3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
23 compat_urllib_request,
31 get_element_by_attribute,
44 class YoutubeBaseInfoExtractor(InfoExtractor):
45 """Provide base functions for Youtube extractors"""
46 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
47 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
48 _NETRC_MACHINE = 'youtube'
49 # If True it will raise an error if no login info is provided
50 _LOGIN_REQUIRED = False
52 def _set_language(self):
54 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
55 # YouTube sets the expire time to about two months
56 expire_time=time.time() + 2 * 30 * 24 * 3600)
58 def _ids_to_results(self, ids):
60 self.url_result(vid_id, 'Youtube', video_id=vid_id)
65 Attempt to log in to YouTube.
66 True is returned if successful or skipped.
67 False is returned if login failed.
69 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
71 (username, password) = self._get_login_info()
72 # No authentication to be performed
74 if self._LOGIN_REQUIRED:
75 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
78 login_page = self._download_webpage(
79 self._LOGIN_URL, None,
80 note='Downloading login page',
81 errnote='unable to fetch login page', fatal=False)
82 if login_page is False:
85 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
86 login_page, 'Login GALX parameter')
90 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
95 'PersistentCookie': 'yes',
97 'bgresponse': 'js_disabled',
98 'checkConnection': '',
99 'checkedDomains': 'youtube',
106 'service': 'youtube',
111 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
113 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
114 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
116 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
117 login_results = self._download_webpage(
119 note='Logging in', errnote='unable to log in', fatal=False)
120 if login_results is False:
123 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
124 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
127 # TODO add SMS and phone call support - these require making a request and then prompting the user
129 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
130 tfa_code = self._get_tfa_info()
133 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
134 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
137 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
139 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
141 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
142 secTok = match.group(1)
143 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
145 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
146 timeStmp = match.group(1)
149 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
151 'smsUserPin': tfa_code,
152 'smsVerifyPin': 'Verify',
154 'PersistentCookie': 'yes',
155 'checkConnection': '',
156 'checkedDomains': 'youtube',
159 'timeStmp': timeStmp,
160 'service': 'youtube',
163 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
164 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
166 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
167 tfa_results = self._download_webpage(
169 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
171 if tfa_results is False:
174 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
175 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
177 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
178 self._downloader.report_warning('unable to log in - did the page structure change?')
180 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
181 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
184 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
185 self._downloader.report_warning('unable to log in: bad username or password')
189 def _real_initialize(self):
190 if self._downloader is None:
193 if not self._login():
197 class YoutubeIE(YoutubeBaseInfoExtractor):
198 IE_DESC = 'YouTube.com'
199 _VALID_URL = r"""(?x)^
201 (?:https?://|//) # http(s):// or protocol-independent URL
202 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
203 (?:www\.)?deturl\.com/www\.youtube\.com/|
204 (?:www\.)?pwnyoutube\.com/|
205 (?:www\.)?yourepeat\.com/|
206 tube\.majestyc\.net/|
207 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
208 (?:.*?\#/)? # handle anchor (#/) redirect urls
209 (?: # the various things that can precede the ID:
210 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
211 |(?: # or the v= param in all its forms
212 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
213 (?:\?|\#!?) # the params delimiter ? or # or #!
214 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
218 |youtu\.be/ # just youtu.be/xxxx
219 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
221 )? # all until now is optional -> you can pass the naked ID
222 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
223 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
224 (?(1).+)? # if we found the ID, everything can follow
226 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
228 '5': {'ext': 'flv', 'width': 400, 'height': 240},
229 '6': {'ext': 'flv', 'width': 450, 'height': 270},
230 '13': {'ext': '3gp'},
231 '17': {'ext': '3gp', 'width': 176, 'height': 144},
232 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
233 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
234 '34': {'ext': 'flv', 'width': 640, 'height': 360},
235 '35': {'ext': 'flv', 'width': 854, 'height': 480},
236 '36': {'ext': '3gp', 'width': 320, 'height': 240},
237 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
238 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
239 '43': {'ext': 'webm', 'width': 640, 'height': 360},
240 '44': {'ext': 'webm', 'width': 854, 'height': 480},
241 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
242 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
243 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
244 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
248 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
249 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
250 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
251 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
252 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
253 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
254 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
256 # Apple HTTP Live Streaming
257 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
258 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
259 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
260 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
261 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
262 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
263 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
266 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
268 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
272 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
275 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
276 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
279 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
280 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
281 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
284 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
285 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
286 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
287 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
288 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
289 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
290 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'},
291 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
301 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
302 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
303 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
304 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
307 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
308 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
310 # Dash webm audio with opus inside
311 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
312 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
313 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
316 '_rtmp': {'protocol': 'rtmp'},
322 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
326 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
327 'uploader': 'Philipp Hagemeister',
328 'uploader_id': 'phihag',
329 'upload_date': '20121002',
330 '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 .',
331 'categories': ['Science & Technology'],
332 'tags': ['youtube-dl'],
334 'dislike_count': int,
340 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
341 'note': 'Test generic use_cipher_signature video (#897)',
345 'upload_date': '20120506',
346 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
347 'description': 'md5:782e8651347686cba06e58f71ab51773',
348 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
349 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
350 'iconic ep', 'iconic', 'love', 'it'],
351 'uploader': 'Icona Pop',
352 'uploader_id': 'IconaPop',
356 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
357 'note': 'Test VEVO video with age protection (#956)',
361 'upload_date': '20130703',
362 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
363 'description': 'md5:64249768eec3bc4276236606ea996373',
364 'uploader': 'justintimberlakeVEVO',
365 'uploader_id': 'justintimberlakeVEVO',
369 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
370 'note': 'Embed-only video (#1746)',
374 'upload_date': '20120608',
375 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
376 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
377 'uploader': 'SET India',
378 'uploader_id': 'setindia'
382 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
383 'note': '256k DASH audio (format 141) via DASH manifest',
387 'upload_date': '20121002',
388 'uploader_id': '8KVIDEO',
390 'uploader': '8KVIDEO',
391 'title': 'UHDTV TEST 8K VIDEO.mp4'
394 'youtube_include_dash_manifest': True,
398 # DASH manifest with encrypted signature
400 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
404 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
405 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
406 'uploader': 'AfrojackVEVO',
407 'uploader_id': 'AfrojackVEVO',
408 'upload_date': '20131011',
411 'youtube_include_dash_manifest': True,
415 # JS player signature function name containing $
417 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
421 'title': 'Taylor Swift - Shake It Off',
422 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
423 'uploader': 'TaylorSwiftVEVO',
424 'uploader_id': 'TaylorSwiftVEVO',
425 'upload_date': '20140818',
428 'youtube_include_dash_manifest': True,
434 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
438 'upload_date': '20100909',
439 'uploader': 'The Amazing Atheist',
440 'uploader_id': 'TheAmazingAtheist',
441 'title': 'Burning Everyone\'s Koran',
442 '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',
445 # Normal age-gate video (No vevo, embed allowed)
447 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
451 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
452 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
453 'uploader': 'The Witcher',
454 'uploader_id': 'WitcherGame',
455 'upload_date': '20140605',
458 # Age-gate video with encrypted signature
460 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
464 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
465 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
466 'uploader': 'LloydVEVO',
467 'uploader_id': 'LloydVEVO',
468 'upload_date': '20110629',
471 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
473 'url': '__2ABJjxzNo',
477 'upload_date': '20100430',
478 'uploader_id': 'deadmau5',
479 'description': 'md5:12c56784b8032162bb936a5f76d55360',
480 'uploader': 'deadmau5',
481 'title': 'Deadmau5 - Some Chords (HD)',
483 'expected_warnings': [
484 'DASH manifest missing',
487 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
489 'url': 'lqQg6PlCWgI',
493 'upload_date': '20120731',
494 'uploader_id': 'olympic',
495 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
496 'uploader': 'Olympics',
497 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
500 'skip_download': 'requires avconv',
505 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
509 'stretched_ratio': 16 / 9.,
510 'upload_date': '20110310',
511 'uploader_id': 'AllenMeow',
512 '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',
514 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
517 # url_encoded_fmt_stream_map is empty string
519 'url': 'qEJwOuvDf7I',
523 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
525 'upload_date': '20150404',
526 'uploader_id': 'spbelect',
527 'uploader': 'Наблюдатели Петербурга',
530 'skip_download': 'requires avconv',
533 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
535 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
539 'title': 'md5:7b81415841e02ecd4313668cde88737a',
540 'description': 'md5:116377fd2963b81ec4ce64b542173306',
541 'upload_date': '20150625',
542 'uploader_id': 'dorappi2000',
543 'uploader': 'dorappi2000',
544 'formats': 'mincount:33',
547 # DASH manifest with segment_list
549 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
550 'md5': '8ce563a1d667b599d21064e982ab9e31',
554 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
555 'uploader': 'Airtek',
556 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
557 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
558 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
561 'youtube_include_dash_manifest': True,
562 'format': '135', # bestvideo
567 def __init__(self, *args, **kwargs):
568 super(YoutubeIE, self).__init__(*args, **kwargs)
569 self._player_cache = {}
571 def report_video_info_webpage_download(self, video_id):
572 """Report attempt to download video info webpage."""
573 self.to_screen('%s: Downloading video info webpage' % video_id)
575 def report_information_extraction(self, video_id):
576 """Report attempt to extract video information."""
577 self.to_screen('%s: Extracting video information' % video_id)
579 def report_unavailable_format(self, video_id, format):
580 """Report extracted video URL."""
581 self.to_screen('%s: Format %s not available' % (video_id, format))
583 def report_rtmp_download(self):
584 """Indicate the download will use the RTMP protocol."""
585 self.to_screen('RTMP download detected')
587 def _signature_cache_id(self, example_sig):
588 """ Return a string representation of a signature """
589 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
591 def _extract_signature_function(self, video_id, player_url, example_sig):
593 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
596 raise ExtractorError('Cannot identify player %r' % player_url)
597 player_type = id_m.group('ext')
598 player_id = id_m.group('id')
600 # Read from filesystem cache
601 func_id = '%s_%s_%s' % (
602 player_type, player_id, self._signature_cache_id(example_sig))
603 assert os.path.basename(func_id) == func_id
605 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
606 if cache_spec is not None:
607 return lambda s: ''.join(s[i] for i in cache_spec)
610 'Downloading player %s' % player_url
611 if self._downloader.params.get('verbose') else
612 'Downloading %s player %s' % (player_type, player_id)
614 if player_type == 'js':
615 code = self._download_webpage(
616 player_url, video_id,
618 errnote='Download of %s failed' % player_url)
619 res = self._parse_sig_js(code)
620 elif player_type == 'swf':
621 urlh = self._request_webpage(
622 player_url, video_id,
624 errnote='Download of %s failed' % player_url)
626 res = self._parse_sig_swf(code)
628 assert False, 'Invalid player type %r' % player_type
630 test_string = ''.join(map(compat_chr, range(len(example_sig))))
631 cache_res = res(test_string)
632 cache_spec = [ord(c) for c in cache_res]
634 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
637 def _print_sig_code(self, func, example_sig):
638 def gen_sig_code(idxs):
639 def _genslice(start, end, step):
640 starts = '' if start == 0 else str(start)
641 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
642 steps = '' if step == 1 else (':%d' % step)
643 return 's[%s%s%s]' % (starts, ends, steps)
646 # Quelch pyflakes warnings - start will be set when step is set
647 start = '(Never used)'
648 for i, prev in zip(idxs[1:], idxs[:-1]):
652 yield _genslice(start, prev, step)
655 if i - prev in [-1, 1]:
664 yield _genslice(start, i, step)
666 test_string = ''.join(map(compat_chr, range(len(example_sig))))
667 cache_res = func(test_string)
668 cache_spec = [ord(c) for c in cache_res]
669 expr_code = ' + '.join(gen_sig_code(cache_spec))
670 signature_id_tuple = '(%s)' % (
671 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
672 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
673 ' return %s\n') % (signature_id_tuple, expr_code)
674 self.to_screen('Extracted signature function:\n' + code)
676 def _parse_sig_js(self, jscode):
677 funcname = self._search_regex(
678 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
679 'Initial JS player signature function name')
681 jsi = JSInterpreter(jscode)
682 initial_function = jsi.extract_function(funcname)
683 return lambda s: initial_function([s])
685 def _parse_sig_swf(self, file_contents):
686 swfi = SWFInterpreter(file_contents)
687 TARGET_CLASSNAME = 'SignatureDecipher'
688 searched_class = swfi.extract_class(TARGET_CLASSNAME)
689 initial_function = swfi.extract_function(searched_class, 'decipher')
690 return lambda s: initial_function([s])
692 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
693 """Turn the encrypted s field into a working signature"""
695 if player_url is None:
696 raise ExtractorError('Cannot decrypt signature without player_url')
698 if player_url.startswith('//'):
699 player_url = 'https:' + player_url
701 player_id = (player_url, self._signature_cache_id(s))
702 if player_id not in self._player_cache:
703 func = self._extract_signature_function(
704 video_id, player_url, s
706 self._player_cache[player_id] = func
707 func = self._player_cache[player_id]
708 if self._downloader.params.get('youtube_print_sig_code'):
709 self._print_sig_code(func, s)
711 except Exception as e:
712 tb = traceback.format_exc()
713 raise ExtractorError(
714 'Signature extraction failed: ' + tb, cause=e)
716 def _get_subtitles(self, video_id, webpage):
718 subs_doc = self._download_xml(
719 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
720 video_id, note=False)
721 except ExtractorError as err:
722 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
726 for track in subs_doc.findall('track'):
727 lang = track.attrib['lang_code']
728 if lang in sub_lang_list:
731 for ext in ['sbv', 'vtt', 'srt']:
732 params = compat_urllib_parse.urlencode({
736 'name': track.attrib['name'].encode('utf-8'),
739 'url': 'https://www.youtube.com/api/timedtext?' + params,
742 sub_lang_list[lang] = sub_formats
743 if not sub_lang_list:
744 self._downloader.report_warning('video doesn\'t have subtitles')
748 def _get_automatic_captions(self, video_id, webpage):
749 """We need the webpage for getting the captions url, pass it as an
750 argument to speed up the process."""
751 self.to_screen('%s: Looking for automatic captions' % video_id)
752 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
753 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
755 self._downloader.report_warning(err_msg)
757 player_config = json.loads(mobj.group(1))
759 args = player_config['args']
760 caption_url = args['ttsurl']
761 timestamp = args['timestamp']
762 # We get the available subtitles
763 list_params = compat_urllib_parse.urlencode({
768 list_url = caption_url + '&' + list_params
769 caption_list = self._download_xml(list_url, video_id)
770 original_lang_node = caption_list.find('track')
771 if original_lang_node is None:
772 self._downloader.report_warning('Video doesn\'t have automatic captions')
774 original_lang = original_lang_node.attrib['lang_code']
775 caption_kind = original_lang_node.attrib.get('kind', '')
778 for lang_node in caption_list.findall('target'):
779 sub_lang = lang_node.attrib['lang_code']
781 for ext in ['sbv', 'vtt', 'srt']:
782 params = compat_urllib_parse.urlencode({
783 'lang': original_lang,
787 'kind': caption_kind,
790 'url': caption_url + '&' + params,
793 sub_lang_list[sub_lang] = sub_formats
795 # An extractor error can be raise by the download process if there are
796 # no automatic captions but there are subtitles
797 except (KeyError, ExtractorError):
798 self._downloader.report_warning(err_msg)
802 def extract_id(cls, url):
803 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
805 raise ExtractorError('Invalid URL: %s' % url)
806 video_id = mobj.group(2)
809 def _extract_from_m3u8(self, manifest_url, video_id):
812 def _get_urls(_manifest):
813 lines = _manifest.split('\n')
814 urls = filter(lambda l: l and not l.startswith('#'),
817 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
818 formats_urls = _get_urls(manifest)
819 for format_url in formats_urls:
820 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
821 url_map[itag] = format_url
824 def _extract_annotations(self, video_id):
825 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
826 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
828 def _parse_dash_manifest(
829 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
830 def decrypt_sig(mobj):
832 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
833 return '/signature/%s' % dec_s
834 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
835 dash_doc = self._download_xml(
836 dash_manifest_url, video_id,
837 note='Downloading DASH manifest',
838 errnote='Could not download DASH manifest',
841 if dash_doc is False:
845 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
846 mime_type = a.attrib.get('mimeType')
847 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
848 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
851 if mime_type == 'text/vtt':
852 # TODO implement WebVTT downloading
854 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
855 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
856 format_id = r.attrib['id']
857 video_url = url_el.text
858 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
860 'format_id': format_id,
862 'width': int_or_none(r.attrib.get('width')),
863 'height': int_or_none(r.attrib.get('height')),
864 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
865 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
866 'filesize': filesize,
867 'fps': int_or_none(r.attrib.get('frameRate')),
869 if segment_list is not None:
871 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
872 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
873 'protocol': 'http_dash_segments',
876 existing_format = next(
878 if fo['format_id'] == format_id)
879 except StopIteration:
880 full_info = self._formats.get(format_id, {}).copy()
882 codecs = r.attrib.get('codecs')
884 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
885 full_info['vcodec'] = codecs
886 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
887 full_info['acodec'] = codecs
888 formats.append(full_info)
890 existing_format.update(f)
892 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
895 def _real_extract(self, url):
897 'http' if self._downloader.params.get('prefer_insecure', False)
902 parsed_url = compat_urllib_parse_urlparse(url)
903 for component in [parsed_url.fragment, parsed_url.query]:
904 query = compat_parse_qs(component)
905 if start_time is None and 't' in query:
906 start_time = parse_duration(query['t'][0])
907 if start_time is None and 'start' in query:
908 start_time = parse_duration(query['start'][0])
909 if end_time is None and 'end' in query:
910 end_time = parse_duration(query['end'][0])
912 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
913 mobj = re.search(self._NEXT_URL_RE, url)
915 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
916 video_id = self.extract_id(url)
919 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
920 video_webpage = self._download_webpage(url, video_id)
922 # Attempt to extract SWF player URL
923 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
925 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
931 def add_dash_mpd(video_info):
932 dash_mpd = video_info.get('dashmpd')
933 if dash_mpd and dash_mpd[0] not in dash_mpds:
934 dash_mpds.append(dash_mpd[0])
939 if re.search(r'player-age-gate-content">', video_webpage) is not None:
941 # We simulate the access to the video from www.youtube.com/v/{video_id}
942 # this can be viewed without login into Youtube
943 url = proto + '://www.youtube.com/embed/%s' % video_id
944 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
945 data = compat_urllib_parse.urlencode({
946 'video_id': video_id,
947 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
948 'sts': self._search_regex(
949 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
951 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
952 video_info_webpage = self._download_webpage(
953 video_info_url, video_id,
954 note='Refetching age-gated info webpage',
955 errnote='unable to download video info webpage')
956 video_info = compat_parse_qs(video_info_webpage)
957 add_dash_mpd(video_info)
961 # Try looking directly into the video webpage
962 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
964 json_code = uppercase_escape(mobj.group(1))
965 ytplayer_config = json.loads(json_code)
966 args = ytplayer_config['args']
967 if args.get('url_encoded_fmt_stream_map'):
968 # Convert to the same format returned by compat_parse_qs
969 video_info = dict((k, [v]) for k, v in args.items())
970 add_dash_mpd(video_info)
971 if args.get('livestream') == '1' or args.get('live_playback') == 1:
973 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
974 # We also try looking in get_video_info since it may contain different dashmpd
975 # URL that points to a DASH manifest with possibly different itag set (some itags
976 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
977 # manifest pointed by get_video_info's dashmpd).
978 # The general idea is to take a union of itags of both DASH manifests (for example
979 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
980 self.report_video_info_webpage_download(video_id)
981 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
983 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
984 % (proto, video_id, el_type))
985 video_info_webpage = self._download_webpage(
987 video_id, note=False,
988 errnote='unable to download video info webpage')
989 get_video_info = compat_parse_qs(video_info_webpage)
990 if get_video_info.get('use_cipher_signature') != ['True']:
991 add_dash_mpd(get_video_info)
993 video_info = get_video_info
994 if 'token' in get_video_info:
996 if 'token' not in video_info:
997 if 'reason' in video_info:
998 if 'The uploader has not made this video available in your country.' in video_info['reason']:
999 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1001 raise ExtractorError('YouTube said: This video is available in %s only' % (
1002 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1004 raise ExtractorError(
1005 'YouTube said: %s' % video_info['reason'][0],
1006 expected=True, video_id=video_id)
1008 raise ExtractorError(
1009 '"token" parameter not in video info for unknown reason',
1012 if 'view_count' in video_info:
1013 view_count = int(video_info['view_count'][0])
1017 # Check for "rental" videos
1018 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1019 raise ExtractorError('"rental" videos not supported')
1021 # Start extracting information
1022 self.report_information_extraction(video_id)
1025 if 'author' not in video_info:
1026 raise ExtractorError('Unable to extract uploader name')
1027 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1030 video_uploader_id = None
1031 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1032 if mobj is not None:
1033 video_uploader_id = mobj.group(1)
1035 self._downloader.report_warning('unable to extract uploader nickname')
1038 if 'title' in video_info:
1039 video_title = video_info['title'][0]
1041 self._downloader.report_warning('Unable to extract video title')
1045 # We try first to get a high quality image:
1046 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1047 video_webpage, re.DOTALL)
1048 if m_thumb is not None:
1049 video_thumbnail = m_thumb.group(1)
1050 elif 'thumbnail_url' not in video_info:
1051 self._downloader.report_warning('unable to extract video thumbnail')
1052 video_thumbnail = None
1053 else: # don't panic if we can't find it
1054 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1057 upload_date = self._html_search_meta(
1058 'datePublished', video_webpage, 'upload date', default=None)
1060 upload_date = self._search_regex(
1061 [r'(?s)id="eow-date.*?>(.*?)</span>',
1062 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1063 video_webpage, 'upload date', default=None)
1065 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1066 upload_date = unified_strdate(upload_date)
1068 m_cat_container = self._search_regex(
1069 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1070 video_webpage, 'categories', default=None)
1072 category = self._html_search_regex(
1073 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1075 video_categories = None if category is None else [category]
1077 video_categories = None
1080 unescapeHTML(m.group('content'))
1081 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1084 video_description = get_element_by_id("eow-description", video_webpage)
1085 if video_description:
1086 video_description = re.sub(r'''(?x)
1088 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1090 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1091 class="yt-uix-redirect-link"\s*>
1094 ''', r'\1', video_description)
1095 video_description = clean_html(video_description)
1097 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1099 video_description = unescapeHTML(fd_mobj.group(1))
1101 video_description = ''
1103 def _extract_count(count_name):
1104 return str_to_int(self._search_regex(
1105 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1106 % re.escape(count_name),
1107 video_webpage, count_name, default=None))
1109 like_count = _extract_count('like')
1110 dislike_count = _extract_count('dislike')
1113 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1114 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1116 if 'length_seconds' not in video_info:
1117 self._downloader.report_warning('unable to extract video duration')
1118 video_duration = None
1120 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1123 video_annotations = None
1124 if self._downloader.params.get('writeannotations', False):
1125 video_annotations = self._extract_annotations(video_id)
1127 def _map_to_format_list(urlmap):
1129 for itag, video_real_url in urlmap.items():
1132 'url': video_real_url,
1133 'player_url': player_url,
1135 if itag in self._formats:
1136 dct.update(self._formats[itag])
1140 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1141 self.report_rtmp_download()
1143 'format_id': '_rtmp',
1145 'url': video_info['conn'][0],
1146 'player_url': player_url,
1148 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1149 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1150 if 'rtmpe%3Dyes' in encoded_url_map:
1151 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1153 for url_data_str in encoded_url_map.split(','):
1154 url_data = compat_parse_qs(url_data_str)
1155 if 'itag' not in url_data or 'url' not in url_data:
1157 format_id = url_data['itag'][0]
1158 url = url_data['url'][0]
1160 if 'sig' in url_data:
1161 url += '&signature=' + url_data['sig'][0]
1162 elif 's' in url_data:
1163 encrypted_sig = url_data['s'][0]
1164 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1166 jsplayer_url_json = self._search_regex(
1168 embed_webpage if age_gate else video_webpage,
1169 'JS player URL (1)', default=None)
1170 if not jsplayer_url_json and not age_gate:
1171 # We need the embed website after all
1172 if embed_webpage is None:
1173 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1174 embed_webpage = self._download_webpage(
1175 embed_url, video_id, 'Downloading embed webpage')
1176 jsplayer_url_json = self._search_regex(
1177 ASSETS_RE, embed_webpage, 'JS player URL')
1179 player_url = json.loads(jsplayer_url_json)
1180 if player_url is None:
1181 player_url_json = self._search_regex(
1182 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1183 video_webpage, 'age gate player URL')
1184 player_url = json.loads(player_url_json)
1186 if self._downloader.params.get('verbose'):
1187 if player_url is None:
1188 player_version = 'unknown'
1189 player_desc = 'unknown'
1191 if player_url.endswith('swf'):
1192 player_version = self._search_regex(
1193 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1194 'flash player', fatal=False)
1195 player_desc = 'flash player %s' % player_version
1197 player_version = self._search_regex(
1198 r'html5player-([^/]+?)(?:/html5player)?\.js',
1200 'html5 player', fatal=False)
1201 player_desc = 'html5 player %s' % player_version
1203 parts_sizes = self._signature_cache_id(encrypted_sig)
1204 self.to_screen('{%s} signature length %s, %s' %
1205 (format_id, parts_sizes, player_desc))
1207 signature = self._decrypt_signature(
1208 encrypted_sig, video_id, player_url, age_gate)
1209 url += '&signature=' + signature
1210 if 'ratebypass' not in url:
1211 url += '&ratebypass=yes'
1212 url_map[format_id] = url
1213 formats = _map_to_format_list(url_map)
1214 elif video_info.get('hlsvp'):
1215 manifest_url = video_info['hlsvp'][0]
1216 url_map = self._extract_from_m3u8(manifest_url, video_id)
1217 formats = _map_to_format_list(url_map)
1219 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1221 # Look for the DASH manifest
1222 if self._downloader.params.get('youtube_include_dash_manifest', True):
1223 dash_mpd_fatal = True
1224 for dash_manifest_url in dash_mpds:
1227 for df in self._parse_dash_manifest(
1228 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1229 # Do not overwrite DASH format found in some previous DASH manifest
1230 if df['format_id'] not in dash_formats:
1231 dash_formats[df['format_id']] = df
1232 # Additional DASH manifests may end up in HTTP Error 403 therefore
1233 # allow them to fail without bug report message if we already have
1234 # some DASH manifest succeeded. This is temporary workaround to reduce
1235 # burst of bug reports until we figure out the reason and whether it
1236 # can be fixed at all.
1237 dash_mpd_fatal = False
1238 except (ExtractorError, KeyError) as e:
1239 self.report_warning(
1240 'Skipping DASH manifest: %r' % e, video_id)
1242 # Remove the formats we found through non-DASH, they
1243 # contain less info and it can be wrong, because we use
1244 # fixed values (for example the resolution). See
1245 # https://github.com/rg3/youtube-dl/issues/5774 for an
1247 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1248 formats.extend(dash_formats.values())
1250 # Check for malformed aspect ratio
1251 stretched_m = re.search(
1252 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1255 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1257 if f.get('vcodec') != 'none':
1258 f['stretched_ratio'] = ratio
1260 self._sort_formats(formats)
1264 'uploader': video_uploader,
1265 'uploader_id': video_uploader_id,
1266 'upload_date': upload_date,
1267 'title': video_title,
1268 'thumbnail': video_thumbnail,
1269 'description': video_description,
1270 'categories': video_categories,
1272 'subtitles': video_subtitles,
1273 'automatic_captions': automatic_captions,
1274 'duration': video_duration,
1275 'age_limit': 18 if age_gate else 0,
1276 'annotations': video_annotations,
1277 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1278 'view_count': view_count,
1279 'like_count': like_count,
1280 'dislike_count': dislike_count,
1281 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1284 'start_time': start_time,
1285 'end_time': end_time,
1289 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1290 IE_DESC = 'YouTube.com playlists'
1291 _VALID_URL = r"""(?x)(?:
1296 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1297 \? (?:.*?&)*? (?:p|a|list)=
1301 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1302 # Top tracks, they can also include dots
1307 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1309 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1310 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1311 IE_NAME = 'youtube:playlist'
1313 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1315 'title': 'ytdl test PL',
1316 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1318 'playlist_count': 3,
1320 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1322 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1323 'title': 'YDL_Empty_List',
1325 'playlist_count': 0,
1327 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1328 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1330 'title': '29C3: Not my department',
1331 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1333 'playlist_count': 95,
1335 'note': 'issue #673',
1336 'url': 'PLBB231211A4F62143',
1338 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1339 'id': 'PLBB231211A4F62143',
1341 'playlist_mincount': 26,
1343 'note': 'Large playlist',
1344 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1346 'title': 'Uploads from Cauchemar',
1347 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1349 'playlist_mincount': 799,
1351 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1353 'title': 'YDL_safe_search',
1354 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1356 'playlist_count': 2,
1359 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1360 'playlist_count': 4,
1363 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1366 'note': 'Embedded SWF player',
1367 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1368 'playlist_count': 4,
1371 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1374 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1375 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1377 'title': 'Uploads from Interstellar Movie',
1378 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1380 'playlist_mincout': 21,
1383 def _real_initialize(self):
1386 def _extract_mix(self, playlist_id):
1387 # The mixes are generated from a single video
1388 # the id of the playlist is just 'RD' + video_id
1389 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1390 webpage = self._download_webpage(
1391 url, playlist_id, 'Downloading Youtube mix')
1392 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1394 search_title('playlist-title') or
1395 search_title('title long-title') or
1396 search_title('title'))
1397 title = clean_html(title_span)
1398 ids = orderedSet(re.findall(
1399 r'''(?xs)data-video-username=".*?".*?
1400 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1402 url_results = self._ids_to_results(ids)
1404 return self.playlist_result(url_results, playlist_id, title)
1406 def _extract_playlist(self, playlist_id):
1407 url = self._TEMPLATE_URL % playlist_id
1408 page = self._download_webpage(url, playlist_id)
1410 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1411 match = match.strip()
1412 # Check if the playlist exists or is private
1413 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1414 raise ExtractorError(
1415 'The playlist doesn\'t exist or is private, use --username or '
1416 '--netrc to access it.',
1418 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1419 raise ExtractorError(
1420 'Invalid parameters. Maybe URL is incorrect.',
1422 elif re.match(r'[^<]*Choose your language[^<]*', match):
1425 self.report_warning('Youtube gives an alert message: ' + match)
1427 # Extract the video ids from the playlist pages
1429 more_widget_html = content_html = page
1430 for page_num in itertools.count(1):
1431 matches = re.finditer(self._VIDEO_RE, content_html)
1432 # We remove the duplicates and the link with index 0
1433 # (it's not the first video of the playlist)
1434 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1435 for vid_id in new_ids:
1436 yield self.url_result(vid_id, 'Youtube', video_id=vid_id)
1438 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1442 more = self._download_json(
1443 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1444 'Downloading page #%s' % page_num,
1445 transform_source=uppercase_escape)
1446 content_html = more['content_html']
1447 if not content_html.strip():
1448 # Some webpages show a "Load more" button but they don't
1451 more_widget_html = more['load_more_widget_html']
1453 playlist_title = self._html_search_regex(
1454 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1457 return self.playlist_result(_entries(), playlist_id, playlist_title)
1459 def _real_extract(self, url):
1460 # Extract playlist id
1461 mobj = re.match(self._VALID_URL, url)
1463 raise ExtractorError('Invalid URL: %s' % url)
1464 playlist_id = mobj.group(1) or mobj.group(2)
1466 # Check if it's a video-specific URL
1467 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1468 if 'v' in query_dict:
1469 video_id = query_dict['v'][0]
1470 if self._downloader.params.get('noplaylist'):
1471 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1472 return self.url_result(video_id, 'Youtube', video_id=video_id)
1474 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1476 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1477 # Mixes require a custom extraction process
1478 return self._extract_mix(playlist_id)
1480 return self._extract_playlist(playlist_id)
1483 class YoutubeChannelIE(InfoExtractor):
1484 IE_DESC = 'YouTube.com channels'
1485 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1486 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1487 IE_NAME = 'youtube:channel'
1489 'note': 'paginated channel',
1490 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1491 'playlist_mincount': 91,
1493 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1498 def extract_videos_from_page(page):
1501 for mobj in re.finditer(r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page):
1502 video_id = mobj.group('id')
1503 video_title = unescapeHTML(mobj.group('title'))
1505 idx = ids_in_page.index(video_id)
1506 if video_title and not titles_in_page[idx]:
1507 titles_in_page[idx] = video_title
1509 ids_in_page.append(video_id)
1510 titles_in_page.append(video_title)
1511 return zip(ids_in_page, titles_in_page)
1513 def _real_extract(self, url):
1514 channel_id = self._match_id(url)
1516 url = self._TEMPLATE_URL % channel_id
1518 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1519 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1520 # otherwise fallback on channel by page extraction
1521 channel_page = self._download_webpage(
1522 url + '?view=57', channel_id,
1523 'Downloading channel page', fatal=False)
1524 channel_playlist_id = self._html_search_meta(
1525 'channelId', channel_page, 'channel id', default=None)
1526 if not channel_playlist_id:
1527 channel_playlist_id = self._search_regex(
1528 r'data-channel-external-id="([^"]+)"',
1529 channel_page, 'channel id', default=None)
1530 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1531 playlist_id = 'UU' + channel_playlist_id[2:]
1532 return self.url_result(
1533 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1535 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1536 autogenerated = re.search(r'''(?x)
1538 channel-header-autogenerated-label|
1539 yt-channel-title-autogenerated
1540 )[^"]*"''', channel_page) is not None
1543 # The videos are contained in a single page
1544 # the ajax pages can't be used, they are empty
1547 video_id, 'Youtube', video_id=video_id,
1548 video_title=video_title)
1549 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1550 return self.playlist_result(entries, channel_id)
1553 more_widget_html = content_html = channel_page
1554 for pagenum in itertools.count(1):
1556 for video_id, video_title in self.extract_videos_from_page(content_html):
1557 yield self.url_result(
1558 video_id, 'Youtube', video_id=video_id,
1559 video_title=video_title)
1562 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1567 more = self._download_json(
1568 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1569 'Downloading page #%s' % (pagenum + 1),
1570 transform_source=uppercase_escape)
1571 content_html = more['content_html']
1572 more_widget_html = more['load_more_widget_html']
1574 return self.playlist_result(_entries(), channel_id)
1577 class YoutubeUserIE(YoutubeChannelIE):
1578 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1579 _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_-]+)'
1580 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1581 IE_NAME = 'youtube:user'
1584 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1585 'playlist_mincount': 320,
1587 'title': 'TheLinuxFoundation',
1590 'url': 'ytuser:phihag',
1591 'only_matching': True,
1595 def suitable(cls, url):
1596 # Don't return True if the url can be extracted with other youtube
1597 # extractor, the regex would is too permissive and it would match.
1598 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1599 if any(ie.suitable(url) for ie in other_ies):
1602 return super(YoutubeUserIE, cls).suitable(url)
1605 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1606 IE_DESC = 'YouTube.com searches'
1607 # there doesn't appear to be a real limit, for example if you search for
1608 # 'python' you get more than 8.000.000 results
1609 _MAX_RESULTS = float('inf')
1610 IE_NAME = 'youtube:search'
1611 _SEARCH_KEY = 'ytsearch'
1612 _EXTRA_QUERY_ARGS = {}
1615 def _get_n_results(self, query, n):
1616 """Get a specified number of results for a query"""
1621 for pagenum in itertools.count(1):
1623 'search_query': query.encode('utf-8'),
1627 url_query.update(self._EXTRA_QUERY_ARGS)
1628 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1629 data = self._download_json(
1630 result_url, video_id='query "%s"' % query,
1631 note='Downloading page %s' % pagenum,
1632 errnote='Unable to download API page')
1633 html_content = data[1]['body']['content']
1635 if 'class="search-message' in html_content:
1636 raise ExtractorError(
1637 '[youtube] No video results', expected=True)
1639 new_videos = self._ids_to_results(orderedSet(re.findall(
1640 r'href="/watch\?v=(.{11})', html_content)))
1641 videos += new_videos
1642 if not new_videos or len(videos) > limit:
1647 return self.playlist_result(videos, query)
1650 class YoutubeSearchDateIE(YoutubeSearchIE):
1651 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1652 _SEARCH_KEY = 'ytsearchdate'
1653 IE_DESC = 'YouTube.com searches, newest videos first'
1654 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1657 class YoutubeSearchURLIE(InfoExtractor):
1658 IE_DESC = 'YouTube.com search URLs'
1659 IE_NAME = 'youtube:search_url'
1660 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1662 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1663 'playlist_mincount': 5,
1665 'title': 'youtube-dl test video',
1669 def _real_extract(self, url):
1670 mobj = re.match(self._VALID_URL, url)
1671 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1673 webpage = self._download_webpage(url, query)
1674 result_code = self._search_regex(
1675 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1677 part_codes = re.findall(
1678 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1680 for part_code in part_codes:
1681 part_title = self._html_search_regex(
1682 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1683 part_url_snippet = self._html_search_regex(
1684 r'(?s)href="([^"]+)"', part_code, 'item URL')
1685 part_url = compat_urlparse.urljoin(
1686 'https://www.youtube.com/', part_url_snippet)
1690 'title': part_title,
1694 '_type': 'playlist',
1700 class YoutubeShowIE(InfoExtractor):
1701 IE_DESC = 'YouTube.com (multi-season) shows'
1702 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1703 IE_NAME = 'youtube:show'
1705 'url': 'http://www.youtube.com/show/airdisasters',
1706 'playlist_mincount': 3,
1708 'id': 'airdisasters',
1709 'title': 'Air Disasters',
1713 def _real_extract(self, url):
1714 mobj = re.match(self._VALID_URL, url)
1715 playlist_id = mobj.group('id')
1716 webpage = self._download_webpage(
1717 url, playlist_id, 'Downloading show webpage')
1718 # There's one playlist for each season of the show
1719 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1720 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1723 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1724 for season in m_seasons
1726 title = self._og_search_title(webpage, fatal=False)
1729 '_type': 'playlist',
1736 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1738 Base class for feed extractors
1739 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1741 _LOGIN_REQUIRED = True
1745 return 'youtube:%s' % self._FEED_NAME
1747 def _real_initialize(self):
1750 def _real_extract(self, url):
1751 page = self._download_webpage(
1752 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1754 # The extraction process is the same as for playlists, but the regex
1755 # for the video ids doesn't contain an index
1757 more_widget_html = content_html = page
1758 for page_num in itertools.count(1):
1759 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1761 # 'recommended' feed has infinite 'load more' and each new portion spins
1762 # the same videos in (sometimes) slightly different order, so we'll check
1763 # for unicity and break when portion has no new videos
1764 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1770 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1774 more = self._download_json(
1775 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1776 'Downloading page #%s' % page_num,
1777 transform_source=uppercase_escape)
1778 content_html = more['content_html']
1779 more_widget_html = more['load_more_widget_html']
1781 return self.playlist_result(
1782 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1785 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1786 IE_NAME = 'youtube:watchlater'
1787 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1788 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1790 _TESTS = [] # override PlaylistIE tests
1792 def _real_extract(self, url):
1793 return self._extract_playlist('WL')
1796 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1797 IE_NAME = 'youtube:favorites'
1798 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1799 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1800 _LOGIN_REQUIRED = True
1802 def _real_extract(self, url):
1803 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1804 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1805 return self.url_result(playlist_id, 'YoutubePlaylist')
1808 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1809 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1810 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1811 _FEED_NAME = 'recommended'
1812 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1815 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1816 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1817 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1818 _FEED_NAME = 'subscriptions'
1819 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1822 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1823 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1824 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1825 _FEED_NAME = 'history'
1826 _PLAYLIST_TITLE = 'Youtube History'
1829 class YoutubeTruncatedURLIE(InfoExtractor):
1830 IE_NAME = 'youtube:truncated_url'
1831 IE_DESC = False # Do not list
1832 _VALID_URL = r'''(?x)
1834 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1837 annotation_id=annotation_[^&]+|
1842 attribution_link\?a=[^&]+
1848 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1849 'only_matching': True,
1851 'url': 'http://www.youtube.com/watch?',
1852 'only_matching': True,
1854 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1855 'only_matching': True,
1857 'url': 'https://www.youtube.com/watch?feature=foo',
1858 'only_matching': True,
1860 'url': 'https://www.youtube.com/watch?hl=en-GB',
1861 'only_matching': True,
1864 def _real_extract(self, url):
1865 raise ExtractorError(
1866 'Did you forget to quote the URL? Remember that & is a meta '
1867 'character in most shells, so you want to put the URL in quotes, '
1869 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1870 ' or simply youtube-dl BaW_jenozKc .',
1874 class YoutubeTruncatedIDIE(InfoExtractor):
1875 IE_NAME = 'youtube:truncated_id'
1876 IE_DESC = False # Do not list
1877 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1880 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1881 'only_matching': True,
1884 def _real_extract(self, url):
1885 video_id = self._match_id(url)
1886 raise ExtractorError(
1887 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),