 '--netrc to access it.',
1347 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1348 raise ExtractorError(
1349 'Invalid parameters. Maybe URL is incorrect.',
1351 elif re.match(r'[^<]*Choose your language[^<]*', match):
1354 self.report_warning('Youtube gives an alert message: ' + match)
1356 # Extract the video ids from the playlist pages
1358 more_widget_html = content_html = page
1359 for page_num in itertools.count(1):
1360 matches = re.finditer(self._VIDEO_RE, content_html)
1361 # We remove the duplicates and the link with index 0
1362 # (it's not the first video of the playlist)
1363 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1364 for vid_id in new_ids:
1365 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1367 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1371 more = self._download_json(
1372 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1373 'Downloading page #%s' % page_num,
1374 transform_source=uppercase_escape)
1375 content_html = more['content_html']
1376 if not content_html.strip():
1377 # Some webpages show a "Load more" button but they don't
1380 more_widget_html = more['load_more_widget_html']
1382 playlist_title = self._html_search_regex(
1383 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1386 return self.playlist_result(_entries(), playlist_id, playlist_title)
1388 def _real_extract(self, url):
1389 # Extract playlist id
1390 mobj = re.match(self._VALID_URL, url)
1392 raise ExtractorError('Invalid URL: %s' % url)
1393 playlist_id = mobj.group(1) or mobj.group(2)
1395 # Check if it's a video-specific URL
1396 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1397 if 'v' in query_dict:
1398 video_id = query_dict['v'][0]
1399 if self._downloader.params.get('noplaylist'):
1400 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1401 return self.url_result(video_id, 'Youtube', video_id=video_id)
1403 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1405 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1406 # Mixes require a custom extraction process
1407 return self._extract_mix(playlist_id)
1409 return self._extract_playlist(playlist_id)
1412 class YoutubeChannelIE(InfoExtractor):
1413 IE_DESC = 'YouTube.com channels'
1414 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1415 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1416 IE_NAME = 'youtube:channel'
1418 'note': 'paginated channel',
1419 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1420 'playlist_mincount': 91,
1422 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1427 def extract_videos_from_page(page):
1430 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1431 video_id = mobj.group('id')
1432 video_title = unescapeHTML(mobj.group('title'))
1434 idx = ids_in_page.index(video_id)
1435 if video_title and not titles_in_page[idx]:
1436 titles_in_page[idx] = video_title
1438 ids_in_page.append(video_id)
1439 titles_in_page.append(video_title)
1440 return zip(ids_in_page, titles_in_page)
1442 def _real_extract(self, url):
1443 channel_id = self._match_id(url)
1445 url = self._TEMPLATE_URL % channel_id
1447 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1448 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1449 # otherwise fallback on channel by page extraction
1450 channel_page = self._download_webpage(
1451 url + '?view=57', channel_id,
1452 'Downloading channel page', fatal=False)
1453 channel_playlist_id = self._html_search_meta(
1454 'channelId', channel_page, 'channel id', default=None)
1455 if not channel_playlist_id:
1456 channel_playlist_id = self._search_regex(
1457 r'data-channel-external-id="([^"]+)"',
1458 channel_page, 'channel id', default=None)
1459 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1460 playlist_id = 'UU' + channel_playlist_id[2:]
1461 return self.url_result(
1462 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1464 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1465 autogenerated = re.search(r'''(?x)
1467 channel-header-autogenerated-label|
1468 yt-channel-title-autogenerated
1469 )[^"]*"''', channel_page) is not None
1472 # The videos are contained in a single page
1473 # the ajax pages can't be used, they are empty
1476 video_id, 'Youtube', video_id=video_id,
1477 video_title=video_title)
1478 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1479 return self.playlist_result(entries, channel_id)
1482 more_widget_html = content_html = channel_page
1483 for pagenum in itertools.count(1):
1485 for video_id, video_title in self.extract_videos_from_page(content_html):
1486 yield self.url_result(
1487 video_id, 'Youtube', video_id=video_id,
1488 video_title=video_title)
1491 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1496 more = self._download_json(
1497 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1498 'Downloading page #%s' % (pagenum + 1),
1499 transform_source=uppercase_escape)
1500 content_html = more['content_html']
1501 more_widget_html = more['load_more_widget_html']
1503 return self.playlist_result(_entries(), channel_id)
1506 class YoutubeUserIE(YoutubeChannelIE):
1507 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1508 _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_-]+)'
1509 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1510 IE_NAME = 'youtube:user'
1513 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1514 'playlist_mincount': 320,
1516 'title': 'TheLinuxFoundation',
1519 'url': 'ytuser:phihag',
1520 'only_matching': True,
1524 def suitable(cls, url):
1525 # Don't return True if the url can be extracted with other youtube
1526 # extractor, the regex would is too permissive and it would match.
1527 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1528 if any(ie.suitable(url) for ie in other_ies):
1531 return super(YoutubeUserIE, cls).suitable(url)
1534 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1535 IE_DESC = 'YouTube.com searches'
1536 # there doesn't appear to be a real limit, for example if you search for
1537 # 'python' you get more than 8.000.000 results
1538 _MAX_RESULTS = float('inf')
1539 IE_NAME = 'youtube:search'
1540 _SEARCH_KEY = 'ytsearch'
1541 _EXTRA_QUERY_ARGS = {}
1544 def _get_n_results(self, query, n):
1545 """Get a specified number of results for a query"""
1550 for pagenum in itertools.count(1):
1552 'search_query': query.encode('utf-8'),
1556 url_query.update(self._EXTRA_QUERY_ARGS)
1557 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1558 data = self._download_json(
1559 result_url, video_id='query "%s"' % query,
1560 note='Downloading page %s' % pagenum,
1561 errnote='Unable to download API page')
1562 html_content = data[1]['body']['content']
1564 if 'class="search-message' in html_content:
1565 raise ExtractorError(
1566 '[youtube] No video results', expected=True)
1568 new_videos = self._ids_to_results(orderedSet(re.findall(
1569 r'href="/watch\?v=(.{11})', html_content)))
1570 videos += new_videos
1571 if not new_videos or len(videos) > limit:
1576 return self.playlist_result(videos, query)
1579 class YoutubeSearchDateIE(YoutubeSearchIE):
1580 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1581 _SEARCH_KEY = 'ytsearchdate'
1582 IE_DESC = 'YouTube.com searches, newest videos first'
1583 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1586 class YoutubeSearchURLIE(InfoExtractor):
1587 IE_DESC = 'YouTube.com search URLs'
1588 IE_NAME = 'youtube:search_url'
1589 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1591 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1592 'playlist_mincount': 5,
1594 'title': 'youtube-dl test video',
1598 def _real_extract(self, url):
1599 mobj = re.match(self._VALID_URL, url)
1600 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1602 webpage = self._download_webpage(url, query)
1603 result_code = self._search_regex(
1604 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1606 part_codes = re.findall(
1607 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1609 for part_code in part_codes:
1610 part_title = self._html_search_regex(
1611 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1612 part_url_snippet = self._html_search_regex(
1613 r'(?s)href="([^"]+)"', part_code, 'item URL')
1614 part_url = compat_urlparse.urljoin(
1615 'https://www.youtube.com/', part_url_snippet)
1619 'title': part_title,
1623 '_type': 'playlist',
1629 class YoutubeShowIE(InfoExtractor):
1630 IE_DESC = 'YouTube.com (multi-season) shows'
1631 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1632 IE_NAME = 'youtube:show'
1634 'url': 'http://www.youtube.com/show/airdisasters',
1635 'playlist_mincount': 3,
1637 'id': 'airdisasters',
1638 'title': 'Air Disasters',
1642 def _real_extract(self, url):
1643 mobj = re.match(self._VALID_URL, url)
1644 playlist_id = mobj.group('id')
1645 webpage = self._download_webpage(
1646 url, playlist_id, 'Downloading show webpage')
1647 # There's one playlist for each season of the show
1648 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1649 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1652 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1653 for season in m_seasons
1655 title = self._og_search_title(webpage, fatal=False)
1658 '_type': 'playlist',
1665 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1667 Base class for feed extractors
1668 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1670 _LOGIN_REQUIRED = True
1674 return 'youtube:%s' % self._FEED_NAME
1676 def _real_initialize(self):
1679 def _real_extract(self, url):
1680 page = self._download_webpage(
1681 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1683 # The extraction process is the same as for playlists, but the regex
1684 # for the video ids doesn't contain an index
1686 more_widget_html = content_html = page
1687 for page_num in itertools.count(1):
1688 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1690 # 'recommended' feed has infinite 'load more' and each new portion spins
1691 # the same videos in (sometimes) slightly different order, so we'll check
1692 # for unicity and break when portion has no new videos
1693 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1699 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1703 more = self._download_json(
1704 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1705 'Downloading page #%s' % page_num,
1706 transform_source=uppercase_escape)
1707 content_html = more['content_html']
1708 more_widget_html = more['load_more_widget_html']
1710 return self.playlist_result(
1711 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1714 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1715 IE_NAME = 'youtube:watchlater'
1716 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1717 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1719 _TESTS = [] # override PlaylistIE tests
1721 def _real_extract(self, url):
1722 return self._extract_playlist('WL')
1725 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1726 IE_NAME = 'youtube:favorites'
1727 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1728 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1729 _LOGIN_REQUIRED = True
1731 def _real_extract(self, url):
1732 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1733 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1734 return self.url_result(playlist_id, 'YoutubePlaylist')
1737 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1738 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1739 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1740 _FEED_NAME = 'recommended'
1741 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1744 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1745 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1746 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1747 _FEED_NAME = 'subscriptions'
1748 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1751 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1752 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1753 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1754 _FEED_NAME = 'history'
1755 _PLAYLIST_TITLE = 'Youtube History'
1758 class YoutubeTruncatedURLIE(InfoExtractor):
1759 IE_NAME = 'youtube:truncated_url'
1760 IE_DESC = False # Do not list
1761 _VALID_URL = r'''(?x)
1763 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1766 annotation_id=annotation_[^&]+|
1771 attribution_link\?a=[^&]+
1777 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1778 'only_matching': True,
1780 'url': 'http://www.youtube.com/watch?',
1781 'only_matching': True,
1783 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1784 'only_matching': True,
1786 'url': 'https://www.youtube.com/watch?feature=foo',
1787 'only_matching': True,
1789 'url': 'https://www.youtube.com/watch?hl=en-GB',
1790 'only_matching': True,
1793 def _real_extract(self, url):
1794 raise ExtractorError(
1795 'Did you forget to quote the URL? Remember that & is a meta '
1796 'character in most shells, so you want to put the URL in quotes, '
1798 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1799 ' or simply youtube-dl BaW_jenozKc .',
1803 class YoutubeTruncatedIDIE(InfoExtractor):
1804 IE_NAME = 'youtube:truncated_id'
1805 IE_DESC = False # Do not list
1806 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1809 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1810 'only_matching': True,
1813 def _real_extract(self, url):
1814 video_id = self._match_id(url)
1815 raise ExtractorError(
1816 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),