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',
525 def __init__(self, *args, **kwargs):
526 super(YoutubeIE, self).__init__(*args, **kwargs)
527 self._player_cache = {}
529 def report_video_info_webpage_download(self, video_id):
530 """Report attempt to download video info webpage."""
531 self.to_screen('%s: Downloading video info webpage' % video_id)
533 def report_information_extraction(self, video_id):
534 """Report attempt to extract video information."""
535 self.to_screen('%s: Extracting video information' % video_id)
537 def report_unavailable_format(self, video_id, format):
538 """Report extracted video URL."""
539 self.to_screen('%s: Format %s not available' % (video_id, format))
541 def report_rtmp_download(self):
542 """Indicate the download will use the RTMP protocol."""
543 self.to_screen('RTMP download detected')
545 def _signature_cache_id(self, example_sig):
546 """ Return a string representation of a signature """
547 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
549 def _extract_signature_function(self, video_id, player_url, example_sig):
551 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
554 raise ExtractorError('Cannot identify player %r' % player_url)
555 player_type = id_m.group('ext')
556 player_id = id_m.group('id')
558 # Read from filesystem cache
559 func_id = '%s_%s_%s' % (
560 player_type, player_id, self._signature_cache_id(example_sig))
561 assert os.path.basename(func_id) == func_id
563 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
564 if cache_spec is not None:
565 return lambda s: ''.join(s[i] for i in cache_spec)
568 'Downloading player %s' % player_url
569 if self._downloader.params.get('verbose') else
570 'Downloading %s player %s' % (player_type, player_id)
572 if player_type == 'js':
573 code = self._download_webpage(
574 player_url, video_id,
576 errnote='Download of %s failed' % player_url)
577 res = self._parse_sig_js(code)
578 elif player_type == 'swf':
579 urlh = self._request_webpage(
580 player_url, video_id,
582 errnote='Download of %s failed' % player_url)
584 res = self._parse_sig_swf(code)
586 assert False, 'Invalid player type %r' % player_type
588 test_string = ''.join(map(compat_chr, range(len(example_sig))))
589 cache_res = res(test_string)
590 cache_spec = [ord(c) for c in cache_res]
592 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
595 def _print_sig_code(self, func, example_sig):
596 def gen_sig_code(idxs):
597 def _genslice(start, end, step):
598 starts = '' if start == 0 else str(start)
599 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
600 steps = '' if step == 1 else (':%d' % step)
601 return 's[%s%s%s]' % (starts, ends, steps)
604 # Quelch pyflakes warnings - start will be set when step is set
605 start = '(Never used)'
606 for i, prev in zip(idxs[1:], idxs[:-1]):
610 yield _genslice(start, prev, step)
613 if i - prev in [-1, 1]:
622 yield _genslice(start, i, step)
624 test_string = ''.join(map(compat_chr, range(len(example_sig))))
625 cache_res = func(test_string)
626 cache_spec = [ord(c) for c in cache_res]
627 expr_code = ' + '.join(gen_sig_code(cache_spec))
628 signature_id_tuple = '(%s)' % (
629 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
630 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
631 ' return %s\n') % (signature_id_tuple, expr_code)
632 self.to_screen('Extracted signature function:\n' + code)
634 def _parse_sig_js(self, jscode):
635 funcname = self._search_regex(
636 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
637 'Initial JS player signature function name')
639 jsi = JSInterpreter(jscode)
640 initial_function = jsi.extract_function(funcname)
641 return lambda s: initial_function([s])
643 def _parse_sig_swf(self, file_contents):
644 swfi = SWFInterpreter(file_contents)
645 TARGET_CLASSNAME = 'SignatureDecipher'
646 searched_class = swfi.extract_class(TARGET_CLASSNAME)
647 initial_function = swfi.extract_function(searched_class, 'decipher')
648 return lambda s: initial_function([s])
650 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
651 """Turn the encrypted s field into a working signature"""
653 if player_url is None:
654 raise ExtractorError('Cannot decrypt signature without player_url')
656 if player_url.startswith('//'):
657 player_url = 'https:' + player_url
659 player_id = (player_url, self._signature_cache_id(s))
660 if player_id not in self._player_cache:
661 func = self._extract_signature_function(
662 video_id, player_url, s
664 self._player_cache[player_id] = func
665 func = self._player_cache[player_id]
666 if self._downloader.params.get('youtube_print_sig_code'):
667 self._print_sig_code(func, s)
669 except Exception as e:
670 tb = traceback.format_exc()
671 raise ExtractorError(
672 'Signature extraction failed: ' + tb, cause=e)
674 def _get_subtitles(self, video_id, webpage):
676 subs_doc = self._download_xml(
677 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
678 video_id, note=False)
679 except ExtractorError as err:
680 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
684 for track in subs_doc.findall('track'):
685 lang = track.attrib['lang_code']
686 if lang in sub_lang_list:
689 for ext in ['sbv', 'vtt', 'srt']:
690 params = compat_urllib_parse.urlencode({
694 'name': track.attrib['name'].encode('utf-8'),
697 'url': 'https://www.youtube.com/api/timedtext?' + params,
700 sub_lang_list[lang] = sub_formats
701 if not sub_lang_list:
702 self._downloader.report_warning('video doesn\'t have subtitles')
706 def _get_automatic_captions(self, video_id, webpage):
707 """We need the webpage for getting the captions url, pass it as an
708 argument to speed up the process."""
709 self.to_screen('%s: Looking for automatic captions' % video_id)
710 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
711 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
713 self._downloader.report_warning(err_msg)
715 player_config = json.loads(mobj.group(1))
717 args = player_config['args']
718 caption_url = args['ttsurl']
719 timestamp = args['timestamp']
720 # We get the available subtitles
721 list_params = compat_urllib_parse.urlencode({
726 list_url = caption_url + '&' + list_params
727 caption_list = self._download_xml(list_url, video_id)
728 original_lang_node = caption_list.find('track')
729 if original_lang_node is None:
730 self._downloader.report_warning('Video doesn\'t have automatic captions')
732 original_lang = original_lang_node.attrib['lang_code']
733 caption_kind = original_lang_node.attrib.get('kind', '')
736 for lang_node in caption_list.findall('target'):
737 sub_lang = lang_node.attrib['lang_code']
739 for ext in ['sbv', 'vtt', 'srt']:
740 params = compat_urllib_parse.urlencode({
741 'lang': original_lang,
745 'kind': caption_kind,
748 'url': caption_url + '&' + params,
751 sub_lang_list[sub_lang] = sub_formats
753 # An extractor error can be raise by the download process if there are
754 # no automatic captions but there are subtitles
755 except (KeyError, ExtractorError):
756 self._downloader.report_warning(err_msg)
760 def extract_id(cls, url):
761 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
763 raise ExtractorError('Invalid URL: %s' % url)
764 video_id = mobj.group(2)
767 def _extract_from_m3u8(self, manifest_url, video_id):
770 def _get_urls(_manifest):
771 lines = _manifest.split('\n')
772 urls = filter(lambda l: l and not l.startswith('#'),
775 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
776 formats_urls = _get_urls(manifest)
777 for format_url in formats_urls:
778 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
779 url_map[itag] = format_url
782 def _extract_annotations(self, video_id):
783 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
784 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
786 def _parse_dash_manifest(
787 self, video_id, dash_manifest_url, player_url, age_gate):
788 def decrypt_sig(mobj):
790 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
791 return '/signature/%s' % dec_s
792 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
793 dash_doc = self._download_xml(
794 dash_manifest_url, video_id,
795 note='Downloading DASH manifest',
796 errnote='Could not download DASH manifest')
799 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
800 mime_type = a.attrib.get('mimeType')
801 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
802 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
805 if mime_type == 'text/vtt':
806 # TODO implement WebVTT downloading
808 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
809 format_id = r.attrib['id']
810 video_url = url_el.text
811 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
813 'format_id': format_id,
815 'width': int_or_none(r.attrib.get('width')),
816 'height': int_or_none(r.attrib.get('height')),
817 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
818 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
819 'filesize': filesize,
820 'fps': int_or_none(r.attrib.get('frameRate')),
823 existing_format = next(
825 if fo['format_id'] == format_id)
826 except StopIteration:
827 full_info = self._formats.get(format_id, {}).copy()
829 formats.append(full_info)
831 existing_format.update(f)
833 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
836 def _real_extract(self, url):
838 'http' if self._downloader.params.get('prefer_insecure', False)
841 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
842 mobj = re.search(self._NEXT_URL_RE, url)
844 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
845 video_id = self.extract_id(url)
848 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
849 video_webpage = self._download_webpage(url, video_id)
851 # Attempt to extract SWF player URL
852 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
854 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
860 if re.search(r'player-age-gate-content">', video_webpage) is not None:
862 # We simulate the access to the video from www.youtube.com/v/{video_id}
863 # this can be viewed without login into Youtube
864 url = proto + '://www.youtube.com/embed/%s' % video_id
865 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
866 data = compat_urllib_parse.urlencode({
867 'video_id': video_id,
868 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
869 'sts': self._search_regex(
870 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
872 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
873 video_info_webpage = self._download_webpage(
874 video_info_url, video_id,
875 note='Refetching age-gated info webpage',
876 errnote='unable to download video info webpage')
877 video_info = compat_parse_qs(video_info_webpage)
881 # Try looking directly into the video webpage
882 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
884 raise ValueError('Could not find ytplayer.config') # caught below
885 json_code = uppercase_escape(mobj.group(1))
886 ytplayer_config = json.loads(json_code)
887 args = ytplayer_config['args']
888 # Convert to the same format returned by compat_parse_qs
889 video_info = dict((k, [v]) for k, v in args.items())
890 if not args.get('url_encoded_fmt_stream_map'):
891 raise ValueError('No stream_map present') # caught below
893 # We fallback to the get_video_info pages (used by the embed page)
894 self.report_video_info_webpage_download(video_id)
895 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
897 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
898 % (proto, video_id, el_type))
899 video_info_webpage = self._download_webpage(
901 video_id, note=False,
902 errnote='unable to download video info webpage')
903 video_info = compat_parse_qs(video_info_webpage)
904 if 'token' in video_info:
906 if 'token' not in video_info:
907 if 'reason' in video_info:
908 if 'The uploader has not made this video available in your country.' in video_info['reason']:
909 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
910 if regions_allowed is not None:
911 raise ExtractorError('YouTube said: This video is available in %s only' % (
912 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
914 raise ExtractorError(
915 'YouTube said: %s' % video_info['reason'][0],
916 expected=True, video_id=video_id)
918 raise ExtractorError(
919 '"token" parameter not in video info for unknown reason',
922 if 'view_count' in video_info:
923 view_count = int(video_info['view_count'][0])
927 # Check for "rental" videos
928 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
929 raise ExtractorError('"rental" videos not supported')
931 # Start extracting information
932 self.report_information_extraction(video_id)
935 if 'author' not in video_info:
936 raise ExtractorError('Unable to extract uploader name')
937 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
940 video_uploader_id = None
941 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
943 video_uploader_id = mobj.group(1)
945 self._downloader.report_warning('unable to extract uploader nickname')
948 if 'title' in video_info:
949 video_title = video_info['title'][0]
951 self._downloader.report_warning('Unable to extract video title')
955 # We try first to get a high quality image:
956 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
957 video_webpage, re.DOTALL)
958 if m_thumb is not None:
959 video_thumbnail = m_thumb.group(1)
960 elif 'thumbnail_url' not in video_info:
961 self._downloader.report_warning('unable to extract video thumbnail')
962 video_thumbnail = None
963 else: # don't panic if we can't find it
964 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
968 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
971 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
974 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
975 upload_date = unified_strdate(upload_date)
977 m_cat_container = self._search_regex(
978 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
979 video_webpage, 'categories', default=None)
981 category = self._html_search_regex(
982 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
984 video_categories = None if category is None else [category]
986 video_categories = None
989 video_description = get_element_by_id("eow-description", video_webpage)
990 if video_description:
991 video_description = re.sub(r'''(?x)
993 (?:[a-zA-Z-]+="[^"]+"\s+)*?
995 (?:[a-zA-Z-]+="[^"]+"\s+)*?
996 class="yt-uix-redirect-link"\s*>
999 ''', r'\1', video_description)
1000 video_description = clean_html(video_description)
1002 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1004 video_description = unescapeHTML(fd_mobj.group(1))
1006 video_description = ''
1008 def _extract_count(count_name):
1009 return str_to_int(self._search_regex(
1010 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1011 % re.escape(count_name),
1012 video_webpage, count_name, default=None))
1014 like_count = _extract_count('like')
1015 dislike_count = _extract_count('dislike')
1018 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1019 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1021 if 'length_seconds' not in video_info:
1022 self._downloader.report_warning('unable to extract video duration')
1023 video_duration = None
1025 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
1028 video_annotations = None
1029 if self._downloader.params.get('writeannotations', False):
1030 video_annotations = self._extract_annotations(video_id)
1032 def _map_to_format_list(urlmap):
1034 for itag, video_real_url in urlmap.items():
1037 'url': video_real_url,
1038 'player_url': player_url,
1040 if itag in self._formats:
1041 dct.update(self._formats[itag])
1045 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1046 self.report_rtmp_download()
1048 'format_id': '_rtmp',
1050 'url': video_info['conn'][0],
1051 'player_url': player_url,
1053 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1054 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1055 if 'rtmpe%3Dyes' in encoded_url_map:
1056 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1058 for url_data_str in encoded_url_map.split(','):
1059 url_data = compat_parse_qs(url_data_str)
1060 if 'itag' not in url_data or 'url' not in url_data:
1062 format_id = url_data['itag'][0]
1063 url = url_data['url'][0]
1065 if 'sig' in url_data:
1066 url += '&signature=' + url_data['sig'][0]
1067 elif 's' in url_data:
1068 encrypted_sig = url_data['s'][0]
1069 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1071 jsplayer_url_json = self._search_regex(
1073 embed_webpage if age_gate else video_webpage,
1074 'JS player URL (1)', default=None)
1075 if not jsplayer_url_json and not age_gate:
1076 # We need the embed website after all
1077 if embed_webpage is None:
1078 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1079 embed_webpage = self._download_webpage(
1080 embed_url, video_id, 'Downloading embed webpage')
1081 jsplayer_url_json = self._search_regex(
1082 ASSETS_RE, embed_webpage, 'JS player URL')
1084 player_url = json.loads(jsplayer_url_json)
1085 if player_url is None:
1086 player_url_json = self._search_regex(
1087 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1088 video_webpage, 'age gate player URL')
1089 player_url = json.loads(player_url_json)
1091 if self._downloader.params.get('verbose'):
1092 if player_url is None:
1093 player_version = 'unknown'
1094 player_desc = 'unknown'
1096 if player_url.endswith('swf'):
1097 player_version = self._search_regex(
1098 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1099 'flash player', fatal=False)
1100 player_desc = 'flash player %s' % player_version
1102 player_version = self._search_regex(
1103 r'html5player-([^/]+?)(?:/html5player)?\.js',
1105 'html5 player', fatal=False)
1106 player_desc = 'html5 player %s' % player_version
1108 parts_sizes = self._signature_cache_id(encrypted_sig)
1109 self.to_screen('{%s} signature length %s, %s' %
1110 (format_id, parts_sizes, player_desc))
1112 signature = self._decrypt_signature(
1113 encrypted_sig, video_id, player_url, age_gate)
1114 url += '&signature=' + signature
1115 if 'ratebypass' not in url:
1116 url += '&ratebypass=yes'
1117 url_map[format_id] = url
1118 formats = _map_to_format_list(url_map)
1119 elif video_info.get('hlsvp'):
1120 manifest_url = video_info['hlsvp'][0]
1121 url_map = self._extract_from_m3u8(manifest_url, video_id)
1122 formats = _map_to_format_list(url_map)
1124 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1126 # Look for the DASH manifest
1127 if self._downloader.params.get('youtube_include_dash_manifest', True):
1128 dash_mpd = video_info.get('dashmpd')
1130 dash_manifest_url = dash_mpd[0]
1132 dash_formats = self._parse_dash_manifest(
1133 video_id, dash_manifest_url, player_url, age_gate)
1134 except (ExtractorError, KeyError) as e:
1135 self.report_warning(
1136 'Skipping DASH manifest: %r' % e, video_id)
1138 # Remove the formats we found through non-DASH, they
1139 # contain less info and it can be wrong, because we use
1140 # fixed values (for example the resolution). See
1141 # https://github.com/rg3/youtube-dl/issues/5774 for an
1143 dash_keys = set(df['format_id'] for df in dash_formats)
1144 formats = [f for f in formats if f['format_id'] not in dash_keys]
1145 formats.extend(dash_formats)
1147 # Check for malformed aspect ratio
1148 stretched_m = re.search(
1149 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1152 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1154 if f.get('vcodec') != 'none':
1155 f['stretched_ratio'] = ratio
1157 self._sort_formats(formats)
1161 'uploader': video_uploader,
1162 'uploader_id': video_uploader_id,
1163 'upload_date': upload_date,
1164 'title': video_title,
1165 'thumbnail': video_thumbnail,
1166 'description': video_description,
1167 'categories': video_categories,
1168 'subtitles': video_subtitles,
1169 'automatic_captions': automatic_captions,
1170 'duration': video_duration,
1171 'age_limit': 18 if age_gate else 0,
1172 'annotations': video_annotations,
1173 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1174 'view_count': view_count,
1175 'like_count': like_count,
1176 'dislike_count': dislike_count,
1177 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1182 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1183 IE_DESC = 'YouTube.com playlists'
1184 _VALID_URL = r"""(?x)(?:
1189 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1190 \? (?:.*?&)*? (?:p|a|list)=
1194 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1195 # Top tracks, they can also include dots
1200 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1202 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1203 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1204 IE_NAME = 'youtube:playlist'
1206 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1208 'title': 'ytdl test PL',
1209 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1211 'playlist_count': 3,
1213 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1215 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1216 'title': 'YDL_Empty_List',
1218 'playlist_count': 0,
1220 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1221 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1223 'title': '29C3: Not my department',
1224 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1226 'playlist_count': 95,
1228 'note': 'issue #673',
1229 'url': 'PLBB231211A4F62143',
1231 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1232 'id': 'PLBB231211A4F62143',
1234 'playlist_mincount': 26,
1236 'note': 'Large playlist',
1237 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1239 'title': 'Uploads from Cauchemar',
1240 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1242 'playlist_mincount': 799,
1244 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1246 'title': 'YDL_safe_search',
1247 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1249 'playlist_count': 2,
1252 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1253 'playlist_count': 4,
1256 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1259 'note': 'Embedded SWF player',
1260 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1261 'playlist_count': 4,
1264 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1267 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1268 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1270 'title': 'Uploads from Interstellar Movie',
1271 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1273 'playlist_mincout': 21,
1276 def _real_initialize(self):
1279 def _extract_mix(self, playlist_id):
1280 # The mixes are generated from a single video
1281 # the id of the playlist is just 'RD' + video_id
1282 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1283 webpage = self._download_webpage(
1284 url, playlist_id, 'Downloading Youtube mix')
1285 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1287 search_title('playlist-title') or
1288 search_title('title long-title') or
1289 search_title('title'))
1290 title = clean_html(title_span)
1291 ids = orderedSet(re.findall(
1292 r'''(?xs)data-video-username=".*?".*?
1293 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1295 url_results = self._ids_to_results(ids)
1297 return self.playlist_result(url_results, playlist_id, title)
1299 def _extract_playlist(self, playlist_id):
1300 url = self._TEMPLATE_URL % playlist_id
1301 page = self._download_webpage(url, playlist_id)
1303 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1304 match = match.strip()
1305 # Check if the playlist exists or is private
1306 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1307 raise ExtractorError(
1308 'The playlist doesn\'t exist or is private, use --username or '
1309 '--netrc to access it.',
1311 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1312 raise ExtractorError(
1313 'Invalid parameters. Maybe URL is incorrect.',
1315 elif re.match(r'[^<]*Choose your language[^<]*', match):
1318 self.report_warning('Youtube gives an alert message: ' + match)
1320 # Extract the video ids from the playlist pages
1322 more_widget_html = content_html = page
1323 for page_num in itertools.count(1):
1324 matches = re.finditer(self._VIDEO_RE, content_html)
1325 # We remove the duplicates and the link with index 0
1326 # (it's not the first video of the playlist)
1327 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1328 for vid_id in new_ids:
1329 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1331 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1335 more = self._download_json(
1336 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1337 'Downloading page #%s' % page_num,
1338 transform_source=uppercase_escape)
1339 content_html = more['content_html']
1340 if not content_html.strip():
1341 # Some webpages show a "Load more" button but they don't
1344 more_widget_html = more['load_more_widget_html']
1346 playlist_title = self._html_search_regex(
1347 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1350 return self.playlist_result(_entries(), playlist_id, playlist_title)
1352 def _real_extract(self, url):
1353 # Extract playlist id
1354 mobj = re.match(self._VALID_URL, url)
1356 raise ExtractorError('Invalid URL: %s' % url)
1357 playlist_id = mobj.group(1) or mobj.group(2)
1359 # Check if it's a video-specific URL
1360 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1361 if 'v' in query_dict:
1362 video_id = query_dict['v'][0]
1363 if self._downloader.params.get('noplaylist'):
1364 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1365 return self.url_result(video_id, 'Youtube', video_id=video_id)
1367 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1369 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1370 # Mixes require a custom extraction process
1371 return self._extract_mix(playlist_id)
1373 return self._extract_playlist(playlist_id)
1376 class YoutubeChannelIE(InfoExtractor):
1377 IE_DESC = 'YouTube.com channels'
1378 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1379 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1380 IE_NAME = 'youtube:channel'
1382 'note': 'paginated channel',
1383 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1384 'playlist_mincount': 91,
1386 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1391 def extract_videos_from_page(page):
1394 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1395 video_id = mobj.group('id')
1396 video_title = unescapeHTML(mobj.group('title'))
1398 idx = ids_in_page.index(video_id)
1399 if video_title and not titles_in_page[idx]:
1400 titles_in_page[idx] = video_title
1402 ids_in_page.append(video_id)
1403 titles_in_page.append(video_title)
1404 return zip(ids_in_page, titles_in_page)
1406 def _real_extract(self, url):
1407 channel_id = self._match_id(url)
1409 url = self._TEMPLATE_URL % channel_id
1411 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1412 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1413 # otherwise fallback on channel by page extraction
1414 channel_page = self._download_webpage(
1415 url + '?view=57', channel_id,
1416 'Downloading channel page', fatal=False)
1417 channel_playlist_id = self._html_search_meta(
1418 'channelId', channel_page, 'channel id', default=None)
1419 if not channel_playlist_id:
1420 channel_playlist_id = self._search_regex(
1421 r'data-channel-external-id="([^"]+)"',
1422 channel_page, 'channel id', default=None)
1423 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1424 playlist_id = 'UU' + channel_playlist_id[2:]
1425 return self.url_result(
1426 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1428 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1429 autogenerated = re.search(r'''(?x)
1431 channel-header-autogenerated-label|
1432 yt-channel-title-autogenerated
1433 )[^"]*"''', channel_page) is not None
1436 # The videos are contained in a single page
1437 # the ajax pages can't be used, they are empty
1440 video_id, 'Youtube', video_id=video_id,
1441 video_title=video_title)
1442 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1443 return self.playlist_result(entries, channel_id)
1446 more_widget_html = content_html = channel_page
1447 for pagenum in itertools.count(1):
1449 for video_id, video_title in self.extract_videos_from_page(content_html):
1450 yield self.url_result(
1451 video_id, 'Youtube', video_id=video_id,
1452 video_title=video_title)
1455 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1460 more = self._download_json(
1461 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1462 'Downloading page #%s' % (pagenum + 1),
1463 transform_source=uppercase_escape)
1464 content_html = more['content_html']
1465 more_widget_html = more['load_more_widget_html']
1467 return self.playlist_result(_entries(), channel_id)
1470 class YoutubeUserIE(YoutubeChannelIE):
1471 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1472 _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_-]+)'
1473 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1474 IE_NAME = 'youtube:user'
1477 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1478 'playlist_mincount': 320,
1480 'title': 'TheLinuxFoundation',
1483 'url': 'ytuser:phihag',
1484 'only_matching': True,
1488 def suitable(cls, url):
1489 # Don't return True if the url can be extracted with other youtube
1490 # extractor, the regex would is too permissive and it would match.
1491 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1492 if any(ie.suitable(url) for ie in other_ies):
1495 return super(YoutubeUserIE, cls).suitable(url)
1498 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1499 IE_DESC = 'YouTube.com searches'
1500 # there doesn't appear to be a real limit, for example if you search for
1501 # 'python' you get more than 8.000.000 results
1502 _MAX_RESULTS = float('inf')
1503 IE_NAME = 'youtube:search'
1504 _SEARCH_KEY = 'ytsearch'
1505 _EXTRA_QUERY_ARGS = {}
1508 def _get_n_results(self, query, n):
1509 """Get a specified number of results for a query"""
1514 for pagenum in itertools.count(1):
1516 'search_query': query.encode('utf-8'),
1520 url_query.update(self._EXTRA_QUERY_ARGS)
1521 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1522 data = self._download_json(
1523 result_url, video_id='query "%s"' % query,
1524 note='Downloading page %s' % pagenum,
1525 errnote='Unable to download API page')
1526 html_content = data[1]['body']['content']
1528 if 'class="search-message' in html_content:
1529 raise ExtractorError(
1530 '[youtube] No video results', expected=True)
1532 new_videos = self._ids_to_results(orderedSet(re.findall(
1533 r'href="/watch\?v=(.{11})', html_content)))
1534 videos += new_videos
1535 if not new_videos or len(videos) > limit:
1540 return self.playlist_result(videos, query)
1543 class YoutubeSearchDateIE(YoutubeSearchIE):
1544 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1545 _SEARCH_KEY = 'ytsearchdate'
1546 IE_DESC = 'YouTube.com searches, newest videos first'
1547 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1550 class YoutubeSearchURLIE(InfoExtractor):
1551 IE_DESC = 'YouTube.com search URLs'
1552 IE_NAME = 'youtube:search_url'
1553 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1555 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1556 'playlist_mincount': 5,
1558 'title': 'youtube-dl test video',
1562 def _real_extract(self, url):
1563 mobj = re.match(self._VALID_URL, url)
1564 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1566 webpage = self._download_webpage(url, query)
1567 result_code = self._search_regex(
1568 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1570 part_codes = re.findall(
1571 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1573 for part_code in part_codes:
1574 part_title = self._html_search_regex(
1575 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1576 part_url_snippet = self._html_search_regex(
1577 r'(?s)href="([^"]+)"', part_code, 'item URL')
1578 part_url = compat_urlparse.urljoin(
1579 'https://www.youtube.com/', part_url_snippet)
1583 'title': part_title,
1587 '_type': 'playlist',
1593 class YoutubeShowIE(InfoExtractor):
1594 IE_DESC = 'YouTube.com (multi-season) shows'
1595 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1596 IE_NAME = 'youtube:show'
1598 'url': 'http://www.youtube.com/show/airdisasters',
1599 'playlist_mincount': 3,
1601 'id': 'airdisasters',
1602 'title': 'Air Disasters',
1606 def _real_extract(self, url):
1607 mobj = re.match(self._VALID_URL, url)
1608 playlist_id = mobj.group('id')
1609 webpage = self._download_webpage(
1610 url, playlist_id, 'Downloading show webpage')
1611 # There's one playlist for each season of the show
1612 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1613 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1616 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1617 for season in m_seasons
1619 title = self._og_search_title(webpage, fatal=False)
1622 '_type': 'playlist',
1629 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1631 Base class for feed extractors
1632 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1634 _LOGIN_REQUIRED = True
1638 return 'youtube:%s' % self._FEED_NAME
1640 def _real_initialize(self):
1643 def _real_extract(self, url):
1644 page = self._download_webpage(
1645 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1647 # The extraction process is the same as for playlists, but the regex
1648 # for the video ids doesn't contain an index
1650 more_widget_html = content_html = page
1651 for page_num in itertools.count(1):
1652 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1654 # 'recommended' feed has infinite 'load more' and each new portion spins
1655 # the same videos in (sometimes) slightly different order, so we'll check
1656 # for unicity and break when portion has no new videos
1657 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1663 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1667 more = self._download_json(
1668 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1669 'Downloading page #%s' % page_num,
1670 transform_source=uppercase_escape)
1671 content_html = more['content_html']
1672 more_widget_html = more['load_more_widget_html']
1674 return self.playlist_result(
1675 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1678 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1679 IE_NAME = 'youtube:watchlater'
1680 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1681 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1683 _TESTS = [] # override PlaylistIE tests
1685 def _real_extract(self, url):
1686 return self._extract_playlist('WL')
1689 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1690 IE_NAME = 'youtube:favorites'
1691 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1692 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1693 _LOGIN_REQUIRED = True
1695 def _real_extract(self, url):
1696 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1697 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1698 return self.url_result(playlist_id, 'YoutubePlaylist')
1701 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1702 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1703 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1704 _FEED_NAME = 'recommended'
1705 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1708 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1709 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1710 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1711 _FEED_NAME = 'subscriptions'
1712 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1715 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1716 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1717 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1718 _FEED_NAME = 'history'
1719 _PLAYLIST_TITLE = 'Youtube History'
1722 class YoutubeTruncatedURLIE(InfoExtractor):
1723 IE_NAME = 'youtube:truncated_url'
1724 IE_DESC = False # Do not list
1725 _VALID_URL = r'''(?x)
1727 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1730 annotation_id=annotation_[^&]+|
1735 attribution_link\?a=[^&]+
1741 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1742 'only_matching': True,
1744 'url': 'http://www.youtube.com/watch?',
1745 'only_matching': True,
1747 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1748 'only_matching': True,
1750 'url': 'https://www.youtube.com/watch?feature=foo',
1751 'only_matching': True,
1753 'url': 'https://www.youtube.com/watch?hl=en-GB',
1754 'only_matching': True,
1757 def _real_extract(self, url):
1758 raise ExtractorError(
1759 'Did you forget to quote the URL? Remember that & is a meta '
1760 'character in most shells, so you want to put the URL in quotes, '
1762 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1763 ' or simply youtube-dl BaW_jenozKc .',
1767 class YoutubeTruncatedIDIE(InfoExtractor):
1768 IE_NAME = 'youtube:truncated_id'
1769 IE_DESC = False # Do not list
1770 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1773 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1774 'only_matching': True,
1777 def _real_extract(self, url):
1778 video_id = self._match_id(url)
1779 raise ExtractorError(
1780 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),