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,
38 class YoutubeBaseInfoExtractor(InfoExtractor):
39 """Provide base functions for Youtube extractors"""
40 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
41 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
42 _NETRC_MACHINE = 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED = False
46 def _set_language(self):
48 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
49 # YouTube sets the expire time to about two months
50 expire_time=time.time() + 2 * 30 * 24 * 3600)
52 def _ids_to_results(self, ids):
54 self.url_result(vid_id, 'Youtube', video_id=vid_id)
59 Attempt to log in to YouTube.
60 True is returned if successful or skipped.
61 False is returned if login failed.
63 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
65 (username, password) = self._get_login_info()
66 # No authentication to be performed
68 if self._LOGIN_REQUIRED:
69 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
72 login_page = self._download_webpage(
73 self._LOGIN_URL, None,
74 note='Downloading login page',
75 errnote='unable to fetch login page', fatal=False)
76 if login_page is False:
79 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
80 login_page, 'Login GALX parameter')
84 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
89 'PersistentCookie': 'yes',
91 'bgresponse': 'js_disabled',
92 'checkConnection': '',
93 'checkedDomains': 'youtube',
100 'service': 'youtube',
105 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
107 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
108 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
110 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
111 login_results = self._download_webpage(
113 note='Logging in', errnote='unable to log in', fatal=False)
114 if login_results is False:
117 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
118 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
121 # TODO add SMS and phone call support - these require making a request and then prompting the user
123 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
124 tfa_code = self._get_tfa_info()
127 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
128 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
131 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
133 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
135 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
136 secTok = match.group(1)
137 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
139 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
140 timeStmp = match.group(1)
143 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
145 'smsUserPin': tfa_code,
146 'smsVerifyPin': 'Verify',
148 'PersistentCookie': 'yes',
149 'checkConnection': '',
150 'checkedDomains': 'youtube',
153 'timeStmp': timeStmp,
154 'service': 'youtube',
157 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
158 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
160 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
161 tfa_results = self._download_webpage(
163 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
165 if tfa_results is False:
168 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
169 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
171 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
172 self._downloader.report_warning('unable to log in - did the page structure change?')
174 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
175 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
178 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
179 self._downloader.report_warning('unable to log in: bad username or password')
183 def _real_initialize(self):
184 if self._downloader is None:
187 if not self._login():
191 class YoutubeIE(YoutubeBaseInfoExtractor):
192 IE_DESC = 'YouTube.com'
193 _VALID_URL = r"""(?x)^
195 (?:https?://|//) # http(s):// or protocol-independent URL
196 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
197 (?:www\.)?deturl\.com/www\.youtube\.com/|
198 (?:www\.)?pwnyoutube\.com/|
199 (?:www\.)?yourepeat\.com/|
200 tube\.majestyc\.net/|
201 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
202 (?:.*?\#/)? # handle anchor (#/) redirect urls
203 (?: # the various things that can precede the ID:
204 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
205 |(?: # or the v= param in all its forms
206 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
207 (?:\?|\#!?) # the params delimiter ? or # or #!
208 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
212 |youtu\.be/ # just youtu.be/xxxx
213 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
215 )? # all until now is optional -> you can pass the naked ID
216 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
217 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
218 (?(1).+)? # if we found the ID, everything can follow
220 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
222 '5': {'ext': 'flv', 'width': 400, 'height': 240},
223 '6': {'ext': 'flv', 'width': 450, 'height': 270},
224 '13': {'ext': '3gp'},
225 '17': {'ext': '3gp', 'width': 176, 'height': 144},
226 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
227 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
228 '34': {'ext': 'flv', 'width': 640, 'height': 360},
229 '35': {'ext': 'flv', 'width': 854, 'height': 480},
230 '36': {'ext': '3gp', 'width': 320, 'height': 240},
231 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
232 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
233 '43': {'ext': 'webm', 'width': 640, 'height': 360},
234 '44': {'ext': 'webm', 'width': 854, 'height': 480},
235 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
236 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
240 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
241 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
242 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
243 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
244 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
245 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
246 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
248 # Apple HTTP Live Streaming
249 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
250 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
251 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
252 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
253 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
254 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
255 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
258 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
263 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
264 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
265 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
266 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
267 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
268 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
271 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
272 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
273 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
276 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
278 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
279 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
280 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
281 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
282 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
283 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
289 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
293 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
294 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
295 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
296 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
299 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
300 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
302 # Dash webm audio with opus inside
303 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
304 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
305 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
308 '_rtmp': {'protocol': 'rtmp'},
314 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
318 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
319 'uploader': 'Philipp Hagemeister',
320 'uploader_id': 'phihag',
321 'upload_date': '20121002',
322 '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 .',
323 'categories': ['Science & Technology'],
325 'dislike_count': int,
329 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
330 'note': 'Test generic use_cipher_signature video (#897)',
334 'upload_date': '20120506',
335 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
336 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
337 'uploader': 'Icona Pop',
338 'uploader_id': 'IconaPop',
342 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
343 'note': 'Test VEVO video with age protection (#956)',
347 'upload_date': '20130703',
348 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
349 'description': 'md5:64249768eec3bc4276236606ea996373',
350 'uploader': 'justintimberlakeVEVO',
351 'uploader_id': 'justintimberlakeVEVO',
355 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
356 'note': 'Embed-only video (#1746)',
360 'upload_date': '20120608',
361 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
362 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
363 'uploader': 'SET India',
364 'uploader_id': 'setindia'
368 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
369 'note': '256k DASH audio (format 141) via DASH manifest',
373 'upload_date': '20121002',
374 'uploader_id': '8KVIDEO',
376 'uploader': '8KVIDEO',
377 'title': 'UHDTV TEST 8K VIDEO.mp4'
380 'youtube_include_dash_manifest': True,
384 # DASH manifest with encrypted signature
386 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
390 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
391 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
392 'uploader': 'AfrojackVEVO',
393 'uploader_id': 'AfrojackVEVO',
394 'upload_date': '20131011',
397 'youtube_include_dash_manifest': True,
401 # JS player signature function name containing $
403 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
407 'title': 'Taylor Swift - Shake It Off',
408 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
409 'uploader': 'TaylorSwiftVEVO',
410 'uploader_id': 'TaylorSwiftVEVO',
411 'upload_date': '20140818',
414 'youtube_include_dash_manifest': True,
420 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
424 'upload_date': '20100909',
425 'uploader': 'The Amazing Atheist',
426 'uploader_id': 'TheAmazingAtheist',
427 'title': 'Burning Everyone\'s Koran',
428 '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',
431 # Normal age-gate video (No vevo, embed allowed)
433 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
437 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
438 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
439 'uploader': 'The Witcher',
440 'uploader_id': 'WitcherGame',
441 'upload_date': '20140605',
444 # Age-gate video with encrypted signature
446 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
450 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
451 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
452 'uploader': 'LloydVEVO',
453 'uploader_id': 'LloydVEVO',
454 'upload_date': '20110629',
457 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
459 'url': '__2ABJjxzNo',
463 'upload_date': '20100430',
464 'uploader_id': 'deadmau5',
465 'description': 'md5:12c56784b8032162bb936a5f76d55360',
466 'uploader': 'deadmau5',
467 'title': 'Deadmau5 - Some Chords (HD)',
469 'expected_warnings': [
470 'DASH manifest missing',
473 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
475 'url': 'lqQg6PlCWgI',
479 'upload_date': '20120731',
480 'uploader_id': 'olympic',
481 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
482 'uploader': 'Olympics',
483 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
486 'skip_download': 'requires avconv',
491 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
495 'stretched_ratio': 16 / 9.,
496 'upload_date': '20110310',
497 'uploader_id': 'AllenMeow',
498 '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',
500 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
503 # url_encoded_fmt_stream_map is empty string
505 'url': 'qEJwOuvDf7I',
509 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
511 'upload_date': '20150404',
512 'uploader_id': 'spbelect',
513 'uploader': 'Наблюдатели Петербурга',
516 'skip_download': 'requires avconv',
521 def __init__(self, *args, **kwargs):
522 super(YoutubeIE, self).__init__(*args, **kwargs)
523 self._player_cache = {}
525 def report_video_info_webpage_download(self, video_id):
526 """Report attempt to download video info webpage."""
527 self.to_screen('%s: Downloading video info webpage' % video_id)
529 def report_information_extraction(self, video_id):
530 """Report attempt to extract video information."""
531 self.to_screen('%s: Extracting video information' % video_id)
533 def report_unavailable_format(self, video_id, format):
534 """Report extracted video URL."""
535 self.to_screen('%s: Format %s not available' % (video_id, format))
537 def report_rtmp_download(self):
538 """Indicate the download will use the RTMP protocol."""
539 self.to_screen('RTMP download detected')
541 def _signature_cache_id(self, example_sig):
542 """ Return a string representation of a signature """
543 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
545 def _extract_signature_function(self, video_id, player_url, example_sig):
547 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
550 raise ExtractorError('Cannot identify player %r' % player_url)
551 player_type = id_m.group('ext')
552 player_id = id_m.group('id')
554 # Read from filesystem cache
555 func_id = '%s_%s_%s' % (
556 player_type, player_id, self._signature_cache_id(example_sig))
557 assert os.path.basename(func_id) == func_id
559 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
560 if cache_spec is not None:
561 return lambda s: ''.join(s[i] for i in cache_spec)
564 'Downloading player %s' % player_url
565 if self._downloader.params.get('verbose') else
566 'Downloading %s player %s' % (player_type, player_id)
568 if player_type == 'js':
569 code = self._download_webpage(
570 player_url, video_id,
572 errnote='Download of %s failed' % player_url)
573 res = self._parse_sig_js(code)
574 elif player_type == 'swf':
575 urlh = self._request_webpage(
576 player_url, video_id,
578 errnote='Download of %s failed' % player_url)
580 res = self._parse_sig_swf(code)
582 assert False, 'Invalid player type %r' % player_type
584 test_string = ''.join(map(compat_chr, range(len(example_sig))))
585 cache_res = res(test_string)
586 cache_spec = [ord(c) for c in cache_res]
588 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
591 def _print_sig_code(self, func, example_sig):
592 def gen_sig_code(idxs):
593 def _genslice(start, end, step):
594 starts = '' if start == 0 else str(start)
595 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
596 steps = '' if step == 1 else (':%d' % step)
597 return 's[%s%s%s]' % (starts, ends, steps)
600 # Quelch pyflakes warnings - start will be set when step is set
601 start = '(Never used)'
602 for i, prev in zip(idxs[1:], idxs[:-1]):
606 yield _genslice(start, prev, step)
609 if i - prev in [-1, 1]:
618 yield _genslice(start, i, step)
620 test_string = ''.join(map(compat_chr, range(len(example_sig))))
621 cache_res = func(test_string)
622 cache_spec = [ord(c) for c in cache_res]
623 expr_code = ' + '.join(gen_sig_code(cache_spec))
624 signature_id_tuple = '(%s)' % (
625 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
626 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
627 ' return %s\n') % (signature_id_tuple, expr_code)
628 self.to_screen('Extracted signature function:\n' + code)
630 def _parse_sig_js(self, jscode):
631 funcname = self._search_regex(
632 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
633 'Initial JS player signature function name')
635 jsi = JSInterpreter(jscode)
636 initial_function = jsi.extract_function(funcname)
637 return lambda s: initial_function([s])
639 def _parse_sig_swf(self, file_contents):
640 swfi = SWFInterpreter(file_contents)
641 TARGET_CLASSNAME = 'SignatureDecipher'
642 searched_class = swfi.extract_class(TARGET_CLASSNAME)
643 initial_function = swfi.extract_function(searched_class, 'decipher')
644 return lambda s: initial_function([s])
646 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
647 """Turn the encrypted s field into a working signature"""
649 if player_url is None:
650 raise ExtractorError('Cannot decrypt signature without player_url')
652 if player_url.startswith('//'):
653 player_url = 'https:' + player_url
655 player_id = (player_url, self._signature_cache_id(s))
656 if player_id not in self._player_cache:
657 func = self._extract_signature_function(
658 video_id, player_url, s
660 self._player_cache[player_id] = func
661 func = self._player_cache[player_id]
662 if self._downloader.params.get('youtube_print_sig_code'):
663 self._print_sig_code(func, s)
665 except Exception as e:
666 tb = traceback.format_exc()
667 raise ExtractorError(
668 'Signature extraction failed: ' + tb, cause=e)
670 def _get_subtitles(self, video_id, webpage):
672 subs_doc = self._download_xml(
673 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
674 video_id, note=False)
675 except ExtractorError as err:
676 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
680 for track in subs_doc.findall('track'):
681 lang = track.attrib['lang_code']
682 if lang in sub_lang_list:
685 for ext in ['sbv', 'vtt', 'srt']:
686 params = compat_urllib_parse.urlencode({
690 'name': track.attrib['name'].encode('utf-8'),
693 'url': 'https://www.youtube.com/api/timedtext?' + params,
696 sub_lang_list[lang] = sub_formats
697 if not sub_lang_list:
698 self._downloader.report_warning('video doesn\'t have subtitles')
702 def _get_automatic_captions(self, video_id, webpage):
703 """We need the webpage for getting the captions url, pass it as an
704 argument to speed up the process."""
705 self.to_screen('%s: Looking for automatic captions' % video_id)
706 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
707 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
709 self._downloader.report_warning(err_msg)
711 player_config = json.loads(mobj.group(1))
713 args = player_config['args']
714 caption_url = args['ttsurl']
715 timestamp = args['timestamp']
716 # We get the available subtitles
717 list_params = compat_urllib_parse.urlencode({
722 list_url = caption_url + '&' + list_params
723 caption_list = self._download_xml(list_url, video_id)
724 original_lang_node = caption_list.find('track')
725 if original_lang_node is None:
726 self._downloader.report_warning('Video doesn\'t have automatic captions')
728 original_lang = original_lang_node.attrib['lang_code']
729 caption_kind = original_lang_node.attrib.get('kind', '')
732 for lang_node in caption_list.findall('target'):
733 sub_lang = lang_node.attrib['lang_code']
735 for ext in ['sbv', 'vtt', 'srt']:
736 params = compat_urllib_parse.urlencode({
737 'lang': original_lang,
741 'kind': caption_kind,
744 'url': caption_url + '&' + params,
747 sub_lang_list[sub_lang] = sub_formats
749 # An extractor error can be raise by the download process if there are
750 # no automatic captions but there are subtitles
751 except (KeyError, ExtractorError):
752 self._downloader.report_warning(err_msg)
756 def extract_id(cls, url):
757 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
759 raise ExtractorError('Invalid URL: %s' % url)
760 video_id = mobj.group(2)
763 def _extract_from_m3u8(self, manifest_url, video_id):
766 def _get_urls(_manifest):
767 lines = _manifest.split('\n')
768 urls = filter(lambda l: l and not l.startswith('#'),
771 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
772 formats_urls = _get_urls(manifest)
773 for format_url in formats_urls:
774 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
775 url_map[itag] = format_url
778 def _extract_annotations(self, video_id):
779 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
780 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
782 def _parse_dash_manifest(
783 self, video_id, dash_manifest_url, player_url, age_gate):
784 def decrypt_sig(mobj):
786 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
787 return '/signature/%s' % dec_s
788 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
789 dash_doc = self._download_xml(
790 dash_manifest_url, video_id,
791 note='Downloading DASH manifest',
792 errnote='Could not download DASH manifest')
795 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
796 mime_type = a.attrib.get('mimeType')
797 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
798 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
801 if mime_type == 'text/vtt':
802 # TODO implement WebVTT downloading
804 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
805 format_id = r.attrib['id']
806 video_url = url_el.text
807 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
809 'format_id': format_id,
811 'width': int_or_none(r.attrib.get('width')),
812 'height': int_or_none(r.attrib.get('height')),
813 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
814 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
815 'filesize': filesize,
816 'fps': int_or_none(r.attrib.get('frameRate')),
819 existing_format = next(
821 if fo['format_id'] == format_id)
822 except StopIteration:
823 full_info = self._formats.get(format_id, {}).copy()
825 formats.append(full_info)
827 existing_format.update(f)
829 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
832 def _real_extract(self, url):
834 'http' if self._downloader.params.get('prefer_insecure', False)
837 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
838 mobj = re.search(self._NEXT_URL_RE, url)
840 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
841 video_id = self.extract_id(url)
844 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
845 video_webpage = self._download_webpage(url, video_id)
847 # Attempt to extract SWF player URL
848 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
850 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
856 if re.search(r'player-age-gate-content">', video_webpage) is not None:
858 # We simulate the access to the video from www.youtube.com/v/{video_id}
859 # this can be viewed without login into Youtube
860 url = proto + '://www.youtube.com/embed/%s' % video_id
861 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
862 data = compat_urllib_parse.urlencode({
863 'video_id': video_id,
864 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
865 'sts': self._search_regex(
866 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
868 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
869 video_info_webpage = self._download_webpage(
870 video_info_url, video_id,
871 note='Refetching age-gated info webpage',
872 errnote='unable to download video info webpage')
873 video_info = compat_parse_qs(video_info_webpage)
877 # Try looking directly into the video webpage
878 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
880 raise ValueError('Could not find ytplayer.config') # caught below
881 json_code = uppercase_escape(mobj.group(1))
882 ytplayer_config = json.loads(json_code)
883 args = ytplayer_config['args']
884 # Convert to the same format returned by compat_parse_qs
885 video_info = dict((k, [v]) for k, v in args.items())
886 if not args.get('url_encoded_fmt_stream_map'):
887 raise ValueError('No stream_map present') # caught below
889 # We fallback to the get_video_info pages (used by the embed page)
890 self.report_video_info_webpage_download(video_id)
891 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
893 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
894 % (proto, video_id, el_type))
895 video_info_webpage = self._download_webpage(
897 video_id, note=False,
898 errnote='unable to download video info webpage')
899 video_info = compat_parse_qs(video_info_webpage)
900 if 'token' in video_info:
902 if 'token' not in video_info:
903 if 'reason' in video_info:
904 raise ExtractorError(
905 'YouTube said: %s' % video_info['reason'][0],
906 expected=True, video_id=video_id)
908 raise ExtractorError(
909 '"token" parameter not in video info for unknown reason',
912 if 'view_count' in video_info:
913 view_count = int(video_info['view_count'][0])
917 # Check for "rental" videos
918 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
919 raise ExtractorError('"rental" videos not supported')
921 # Start extracting information
922 self.report_information_extraction(video_id)
925 if 'author' not in video_info:
926 raise ExtractorError('Unable to extract uploader name')
927 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
930 video_uploader_id = None
931 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
933 video_uploader_id = mobj.group(1)
935 self._downloader.report_warning('unable to extract uploader nickname')
938 if 'title' in video_info:
939 video_title = video_info['title'][0]
941 self._downloader.report_warning('Unable to extract video title')
945 # We try first to get a high quality image:
946 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
947 video_webpage, re.DOTALL)
948 if m_thumb is not None:
949 video_thumbnail = m_thumb.group(1)
950 elif 'thumbnail_url' not in video_info:
951 self._downloader.report_warning('unable to extract video thumbnail')
952 video_thumbnail = None
953 else: # don't panic if we can't find it
954 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
958 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
961 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
964 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
965 upload_date = unified_strdate(upload_date)
967 m_cat_container = self._search_regex(
968 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
969 video_webpage, 'categories', default=None)
971 category = self._html_search_regex(
972 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
974 video_categories = None if category is None else [category]
976 video_categories = None
979 video_description = get_element_by_id("eow-description", video_webpage)
980 if video_description:
981 video_description = re.sub(r'''(?x)
983 (?:[a-zA-Z-]+="[^"]+"\s+)*?
985 (?:[a-zA-Z-]+="[^"]+"\s+)*?
986 class="yt-uix-redirect-link"\s*>
989 ''', r'\1', video_description)
990 video_description = clean_html(video_description)
992 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
994 video_description = unescapeHTML(fd_mobj.group(1))
996 video_description = ''
998 def _extract_count(count_name):
999 count = self._search_regex(
1000 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
1001 video_webpage, count_name, default=None)
1002 if count is not None:
1003 return int(count.replace(',', ''))
1005 like_count = _extract_count('like')
1006 dislike_count = _extract_count('dislike')
1009 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1010 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1012 if 'length_seconds' not in video_info:
1013 self._downloader.report_warning('unable to extract video duration')
1014 video_duration = None
1016 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
1019 video_annotations = None
1020 if self._downloader.params.get('writeannotations', False):
1021 video_annotations = self._extract_annotations(video_id)
1023 def _map_to_format_list(urlmap):
1025 for itag, video_real_url in urlmap.items():
1028 'url': video_real_url,
1029 'player_url': player_url,
1031 if itag in self._formats:
1032 dct.update(self._formats[itag])
1036 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1037 self.report_rtmp_download()
1039 'format_id': '_rtmp',
1041 'url': video_info['conn'][0],
1042 'player_url': player_url,
1044 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1045 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1046 if 'rtmpe%3Dyes' in encoded_url_map:
1047 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1049 for url_data_str in encoded_url_map.split(','):
1050 url_data = compat_parse_qs(url_data_str)
1051 if 'itag' not in url_data or 'url' not in url_data:
1053 format_id = url_data['itag'][0]
1054 url = url_data['url'][0]
1056 if 'sig' in url_data:
1057 url += '&signature=' + url_data['sig'][0]
1058 elif 's' in url_data:
1059 encrypted_sig = url_data['s'][0]
1060 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1062 jsplayer_url_json = self._search_regex(
1064 embed_webpage if age_gate else video_webpage,
1065 'JS player URL (1)', default=None)
1066 if not jsplayer_url_json and not age_gate:
1067 # We need the embed website after all
1068 if embed_webpage is None:
1069 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1070 embed_webpage = self._download_webpage(
1071 embed_url, video_id, 'Downloading embed webpage')
1072 jsplayer_url_json = self._search_regex(
1073 ASSETS_RE, embed_webpage, 'JS player URL')
1075 player_url = json.loads(jsplayer_url_json)
1076 if player_url is None:
1077 player_url_json = self._search_regex(
1078 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1079 video_webpage, 'age gate player URL')
1080 player_url = json.loads(player_url_json)
1082 if self._downloader.params.get('verbose'):
1083 if player_url is None:
1084 player_version = 'unknown'
1085 player_desc = 'unknown'
1087 if player_url.endswith('swf'):
1088 player_version = self._search_regex(
1089 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1090 'flash player', fatal=False)
1091 player_desc = 'flash player %s' % player_version
1093 player_version = self._search_regex(
1094 r'html5player-([^/]+?)(?:/html5player)?\.js',
1096 'html5 player', fatal=False)
1097 player_desc = 'html5 player %s' % player_version
1099 parts_sizes = self._signature_cache_id(encrypted_sig)
1100 self.to_screen('{%s} signature length %s, %s' %
1101 (format_id, parts_sizes, player_desc))
1103 signature = self._decrypt_signature(
1104 encrypted_sig, video_id, player_url, age_gate)
1105 url += '&signature=' + signature
1106 if 'ratebypass' not in url:
1107 url += '&ratebypass=yes'
1108 url_map[format_id] = url
1109 formats = _map_to_format_list(url_map)
1110 elif video_info.get('hlsvp'):
1111 manifest_url = video_info['hlsvp'][0]
1112 url_map = self._extract_from_m3u8(manifest_url, video_id)
1113 formats = _map_to_format_list(url_map)
1115 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1117 # Look for the DASH manifest
1118 if self._downloader.params.get('youtube_include_dash_manifest', True):
1119 dash_mpd = video_info.get('dashmpd')
1121 dash_manifest_url = dash_mpd[0]
1123 dash_formats = self._parse_dash_manifest(
1124 video_id, dash_manifest_url, player_url, age_gate)
1125 except (ExtractorError, KeyError) as e:
1126 self.report_warning(
1127 'Skipping DASH manifest: %r' % e, video_id)
1129 # Hide the formats we found through non-DASH
1130 dash_keys = set(df['format_id'] for df in dash_formats)
1132 if f['format_id'] in dash_keys:
1133 f['format_id'] = 'nondash-%s' % f['format_id']
1134 f['preference'] = f.get('preference', 0) - 10000
1135 formats.extend(dash_formats)
1137 # Check for malformed aspect ratio
1138 stretched_m = re.search(
1139 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1142 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1144 if f.get('vcodec') != 'none':
1145 f['stretched_ratio'] = ratio
1147 self._sort_formats(formats)
1151 'uploader': video_uploader,
1152 'uploader_id': video_uploader_id,
1153 'upload_date': upload_date,
1154 'title': video_title,
1155 'thumbnail': video_thumbnail,
1156 'description': video_description,
1157 'categories': video_categories,
1158 'subtitles': video_subtitles,
1159 'automatic_captions': automatic_captions,
1160 'duration': video_duration,
1161 'age_limit': 18 if age_gate else 0,
1162 'annotations': video_annotations,
1163 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1164 'view_count': view_count,
1165 'like_count': like_count,
1166 'dislike_count': dislike_count,
1167 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1172 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1173 IE_DESC = 'YouTube.com playlists'
1174 _VALID_URL = r"""(?x)(?:
1179 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1180 \? (?:.*?&)*? (?:p|a|list)=
1184 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1185 # Top tracks, they can also include dots
1190 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1192 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1193 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1194 IE_NAME = 'youtube:playlist'
1196 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1198 'title': 'ytdl test PL',
1199 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1201 'playlist_count': 3,
1203 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1205 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1206 'title': 'YDL_Empty_List',
1208 'playlist_count': 0,
1210 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1211 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1213 'title': '29C3: Not my department',
1214 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1216 'playlist_count': 95,
1218 'note': 'issue #673',
1219 'url': 'PLBB231211A4F62143',
1221 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1222 'id': 'PLBB231211A4F62143',
1224 'playlist_mincount': 26,
1226 'note': 'Large playlist',
1227 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1229 'title': 'Uploads from Cauchemar',
1230 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1232 'playlist_mincount': 799,
1234 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1236 'title': 'YDL_safe_search',
1237 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1239 'playlist_count': 2,
1242 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1243 'playlist_count': 4,
1246 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1249 'note': 'Embedded SWF player',
1250 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1251 'playlist_count': 4,
1254 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1257 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1258 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1260 'title': 'Uploads from Interstellar Movie',
1261 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1263 'playlist_mincout': 21,
1266 def _real_initialize(self):
1269 def _extract_mix(self, playlist_id):
1270 # The mixes are generated from a single video
1271 # the id of the playlist is just 'RD' + video_id
1272 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1273 webpage = self._download_webpage(
1274 url, playlist_id, 'Downloading Youtube mix')
1275 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1277 search_title('playlist-title') or
1278 search_title('title long-title') or
1279 search_title('title'))
1280 title = clean_html(title_span)
1281 ids = orderedSet(re.findall(
1282 r'''(?xs)data-video-username=".*?".*?
1283 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1285 url_results = self._ids_to_results(ids)
1287 return self.playlist_result(url_results, playlist_id, title)
1289 def _extract_playlist(self, playlist_id):
1290 url = self._TEMPLATE_URL % playlist_id
1291 page = self._download_webpage(url, playlist_id)
1292 more_widget_html = content_html = page
1294 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1295 match = match.strip()
1296 # Check if the playlist exists or is private
1297 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1298 raise ExtractorError(
1299 'The playlist doesn\'t exist or is private, use --username or '
1300 '--netrc to access it.',
1302 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1303 raise ExtractorError(
1304 'Invalid parameters. Maybe URL is incorrect.',
1306 elif re.match(r'[^<]*Choose your language[^<]*', match):
1309 self.report_warning('Youtube gives an alert message: ' + match)
1311 # Extract the video ids from the playlist pages
1314 for page_num in itertools.count(1):
1315 matches = re.finditer(self._VIDEO_RE, content_html)
1316 # We remove the duplicates and the link with index 0
1317 # (it's not the first video of the playlist)
1318 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1321 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1325 more = self._download_json(
1326 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1327 'Downloading page #%s' % page_num,
1328 transform_source=uppercase_escape)
1329 content_html = more['content_html']
1330 if not content_html.strip():
1331 # Some webpages show a "Load more" button but they don't
1334 more_widget_html = more['load_more_widget_html']
1336 playlist_title = self._html_search_regex(
1337 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1340 url_results = self._ids_to_results(ids)
1341 return self.playlist_result(url_results, playlist_id, playlist_title)
1343 def _real_extract(self, url):
1344 # Extract playlist id
1345 mobj = re.match(self._VALID_URL, url)
1347 raise ExtractorError('Invalid URL: %s' % url)
1348 playlist_id = mobj.group(1) or mobj.group(2)
1350 # Check if it's a video-specific URL
1351 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1352 if 'v' in query_dict:
1353 video_id = query_dict['v'][0]
1354 if self._downloader.params.get('noplaylist'):
1355 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1356 return self.url_result(video_id, 'Youtube', video_id=video_id)
1358 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1360 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1361 # Mixes require a custom extraction process
1362 return self._extract_mix(playlist_id)
1364 return self._extract_playlist(playlist_id)
1367 class YoutubeChannelIE(InfoExtractor):
1368 IE_DESC = 'YouTube.com channels'
1369 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1370 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1371 IE_NAME = 'youtube:channel'
1373 'note': 'paginated channel',
1374 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1375 'playlist_mincount': 91,
1377 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1382 def extract_videos_from_page(page):
1385 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1386 video_id = mobj.group('id')
1387 video_title = unescapeHTML(mobj.group('title'))
1389 idx = ids_in_page.index(video_id)
1390 if video_title and not titles_in_page[idx]:
1391 titles_in_page[idx] = video_title
1393 ids_in_page.append(video_id)
1394 titles_in_page.append(video_title)
1395 return zip(ids_in_page, titles_in_page)
1397 def _real_extract(self, url):
1398 channel_id = self._match_id(url)
1400 url = self._TEMPLATE_URL % channel_id
1401 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1402 autogenerated = re.search(r'''(?x)
1404 channel-header-autogenerated-label|
1405 yt-channel-title-autogenerated
1406 )[^"]*"''', channel_page) is not None
1409 # The videos are contained in a single page
1410 # the ajax pages can't be used, they are empty
1413 video_id, 'Youtube', video_id=video_id,
1414 video_title=video_title)
1415 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1416 return self.playlist_result(entries, channel_id)
1419 more_widget_html = content_html = channel_page
1420 for pagenum in itertools.count(1):
1422 for video_id, video_title in self.extract_videos_from_page(content_html):
1423 yield self.url_result(
1424 video_id, 'Youtube', video_id=video_id,
1425 video_title=video_title)
1428 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1433 more = self._download_json(
1434 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1435 'Downloading page #%s' % (pagenum + 1),
1436 transform_source=uppercase_escape)
1437 content_html = more['content_html']
1438 more_widget_html = more['load_more_widget_html']
1440 return self.playlist_result(_entries(), channel_id)
1443 class YoutubeUserIE(YoutubeChannelIE):
1444 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1445 _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_-]+)'
1446 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1447 IE_NAME = 'youtube:user'
1450 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1451 'playlist_mincount': 320,
1453 'title': 'TheLinuxFoundation',
1456 'url': 'ytuser:phihag',
1457 'only_matching': True,
1461 def suitable(cls, url):
1462 # Don't return True if the url can be extracted with other youtube
1463 # extractor, the regex would is too permissive and it would match.
1464 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1465 if any(ie.suitable(url) for ie in other_ies):
1468 return super(YoutubeUserIE, cls).suitable(url)
1471 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1472 IE_DESC = 'YouTube.com searches'
1473 # there doesn't appear to be a real limit, for example if you search for
1474 # 'python' you get more than 8.000.000 results
1475 _MAX_RESULTS = float('inf')
1476 IE_NAME = 'youtube:search'
1477 _SEARCH_KEY = 'ytsearch'
1478 _EXTRA_QUERY_ARGS = {}
1481 def _get_n_results(self, query, n):
1482 """Get a specified number of results for a query"""
1487 for pagenum in itertools.count(1):
1489 'search_query': query,
1493 url_query.update(self._EXTRA_QUERY_ARGS)
1494 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1495 data = self._download_json(
1496 result_url, video_id='query "%s"' % query,
1497 note='Downloading page %s' % pagenum,
1498 errnote='Unable to download API page')
1499 html_content = data[1]['body']['content']
1501 if 'class="search-message' in html_content:
1502 raise ExtractorError(
1503 '[youtube] No video results', expected=True)
1505 new_videos = self._ids_to_results(orderedSet(re.findall(
1506 r'href="/watch\?v=(.{11})', html_content)))
1507 videos += new_videos
1508 if not new_videos or len(videos) > limit:
1513 return self.playlist_result(videos, query)
1516 class YoutubeSearchDateIE(YoutubeSearchIE):
1517 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1518 _SEARCH_KEY = 'ytsearchdate'
1519 IE_DESC = 'YouTube.com searches, newest videos first'
1520 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1523 class YoutubeSearchURLIE(InfoExtractor):
1524 IE_DESC = 'YouTube.com search URLs'
1525 IE_NAME = 'youtube:search_url'
1526 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1528 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1529 'playlist_mincount': 5,
1531 'title': 'youtube-dl test video',
1535 def _real_extract(self, url):
1536 mobj = re.match(self._VALID_URL, url)
1537 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1539 webpage = self._download_webpage(url, query)
1540 result_code = self._search_regex(
1541 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1543 part_codes = re.findall(
1544 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1546 for part_code in part_codes:
1547 part_title = self._html_search_regex(
1548 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1549 part_url_snippet = self._html_search_regex(
1550 r'(?s)href="([^"]+)"', part_code, 'item URL')
1551 part_url = compat_urlparse.urljoin(
1552 'https://www.youtube.com/', part_url_snippet)
1556 'title': part_title,
1560 '_type': 'playlist',
1566 class YoutubeShowIE(InfoExtractor):
1567 IE_DESC = 'YouTube.com (multi-season) shows'
1568 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1569 IE_NAME = 'youtube:show'
1571 'url': 'http://www.youtube.com/show/airdisasters',
1572 'playlist_mincount': 3,
1574 'id': 'airdisasters',
1575 'title': 'Air Disasters',
1579 def _real_extract(self, url):
1580 mobj = re.match(self._VALID_URL, url)
1581 playlist_id = mobj.group('id')
1582 webpage = self._download_webpage(
1583 url, playlist_id, 'Downloading show webpage')
1584 # There's one playlist for each season of the show
1585 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1586 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1589 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1590 for season in m_seasons
1592 title = self._og_search_title(webpage, fatal=False)
1595 '_type': 'playlist',
1602 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1604 Base class for feed extractors
1605 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1607 _LOGIN_REQUIRED = True
1611 return 'youtube:%s' % self._FEED_NAME
1613 def _real_initialize(self):
1616 def _real_extract(self, url):
1617 page = self._download_webpage(
1618 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1620 # The extraction process is the same as for playlists, but the regex
1621 # for the video ids doesn't contain an index
1623 more_widget_html = content_html = page
1624 for page_num in itertools.count(1):
1625 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1627 # 'recommended' feed has infinite 'load more' and each new portion spins
1628 # the same videos in (sometimes) slightly different order, so we'll check
1629 # for unicity and break when portion has no new videos
1630 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1636 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1640 more = self._download_json(
1641 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1642 'Downloading page #%s' % page_num,
1643 transform_source=uppercase_escape)
1644 content_html = more['content_html']
1645 more_widget_html = more['load_more_widget_html']
1647 return self.playlist_result(
1648 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1651 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1652 IE_NAME = 'youtube:watchlater'
1653 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1654 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1656 _TESTS = [] # override PlaylistIE tests
1658 def _real_extract(self, url):
1659 return self._extract_playlist('WL')
1662 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1663 IE_NAME = 'youtube:favorites'
1664 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1665 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1666 _LOGIN_REQUIRED = True
1668 def _real_extract(self, url):
1669 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1670 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1671 return self.url_result(playlist_id, 'YoutubePlaylist')
1674 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1675 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1676 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1677 _FEED_NAME = 'recommended'
1678 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1681 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1682 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1683 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1684 _FEED_NAME = 'subscriptions'
1685 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1688 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1689 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1690 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1691 _FEED_NAME = 'history'
1692 _PLAYLIST_TITLE = 'Youtube History'
1695 class YoutubeTruncatedURLIE(InfoExtractor):
1696 IE_NAME = 'youtube:truncated_url'
1697 IE_DESC = False # Do not list
1698 _VALID_URL = r'''(?x)
1700 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1703 annotation_id=annotation_[^&]+|
1708 attribution_link\?a=[^&]+
1714 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1715 'only_matching': True,
1717 'url': 'http://www.youtube.com/watch?',
1718 'only_matching': True,
1720 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1721 'only_matching': True,
1723 'url': 'https://www.youtube.com/watch?feature=foo',
1724 'only_matching': True,
1726 'url': 'https://www.youtube.com/watch?hl=en-GB',
1727 'only_matching': True,
1730 def _real_extract(self, url):
1731 raise ExtractorError(
1732 'Did you forget to quote the URL? Remember that & is a meta '
1733 'character in most shells, so you want to put the URL in quotes, '
1735 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1736 ' or simply youtube-dl BaW_jenozKc .',
1740 class YoutubeTruncatedIDIE(InfoExtractor):
1741 IE_NAME = 'youtube:truncated_id'
1742 IE_DESC = False # Do not list
1743 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1746 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1747 'only_matching': True,
1750 def _real_extract(self, url):
1751 video_id = self._match_id(url)
1752 raise ExtractorError(
1753 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),