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},
237 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
238 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
242 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
243 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
244 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
245 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
246 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
247 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
248 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
250 # Apple HTTP Live Streaming
251 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
252 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
253 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
254 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
255 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
256 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
257 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
260 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
263 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
264 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
265 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
266 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
268 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
269 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
270 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
273 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
274 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
275 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
278 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
279 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
280 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
281 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
282 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
283 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
284 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
285 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
289 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
295 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
296 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
297 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
298 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
301 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
302 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
304 # Dash webm audio with opus inside
305 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
306 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
307 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
310 '_rtmp': {'protocol': 'rtmp'},
316 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
320 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
321 'uploader': 'Philipp Hagemeister',
322 'uploader_id': 'phihag',
323 'upload_date': '20121002',
324 '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 .',
325 'categories': ['Science & Technology'],
327 'dislike_count': int,
331 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
332 'note': 'Test generic use_cipher_signature video (#897)',
336 'upload_date': '20120506',
337 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
338 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
339 'uploader': 'Icona Pop',
340 'uploader_id': 'IconaPop',
344 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
345 'note': 'Test VEVO video with age protection (#956)',
349 'upload_date': '20130703',
350 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
351 'description': 'md5:64249768eec3bc4276236606ea996373',
352 'uploader': 'justintimberlakeVEVO',
353 'uploader_id': 'justintimberlakeVEVO',
357 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
358 'note': 'Embed-only video (#1746)',
362 'upload_date': '20120608',
363 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
364 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
365 'uploader': 'SET India',
366 'uploader_id': 'setindia'
370 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
371 'note': '256k DASH audio (format 141) via DASH manifest',
375 'upload_date': '20121002',
376 'uploader_id': '8KVIDEO',
378 'uploader': '8KVIDEO',
379 'title': 'UHDTV TEST 8K VIDEO.mp4'
382 'youtube_include_dash_manifest': True,
386 # DASH manifest with encrypted signature
388 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
392 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
393 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
394 'uploader': 'AfrojackVEVO',
395 'uploader_id': 'AfrojackVEVO',
396 'upload_date': '20131011',
399 'youtube_include_dash_manifest': True,
403 # JS player signature function name containing $
405 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
409 'title': 'Taylor Swift - Shake It Off',
410 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
411 'uploader': 'TaylorSwiftVEVO',
412 'uploader_id': 'TaylorSwiftVEVO',
413 'upload_date': '20140818',
416 'youtube_include_dash_manifest': True,
422 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
426 'upload_date': '20100909',
427 'uploader': 'The Amazing Atheist',
428 'uploader_id': 'TheAmazingAtheist',
429 'title': 'Burning Everyone\'s Koran',
430 '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',
433 # Normal age-gate video (No vevo, embed allowed)
435 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
439 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
440 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
441 'uploader': 'The Witcher',
442 'uploader_id': 'WitcherGame',
443 'upload_date': '20140605',
446 # Age-gate video with encrypted signature
448 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
452 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
453 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
454 'uploader': 'LloydVEVO',
455 'uploader_id': 'LloydVEVO',
456 'upload_date': '20110629',
459 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
461 'url': '__2ABJjxzNo',
465 'upload_date': '20100430',
466 'uploader_id': 'deadmau5',
467 'description': 'md5:12c56784b8032162bb936a5f76d55360',
468 'uploader': 'deadmau5',
469 'title': 'Deadmau5 - Some Chords (HD)',
471 'expected_warnings': [
472 'DASH manifest missing',
475 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
477 'url': 'lqQg6PlCWgI',
481 'upload_date': '20120731',
482 'uploader_id': 'olympic',
483 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
484 'uploader': 'Olympics',
485 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
488 'skip_download': 'requires avconv',
493 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
497 'stretched_ratio': 16 / 9.,
498 'upload_date': '20110310',
499 'uploader_id': 'AllenMeow',
500 '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',
502 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
505 # url_encoded_fmt_stream_map is empty string
507 'url': 'qEJwOuvDf7I',
511 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
513 'upload_date': '20150404',
514 'uploader_id': 'spbelect',
515 'uploader': 'Наблюдатели Петербурга',
518 'skip_download': 'requires avconv',
523 def __init__(self, *args, **kwargs):
524 super(YoutubeIE, self).__init__(*args, **kwargs)
525 self._player_cache = {}
527 def report_video_info_webpage_download(self, video_id):
528 """Report attempt to download video info webpage."""
529 self.to_screen('%s: Downloading video info webpage' % video_id)
531 def report_information_extraction(self, video_id):
532 """Report attempt to extract video information."""
533 self.to_screen('%s: Extracting video information' % video_id)
535 def report_unavailable_format(self, video_id, format):
536 """Report extracted video URL."""
537 self.to_screen('%s: Format %s not available' % (video_id, format))
539 def report_rtmp_download(self):
540 """Indicate the download will use the RTMP protocol."""
541 self.to_screen('RTMP download detected')
543 def _signature_cache_id(self, example_sig):
544 """ Return a string representation of a signature """
545 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
547 def _extract_signature_function(self, video_id, player_url, example_sig):
549 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
552 raise ExtractorError('Cannot identify player %r' % player_url)
553 player_type = id_m.group('ext')
554 player_id = id_m.group('id')
556 # Read from filesystem cache
557 func_id = '%s_%s_%s' % (
558 player_type, player_id, self._signature_cache_id(example_sig))
559 assert os.path.basename(func_id) == func_id
561 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
562 if cache_spec is not None:
563 return lambda s: ''.join(s[i] for i in cache_spec)
566 'Downloading player %s' % player_url
567 if self._downloader.params.get('verbose') else
568 'Downloading %s player %s' % (player_type, player_id)
570 if player_type == 'js':
571 code = self._download_webpage(
572 player_url, video_id,
574 errnote='Download of %s failed' % player_url)
575 res = self._parse_sig_js(code)
576 elif player_type == 'swf':
577 urlh = self._request_webpage(
578 player_url, video_id,
580 errnote='Download of %s failed' % player_url)
582 res = self._parse_sig_swf(code)
584 assert False, 'Invalid player type %r' % player_type
586 test_string = ''.join(map(compat_chr, range(len(example_sig))))
587 cache_res = res(test_string)
588 cache_spec = [ord(c) for c in cache_res]
590 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
593 def _print_sig_code(self, func, example_sig):
594 def gen_sig_code(idxs):
595 def _genslice(start, end, step):
596 starts = '' if start == 0 else str(start)
597 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
598 steps = '' if step == 1 else (':%d' % step)
599 return 's[%s%s%s]' % (starts, ends, steps)
602 # Quelch pyflakes warnings - start will be set when step is set
603 start = '(Never used)'
604 for i, prev in zip(idxs[1:], idxs[:-1]):
608 yield _genslice(start, prev, step)
611 if i - prev in [-1, 1]:
620 yield _genslice(start, i, step)
622 test_string = ''.join(map(compat_chr, range(len(example_sig))))
623 cache_res = func(test_string)
624 cache_spec = [ord(c) for c in cache_res]
625 expr_code = ' + '.join(gen_sig_code(cache_spec))
626 signature_id_tuple = '(%s)' % (
627 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
628 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
629 ' return %s\n') % (signature_id_tuple, expr_code)
630 self.to_screen('Extracted signature function:\n' + code)
632 def _parse_sig_js(self, jscode):
633 funcname = self._search_regex(
634 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
635 'Initial JS player signature function name')
637 jsi = JSInterpreter(jscode)
638 initial_function = jsi.extract_function(funcname)
639 return lambda s: initial_function([s])
641 def _parse_sig_swf(self, file_contents):
642 swfi = SWFInterpreter(file_contents)
643 TARGET_CLASSNAME = 'SignatureDecipher'
644 searched_class = swfi.extract_class(TARGET_CLASSNAME)
645 initial_function = swfi.extract_function(searched_class, 'decipher')
646 return lambda s: initial_function([s])
648 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
649 """Turn the encrypted s field into a working signature"""
651 if player_url is None:
652 raise ExtractorError('Cannot decrypt signature without player_url')
654 if player_url.startswith('//'):
655 player_url = 'https:' + player_url
657 player_id = (player_url, self._signature_cache_id(s))
658 if player_id not in self._player_cache:
659 func = self._extract_signature_function(
660 video_id, player_url, s
662 self._player_cache[player_id] = func
663 func = self._player_cache[player_id]
664 if self._downloader.params.get('youtube_print_sig_code'):
665 self._print_sig_code(func, s)
667 except Exception as e:
668 tb = traceback.format_exc()
669 raise ExtractorError(
670 'Signature extraction failed: ' + tb, cause=e)
672 def _get_subtitles(self, video_id, webpage):
674 subs_doc = self._download_xml(
675 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
676 video_id, note=False)
677 except ExtractorError as err:
678 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
682 for track in subs_doc.findall('track'):
683 lang = track.attrib['lang_code']
684 if lang in sub_lang_list:
687 for ext in ['sbv', 'vtt', 'srt']:
688 params = compat_urllib_parse.urlencode({
692 'name': track.attrib['name'].encode('utf-8'),
695 'url': 'https://www.youtube.com/api/timedtext?' + params,
698 sub_lang_list[lang] = sub_formats
699 if not sub_lang_list:
700 self._downloader.report_warning('video doesn\'t have subtitles')
704 def _get_automatic_captions(self, video_id, webpage):
705 """We need the webpage for getting the captions url, pass it as an
706 argument to speed up the process."""
707 self.to_screen('%s: Looking for automatic captions' % video_id)
708 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
709 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
711 self._downloader.report_warning(err_msg)
713 player_config = json.loads(mobj.group(1))
715 args = player_config['args']
716 caption_url = args['ttsurl']
717 timestamp = args['timestamp']
718 # We get the available subtitles
719 list_params = compat_urllib_parse.urlencode({
724 list_url = caption_url + '&' + list_params
725 caption_list = self._download_xml(list_url, video_id)
726 original_lang_node = caption_list.find('track')
727 if original_lang_node is None:
728 self._downloader.report_warning('Video doesn\'t have automatic captions')
730 original_lang = original_lang_node.attrib['lang_code']
731 caption_kind = original_lang_node.attrib.get('kind', '')
734 for lang_node in caption_list.findall('target'):
735 sub_lang = lang_node.attrib['lang_code']
737 for ext in ['sbv', 'vtt', 'srt']:
738 params = compat_urllib_parse.urlencode({
739 'lang': original_lang,
743 'kind': caption_kind,
746 'url': caption_url + '&' + params,
749 sub_lang_list[sub_lang] = sub_formats
751 # An extractor error can be raise by the download process if there are
752 # no automatic captions but there are subtitles
753 except (KeyError, ExtractorError):
754 self._downloader.report_warning(err_msg)
758 def extract_id(cls, url):
759 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
761 raise ExtractorError('Invalid URL: %s' % url)
762 video_id = mobj.group(2)
765 def _extract_from_m3u8(self, manifest_url, video_id):
768 def _get_urls(_manifest):
769 lines = _manifest.split('\n')
770 urls = filter(lambda l: l and not l.startswith('#'),
773 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
774 formats_urls = _get_urls(manifest)
775 for format_url in formats_urls:
776 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
777 url_map[itag] = format_url
780 def _extract_annotations(self, video_id):
781 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
782 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
784 def _parse_dash_manifest(
785 self, video_id, dash_manifest_url, player_url, age_gate):
786 def decrypt_sig(mobj):
788 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
789 return '/signature/%s' % dec_s
790 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
791 dash_doc = self._download_xml(
792 dash_manifest_url, video_id,
793 note='Downloading DASH manifest',
794 errnote='Could not download DASH manifest')
797 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
798 mime_type = a.attrib.get('mimeType')
799 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
800 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
803 if mime_type == 'text/vtt':
804 # TODO implement WebVTT downloading
806 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
807 format_id = r.attrib['id']
808 video_url = url_el.text
809 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
811 'format_id': format_id,
813 'width': int_or_none(r.attrib.get('width')),
814 'height': int_or_none(r.attrib.get('height')),
815 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
816 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
817 'filesize': filesize,
818 'fps': int_or_none(r.attrib.get('frameRate')),
821 existing_format = next(
823 if fo['format_id'] == format_id)
824 except StopIteration:
825 full_info = self._formats.get(format_id, {}).copy()
827 codecs = r.attrib.get('codecs')
829 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
830 full_info['vcodec'] = codecs
831 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
832 full_info['acodec'] = codecs
833 formats.append(full_info)
835 existing_format.update(f)
837 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
840 def _real_extract(self, url):
842 'http' if self._downloader.params.get('prefer_insecure', False)
845 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
846 mobj = re.search(self._NEXT_URL_RE, url)
848 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
849 video_id = self.extract_id(url)
852 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
853 video_webpage = self._download_webpage(url, video_id)
855 # Attempt to extract SWF player URL
856 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
858 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
864 if re.search(r'player-age-gate-content">', video_webpage) is not None:
866 # We simulate the access to the video from www.youtube.com/v/{video_id}
867 # this can be viewed without login into Youtube
868 url = proto + '://www.youtube.com/embed/%s' % video_id
869 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
870 data = compat_urllib_parse.urlencode({
871 'video_id': video_id,
872 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
873 'sts': self._search_regex(
874 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
876 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
877 video_info_webpage = self._download_webpage(
878 video_info_url, video_id,
879 note='Refetching age-gated info webpage',
880 errnote='unable to download video info webpage')
881 video_info = compat_parse_qs(video_info_webpage)
885 # Try looking directly into the video webpage
886 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
888 raise ValueError('Could not find ytplayer.config') # caught below
889 json_code = uppercase_escape(mobj.group(1))
890 ytplayer_config = json.loads(json_code)
891 args = ytplayer_config['args']
892 # Convert to the same format returned by compat_parse_qs
893 video_info = dict((k, [v]) for k, v in args.items())
894 if not args.get('url_encoded_fmt_stream_map'):
895 raise ValueError('No stream_map present') # caught below
897 # We fallback to the get_video_info pages (used by the embed page)
898 self.report_video_info_webpage_download(video_id)
899 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
901 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
902 % (proto, video_id, el_type))
903 video_info_webpage = self._download_webpage(
905 video_id, note=False,
906 errnote='unable to download video info webpage')
907 video_info = compat_parse_qs(video_info_webpage)
908 if 'token' in video_info:
910 if 'token' not in video_info:
911 if 'reason' in video_info:
912 raise ExtractorError(
913 'YouTube said: %s' % video_info['reason'][0],
914 expected=True, video_id=video_id)
916 raise ExtractorError(
917 '"token" parameter not in video info for unknown reason',
920 if 'view_count' in video_info:
921 view_count = int(video_info['view_count'][0])
925 # Check for "rental" videos
926 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
927 raise ExtractorError('"rental" videos not supported')
929 # Start extracting information
930 self.report_information_extraction(video_id)
933 if 'author' not in video_info:
934 raise ExtractorError('Unable to extract uploader name')
935 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
938 video_uploader_id = None
939 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
941 video_uploader_id = mobj.group(1)
943 self._downloader.report_warning('unable to extract uploader nickname')
946 if 'title' in video_info:
947 video_title = video_info['title'][0]
949 self._downloader.report_warning('Unable to extract video title')
953 # We try first to get a high quality image:
954 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
955 video_webpage, re.DOTALL)
956 if m_thumb is not None:
957 video_thumbnail = m_thumb.group(1)
958 elif 'thumbnail_url' not in video_info:
959 self._downloader.report_warning('unable to extract video thumbnail')
960 video_thumbnail = None
961 else: # don't panic if we can't find it
962 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
966 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
969 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
972 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
973 upload_date = unified_strdate(upload_date)
975 m_cat_container = self._search_regex(
976 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
977 video_webpage, 'categories', default=None)
979 category = self._html_search_regex(
980 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
982 video_categories = None if category is None else [category]
984 video_categories = None
987 video_description = get_element_by_id("eow-description", video_webpage)
988 if video_description:
989 video_description = re.sub(r'''(?x)
991 (?:[a-zA-Z-]+="[^"]+"\s+)*?
993 (?:[a-zA-Z-]+="[^"]+"\s+)*?
994 class="yt-uix-redirect-link"\s*>
997 ''', r'\1', video_description)
998 video_description = clean_html(video_description)
1000 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1002 video_description = unescapeHTML(fd_mobj.group(1))
1004 video_description = ''
1006 def _extract_count(count_name):
1007 count = self._search_regex(
1008 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
1009 video_webpage, count_name, default=None)
1010 if count is not None:
1011 return int(count.replace(',', ''))
1013 like_count = _extract_count('like')
1014 dislike_count = _extract_count('dislike')
1017 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1018 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1020 if 'length_seconds' not in video_info:
1021 self._downloader.report_warning('unable to extract video duration')
1022 video_duration = None
1024 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
1027 video_annotations = None
1028 if self._downloader.params.get('writeannotations', False):
1029 video_annotations = self._extract_annotations(video_id)
1031 def _map_to_format_list(urlmap):
1033 for itag, video_real_url in urlmap.items():
1036 'url': video_real_url,
1037 'player_url': player_url,
1039 if itag in self._formats:
1040 dct.update(self._formats[itag])
1044 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1045 self.report_rtmp_download()
1047 'format_id': '_rtmp',
1049 'url': video_info['conn'][0],
1050 'player_url': player_url,
1052 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1053 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1054 if 'rtmpe%3Dyes' in encoded_url_map:
1055 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1057 for url_data_str in encoded_url_map.split(','):
1058 url_data = compat_parse_qs(url_data_str)
1059 if 'itag' not in url_data or 'url' not in url_data:
1061 format_id = url_data['itag'][0]
1062 url = url_data['url'][0]
1064 if 'sig' in url_data:
1065 url += '&signature=' + url_data['sig'][0]
1066 elif 's' in url_data:
1067 encrypted_sig = url_data['s'][0]
1068 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1070 jsplayer_url_json = self._search_regex(
1072 embed_webpage if age_gate else video_webpage,
1073 'JS player URL (1)', default=None)
1074 if not jsplayer_url_json and not age_gate:
1075 # We need the embed website after all
1076 if embed_webpage is None:
1077 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1078 embed_webpage = self._download_webpage(
1079 embed_url, video_id, 'Downloading embed webpage')
1080 jsplayer_url_json = self._search_regex(
1081 ASSETS_RE, embed_webpage, 'JS player URL')
1083 player_url = json.loads(jsplayer_url_json)
1084 if player_url is None:
1085 player_url_json = self._search_regex(
1086 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1087 video_webpage, 'age gate player URL')
1088 player_url = json.loads(player_url_json)
1090 if self._downloader.params.get('verbose'):
1091 if player_url is None:
1092 player_version = 'unknown'
1093 player_desc = 'unknown'
1095 if player_url.endswith('swf'):
1096 player_version = self._search_regex(
1097 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1098 'flash player', fatal=False)
1099 player_desc = 'flash player %s' % player_version
1101 player_version = self._search_regex(
1102 r'html5player-([^/]+?)(?:/html5player)?\.js',
1104 'html5 player', fatal=False)
1105 player_desc = 'html5 player %s' % player_version
1107 parts_sizes = self._signature_cache_id(encrypted_sig)
1108 self.to_screen('{%s} signature length %s, %s' %
1109 (format_id, parts_sizes, player_desc))
1111 signature = self._decrypt_signature(
1112 encrypted_sig, video_id, player_url, age_gate)
1113 url += '&signature=' + signature
1114 if 'ratebypass' not in url:
1115 url += '&ratebypass=yes'
1116 url_map[format_id] = url
1117 formats = _map_to_format_list(url_map)
1118 elif video_info.get('hlsvp'):
1119 manifest_url = video_info['hlsvp'][0]
1120 url_map = self._extract_from_m3u8(manifest_url, video_id)
1121 formats = _map_to_format_list(url_map)
1123 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1125 # Look for the DASH manifest
1126 if self._downloader.params.get('youtube_include_dash_manifest', True):
1127 dash_mpd = video_info.get('dashmpd')
1129 dash_manifest_url = dash_mpd[0]
1131 dash_formats = self._parse_dash_manifest(
1132 video_id, dash_manifest_url, player_url, age_gate)
1133 except (ExtractorError, KeyError) as e:
1134 self.report_warning(
1135 'Skipping DASH manifest: %r' % e, video_id)
1137 # Remove the formats we found through non-DASH, they
1138 # contain less info and it can be wrong, because we use
1139 # fixed values (for example the resolution). See
1140 # https://github.com/rg3/youtube-dl/issues/5774 for an
1142 dash_keys = set(df['format_id'] for df in dash_formats)
1143 formats = [f for f in formats if f['format_id'] not in dash_keys]
1144 formats.extend(dash_formats)
1146 # Check for malformed aspect ratio
1147 stretched_m = re.search(
1148 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1151 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1153 if f.get('vcodec') != 'none':
1154 f['stretched_ratio'] = ratio
1156 self._sort_formats(formats)
1160 'uploader': video_uploader,
1161 'uploader_id': video_uploader_id,
1162 'upload_date': upload_date,
1163 'title': video_title,
1164 'thumbnail': video_thumbnail,
1165 'description': video_description,
1166 'categories': video_categories,
1167 'subtitles': video_subtitles,
1168 'automatic_captions': automatic_captions,
1169 'duration': video_duration,
1170 'age_limit': 18 if age_gate else 0,
1171 'annotations': video_annotations,
1172 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1173 'view_count': view_count,
1174 'like_count': like_count,
1175 'dislike_count': dislike_count,
1176 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1181 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1182 IE_DESC = 'YouTube.com playlists'
1183 _VALID_URL = r"""(?x)(?:
1188 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1189 \? (?:.*?&)*? (?:p|a|list)=
1193 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1194 # Top tracks, they can also include dots
1199 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1201 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1202 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1203 IE_NAME = 'youtube:playlist'
1205 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1207 'title': 'ytdl test PL',
1208 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1210 'playlist_count': 3,
1212 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1214 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1215 'title': 'YDL_Empty_List',
1217 'playlist_count': 0,
1219 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1220 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1222 'title': '29C3: Not my department',
1223 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1225 'playlist_count': 95,
1227 'note': 'issue #673',
1228 'url': 'PLBB231211A4F62143',
1230 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1231 'id': 'PLBB231211A4F62143',
1233 'playlist_mincount': 26,
1235 'note': 'Large playlist',
1236 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1238 'title': 'Uploads from Cauchemar',
1239 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1241 'playlist_mincount': 799,
1243 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1245 'title': 'YDL_safe_search',
1246 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1248 'playlist_count': 2,
1251 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1252 'playlist_count': 4,
1255 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1258 'note': 'Embedded SWF player',
1259 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1260 'playlist_count': 4,
1263 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1266 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1267 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1269 'title': 'Uploads from Interstellar Movie',
1270 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1272 'playlist_mincout': 21,
1275 def _real_initialize(self):
1278 def _extract_mix(self, playlist_id):
1279 # The mixes are generated from a single video
1280 # the id of the playlist is just 'RD' + video_id
1281 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1282 webpage = self._download_webpage(
1283 url, playlist_id, 'Downloading Youtube mix')
1284 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1286 search_title('playlist-title') or
1287 search_title('title long-title') or
1288 search_title('title'))
1289 title = clean_html(title_span)
1290 ids = orderedSet(re.findall(
1291 r'''(?xs)data-video-username=".*?".*?
1292 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1294 url_results = self._ids_to_results(ids)
1296 return self.playlist_result(url_results, playlist_id, title)
1298 def _extract_playlist(self, playlist_id):
1299 url = self._TEMPLATE_URL % playlist_id
1300 page = self._download_webpage(url, playlist_id)
1302 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1303 match = match.strip()
1304 # Check if the playlist exists or is private
1305 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1306 raise ExtractorError(
1307 'The playlist doesn\'t exist or is private, use --username or '
1308 '--netrc to access it.',
1310 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1311 raise ExtractorError(
1312 'Invalid parameters. Maybe URL is incorrect.',
1314 elif re.match(r'[^<]*Choose your language[^<]*', match):
1317 self.report_warning('Youtube gives an alert message: ' + match)
1319 # Extract the video ids from the playlist pages
1321 more_widget_html = content_html = page
1322 for page_num in itertools.count(1):
1323 matches = re.finditer(self._VIDEO_RE, content_html)
1324 # We remove the duplicates and the link with index 0
1325 # (it's not the first video of the playlist)
1326 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1327 for vid_id in new_ids:
1328 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1330 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1334 more = self._download_json(
1335 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1336 'Downloading page #%s' % page_num,
1337 transform_source=uppercase_escape)
1338 content_html = more['content_html']
1339 if not content_html.strip():
1340 # Some webpages show a "Load more" button but they don't
1343 more_widget_html = more['load_more_widget_html']
1345 playlist_title = self._html_search_regex(
1346 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1349 return self.playlist_result(_entries(), playlist_id, playlist_title)
1351 def _real_extract(self, url):
1352 # Extract playlist id
1353 mobj = re.match(self._VALID_URL, url)
1355 raise ExtractorError('Invalid URL: %s' % url)
1356 playlist_id = mobj.group(1) or mobj.group(2)
1358 # Check if it's a video-specific URL
1359 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1360 if 'v' in query_dict:
1361 video_id = query_dict['v'][0]
1362 if self._downloader.params.get('noplaylist'):
1363 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1364 return self.url_result(video_id, 'Youtube', video_id=video_id)
1366 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1368 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1369 # Mixes require a custom extraction process
1370 return self._extract_mix(playlist_id)
1372 return self._extract_playlist(playlist_id)
1375 class YoutubeChannelIE(InfoExtractor):
1376 IE_DESC = 'YouTube.com channels'
1377 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1378 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1379 IE_NAME = 'youtube:channel'
1381 'note': 'paginated channel',
1382 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1383 'playlist_mincount': 91,
1385 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1390 def extract_videos_from_page(page):
1393 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1394 video_id = mobj.group('id')
1395 video_title = unescapeHTML(mobj.group('title'))
1397 idx = ids_in_page.index(video_id)
1398 if video_title and not titles_in_page[idx]:
1399 titles_in_page[idx] = video_title
1401 ids_in_page.append(video_id)
1402 titles_in_page.append(video_title)
1403 return zip(ids_in_page, titles_in_page)
1405 def _real_extract(self, url):
1406 channel_id = self._match_id(url)
1408 url = self._TEMPLATE_URL % channel_id
1410 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1411 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1412 # otherwise fallback on channel by page extraction
1413 channel_page = self._download_webpage(
1414 url + '?view=57', channel_id,
1415 'Downloading channel page', fatal=False)
1416 channel_playlist_id = self._html_search_meta(
1417 'channelId', channel_page, 'channel id', default=None)
1418 if not channel_playlist_id:
1419 channel_playlist_id = self._search_regex(
1420 r'data-channel-external-id="([^"]+)"',
1421 channel_page, 'channel id', default=None)
1422 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1423 playlist_id = 'UU' + channel_playlist_id[2:]
1424 return self.url_result(
1425 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1427 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1428 autogenerated = re.search(r'''(?x)
1430 channel-header-autogenerated-label|
1431 yt-channel-title-autogenerated
1432 )[^"]*"''', channel_page) is not None
1435 # The videos are contained in a single page
1436 # the ajax pages can't be used, they are empty
1439 video_id, 'Youtube', video_id=video_id,
1440 video_title=video_title)
1441 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1442 return self.playlist_result(entries, channel_id)
1445 more_widget_html = content_html = channel_page
1446 for pagenum in itertools.count(1):
1448 for video_id, video_title in self.extract_videos_from_page(content_html):
1449 yield self.url_result(
1450 video_id, 'Youtube', video_id=video_id,
1451 video_title=video_title)
1454 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1459 more = self._download_json(
1460 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1461 'Downloading page #%s' % (pagenum + 1),
1462 transform_source=uppercase_escape)
1463 content_html = more['content_html']
1464 more_widget_html = more['load_more_widget_html']
1466 return self.playlist_result(_entries(), channel_id)
1469 class YoutubeUserIE(YoutubeChannelIE):
1470 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1471 _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_-]+)'
1472 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1473 IE_NAME = 'youtube:user'
1476 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1477 'playlist_mincount': 320,
1479 'title': 'TheLinuxFoundation',
1482 'url': 'ytuser:phihag',
1483 'only_matching': True,
1487 def suitable(cls, url):
1488 # Don't return True if the url can be extracted with other youtube
1489 # extractor, the regex would is too permissive and it would match.
1490 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1491 if any(ie.suitable(url) for ie in other_ies):
1494 return super(YoutubeUserIE, cls).suitable(url)
1497 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1498 IE_DESC = 'YouTube.com searches'
1499 # there doesn't appear to be a real limit, for example if you search for
1500 # 'python' you get more than 8.000.000 results
1501 _MAX_RESULTS = float('inf')
1502 IE_NAME = 'youtube:search'
1503 _SEARCH_KEY = 'ytsearch'
1504 _EXTRA_QUERY_ARGS = {}
1507 def _get_n_results(self, query, n):
1508 """Get a specified number of results for a query"""
1513 for pagenum in itertools.count(1):
1515 'search_query': query.encode('utf-8'),
1519 url_query.update(self._EXTRA_QUERY_ARGS)
1520 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1521 data = self._download_json(
1522 result_url, video_id='query "%s"' % query,
1523 note='Downloading page %s' % pagenum,
1524 errnote='Unable to download API page')
1525 html_content = data[1]['body']['content']
1527 if 'class="search-message' in html_content:
1528 raise ExtractorError(
1529 '[youtube] No video results', expected=True)
1531 new_videos = self._ids_to_results(orderedSet(re.findall(
1532 r'href="/watch\?v=(.{11})', html_content)))
1533 videos += new_videos
1534 if not new_videos or len(videos) > limit:
1539 return self.playlist_result(videos, query)
1542 class YoutubeSearchDateIE(YoutubeSearchIE):
1543 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1544 _SEARCH_KEY = 'ytsearchdate'
1545 IE_DESC = 'YouTube.com searches, newest videos first'
1546 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1549 class YoutubeSearchURLIE(InfoExtractor):
1550 IE_DESC = 'YouTube.com search URLs'
1551 IE_NAME = 'youtube:search_url'
1552 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1554 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1555 'playlist_mincount': 5,
1557 'title': 'youtube-dl test video',
1561 def _real_extract(self, url):
1562 mobj = re.match(self._VALID_URL, url)
1563 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1565 webpage = self._download_webpage(url, query)
1566 result_code = self._search_regex(
1567 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1569 part_codes = re.findall(
1570 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1572 for part_code in part_codes:
1573 part_title = self._html_search_regex(
1574 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1575 part_url_snippet = self._html_search_regex(
1576 r'(?s)href="([^"]+)"', part_code, 'item URL')
1577 part_url = compat_urlparse.urljoin(
1578 'https://www.youtube.com/', part_url_snippet)
1582 'title': part_title,
1586 '_type': 'playlist',
1592 class YoutubeShowIE(InfoExtractor):
1593 IE_DESC = 'YouTube.com (multi-season) shows'
1594 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1595 IE_NAME = 'youtube:show'
1597 'url': 'http://www.youtube.com/show/airdisasters',
1598 'playlist_mincount': 3,
1600 'id': 'airdisasters',
1601 'title': 'Air Disasters',
1605 def _real_extract(self, url):
1606 mobj = re.match(self._VALID_URL, url)
1607 playlist_id = mobj.group('id')
1608 webpage = self._download_webpage(
1609 url, playlist_id, 'Downloading show webpage')
1610 # There's one playlist for each season of the show
1611 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1612 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1615 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1616 for season in m_seasons
1618 title = self._og_search_title(webpage, fatal=False)
1621 '_type': 'playlist',
1628 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1630 Base class for feed extractors
1631 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1633 _LOGIN_REQUIRED = True
1637 return 'youtube:%s' % self._FEED_NAME
1639 def _real_initialize(self):
1642 def _real_extract(self, url):
1643 page = self._download_webpage(
1644 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1646 # The extraction process is the same as for playlists, but the regex
1647 # for the video ids doesn't contain an index
1649 more_widget_html = content_html = page
1650 for page_num in itertools.count(1):
1651 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1653 # 'recommended' feed has infinite 'load more' and each new portion spins
1654 # the same videos in (sometimes) slightly different order, so we'll check
1655 # for unicity and break when portion has no new videos
1656 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1662 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1666 more = self._download_json(
1667 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1668 'Downloading page #%s' % page_num,
1669 transform_source=uppercase_escape)
1670 content_html = more['content_html']
1671 more_widget_html = more['load_more_widget_html']
1673 return self.playlist_result(
1674 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1677 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1678 IE_NAME = 'youtube:watchlater'
1679 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1680 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1682 _TESTS = [] # override PlaylistIE tests
1684 def _real_extract(self, url):
1685 return self._extract_playlist('WL')
1688 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1689 IE_NAME = 'youtube:favorites'
1690 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1691 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1692 _LOGIN_REQUIRED = True
1694 def _real_extract(self, url):
1695 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1696 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1697 return self.url_result(playlist_id, 'YoutubePlaylist')
1700 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1701 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1702 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1703 _FEED_NAME = 'recommended'
1704 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1707 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1708 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1709 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1710 _FEED_NAME = 'subscriptions'
1711 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1714 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1715 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1716 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1717 _FEED_NAME = 'history'
1718 _PLAYLIST_TITLE = 'Youtube History'
1721 class YoutubeTruncatedURLIE(InfoExtractor):
1722 IE_NAME = 'youtube:truncated_url'
1723 IE_DESC = False # Do not list
1724 _VALID_URL = r'''(?x)
1726 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1729 annotation_id=annotation_[^&]+|
1734 attribution_link\?a=[^&]+
1740 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1741 'only_matching': True,
1743 'url': 'http://www.youtube.com/watch?',
1744 'only_matching': True,
1746 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1747 'only_matching': True,
1749 'url': 'https://www.youtube.com/watch?feature=foo',
1750 'only_matching': True,
1752 'url': 'https://www.youtube.com/watch?hl=en-GB',
1753 'only_matching': True,
1756 def _real_extract(self, url):
1757 raise ExtractorError(
1758 'Did you forget to quote the URL? Remember that & is a meta '
1759 'character in most shells, so you want to put the URL in quotes, '
1761 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1762 ' or simply youtube-dl BaW_jenozKc .',
1766 class YoutubeTruncatedIDIE(InfoExtractor):
1767 IE_NAME = 'youtube:truncated_id'
1768 IE_DESC = False # Do not list
1769 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1772 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1773 'only_matching': True,
1776 def _real_extract(self, url):
1777 video_id = self._match_id(url)
1778 raise ExtractorError(
1779 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),