3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from .subtitles import SubtitlesInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
21 compat_urllib_request,
29 get_element_by_attribute,
40 class YoutubeBaseInfoExtractor(InfoExtractor):
41 """Provide base functions for Youtube extractors"""
42 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
43 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
44 _NETRC_MACHINE = 'youtube'
45 # If True it will raise an error if no login info is provided
46 _LOGIN_REQUIRED = False
48 def _set_language(self):
50 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
51 # YouTube sets the expire time to about two months
52 expire_time=time.time() + 2 * 30 * 24 * 3600)
56 Attempt to log in to YouTube.
57 True is returned if successful or skipped.
58 False is returned if login failed.
60 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
62 (username, password) = self._get_login_info()
63 # No authentication to be performed
65 if self._LOGIN_REQUIRED:
66 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
69 login_page = self._download_webpage(
70 self._LOGIN_URL, None,
71 note='Downloading login page',
72 errnote='unable to fetch login page', fatal=False)
73 if login_page is False:
76 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
77 login_page, 'Login GALX parameter')
81 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
86 'PersistentCookie': 'yes',
88 'bgresponse': 'js_disabled',
89 'checkConnection': '',
90 'checkedDomains': 'youtube',
102 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
104 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
105 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
107 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
108 login_results = self._download_webpage(
110 note='Logging in', errnote='unable to log in', fatal=False)
111 if login_results is False:
114 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
115 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
118 # TODO add SMS and phone call support - these require making a request and then prompting the user
120 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
121 tfa_code = self._get_tfa_info()
124 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
125 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
128 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
130 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
132 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
133 secTok = match.group(1)
134 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
136 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
137 timeStmp = match.group(1)
140 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
142 'smsUserPin': tfa_code,
143 'smsVerifyPin': 'Verify',
145 'PersistentCookie': 'yes',
146 'checkConnection': '',
147 'checkedDomains': 'youtube',
150 'timeStmp': timeStmp,
151 'service': 'youtube',
154 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
155 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
157 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
158 tfa_results = self._download_webpage(
160 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
162 if tfa_results is False:
165 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
166 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
168 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
169 self._downloader.report_warning('unable to log in - did the page structure change?')
171 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
172 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
175 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
176 self._downloader.report_warning('unable to log in: bad username or password')
180 def _real_initialize(self):
181 if self._downloader is None:
184 if not self._login():
188 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
189 IE_DESC = 'YouTube.com'
190 _VALID_URL = r"""(?x)^
192 (?:https?://|//) # http(s):// or protocol-independent URL
193 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
194 (?:www\.)?deturl\.com/www\.youtube\.com/|
195 (?:www\.)?pwnyoutube\.com/|
196 (?:www\.)?yourepeat\.com/|
197 tube\.majestyc\.net/|
198 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
199 (?:.*?\#/)? # handle anchor (#/) redirect urls
200 (?: # the various things that can precede the ID:
201 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
202 |(?: # or the v= param in all its forms
203 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
204 (?:\?|\#!?) # the params delimiter ? or # or #!
205 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
209 |youtu\.be/ # just youtu.be/xxxx
210 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
212 )? # all until now is optional -> you can pass the naked ID
213 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
214 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
215 (?(1).+)? # if we found the ID, everything can follow
217 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
219 '5': {'ext': 'flv', 'width': 400, 'height': 240},
220 '6': {'ext': 'flv', 'width': 450, 'height': 270},
221 '13': {'ext': '3gp'},
222 '17': {'ext': '3gp', 'width': 176, 'height': 144},
223 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
224 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
225 '34': {'ext': 'flv', 'width': 640, 'height': 360},
226 '35': {'ext': 'flv', 'width': 854, 'height': 480},
227 '36': {'ext': '3gp', 'width': 320, 'height': 240},
228 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
229 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
230 '43': {'ext': 'webm', 'width': 640, 'height': 360},
231 '44': {'ext': 'webm', 'width': 854, 'height': 480},
232 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
233 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
237 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
238 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
239 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
240 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
241 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
242 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
243 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
245 # Apple HTTP Live Streaming
246 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
247 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
248 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
249 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
250 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
251 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
252 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
255 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
261 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
263 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
264 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
265 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
268 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
269 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
270 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
273 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
278 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
279 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
280 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
289 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
290 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
291 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
292 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
293 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
296 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
297 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
299 # Dash webm audio with opus inside
300 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
301 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
302 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
305 '_rtmp': {'protocol': 'rtmp'},
311 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
315 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
316 'uploader': 'Philipp Hagemeister',
317 'uploader_id': 'phihag',
318 'upload_date': '20121002',
319 '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 .',
320 'categories': ['Science & Technology'],
322 'dislike_count': int,
326 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
327 'note': 'Test generic use_cipher_signature video (#897)',
331 'upload_date': '20120506',
332 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
333 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
334 'uploader': 'Icona Pop',
335 'uploader_id': 'IconaPop',
339 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
340 'note': 'Test VEVO video with age protection (#956)',
344 'upload_date': '20130703',
345 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
346 'description': 'md5:64249768eec3bc4276236606ea996373',
347 'uploader': 'justintimberlakeVEVO',
348 'uploader_id': 'justintimberlakeVEVO',
352 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
353 'note': 'Embed-only video (#1746)',
357 'upload_date': '20120608',
358 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
359 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
360 'uploader': 'SET India',
361 'uploader_id': 'setindia'
365 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
366 'note': '256k DASH audio (format 141) via DASH manifest',
370 'upload_date': '20121002',
371 'uploader_id': '8KVIDEO',
373 'uploader': '8KVIDEO',
374 'title': 'UHDTV TEST 8K VIDEO.mp4'
377 'youtube_include_dash_manifest': True,
381 # DASH manifest with encrypted signature
383 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
387 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
388 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
389 'uploader': 'AfrojackVEVO',
390 'uploader_id': 'AfrojackVEVO',
391 'upload_date': '20131011',
394 'youtube_include_dash_manifest': True,
398 # JS player signature function name containing $
400 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
404 'title': 'Taylor Swift - Shake It Off',
405 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
406 'uploader': 'TaylorSwiftVEVO',
407 'uploader_id': 'TaylorSwiftVEVO',
408 'upload_date': '20140818',
411 'youtube_include_dash_manifest': True,
417 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
421 'upload_date': '20100909',
422 'uploader': 'The Amazing Atheist',
423 'uploader_id': 'TheAmazingAtheist',
424 'title': 'Burning Everyone\'s Koran',
425 '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',
428 # Normal age-gate video (No vevo, embed allowed)
430 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
434 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
435 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
436 'uploader': 'The Witcher',
437 'uploader_id': 'WitcherGame',
438 'upload_date': '20140605',
441 # Age-gate video with encrypted signature
443 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
447 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
448 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
449 'uploader': 'LloydVEVO',
450 'uploader_id': 'LloydVEVO',
451 'upload_date': '20110629',
454 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
456 'url': '__2ABJjxzNo',
460 'upload_date': '20100430',
461 'uploader_id': 'deadmau5',
462 'description': 'md5:12c56784b8032162bb936a5f76d55360',
463 'uploader': 'deadmau5',
464 'title': 'Deadmau5 - Some Chords (HD)',
466 'expected_warnings': [
467 'DASH manifest missing',
470 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
472 'url': 'lqQg6PlCWgI',
476 'upload_date': '20120731',
477 'uploader_id': 'olympic',
478 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
479 'uploader': 'Olympics',
480 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
483 'skip_download': 'requires avconv',
488 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
492 'stretched_ratio': 16 / 9.,
493 'upload_date': '20110310',
494 'uploader_id': 'AllenMeow',
495 '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',
497 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
502 def __init__(self, *args, **kwargs):
503 super(YoutubeIE, self).__init__(*args, **kwargs)
504 self._player_cache = {}
506 def report_video_info_webpage_download(self, video_id):
507 """Report attempt to download video info webpage."""
508 self.to_screen('%s: Downloading video info webpage' % video_id)
510 def report_information_extraction(self, video_id):
511 """Report attempt to extract video information."""
512 self.to_screen('%s: Extracting video information' % video_id)
514 def report_unavailable_format(self, video_id, format):
515 """Report extracted video URL."""
516 self.to_screen('%s: Format %s not available' % (video_id, format))
518 def report_rtmp_download(self):
519 """Indicate the download will use the RTMP protocol."""
520 self.to_screen('RTMP download detected')
522 def _signature_cache_id(self, example_sig):
523 """ Return a string representation of a signature """
524 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
526 def _extract_signature_function(self, video_id, player_url, example_sig):
528 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
531 raise ExtractorError('Cannot identify player %r' % player_url)
532 player_type = id_m.group('ext')
533 player_id = id_m.group('id')
535 # Read from filesystem cache
536 func_id = '%s_%s_%s' % (
537 player_type, player_id, self._signature_cache_id(example_sig))
538 assert os.path.basename(func_id) == func_id
540 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
541 if cache_spec is not None:
542 return lambda s: ''.join(s[i] for i in cache_spec)
544 if player_type == 'js':
545 code = self._download_webpage(
546 player_url, video_id,
547 note='Downloading %s player %s' % (player_type, player_id),
548 errnote='Download of %s failed' % player_url)
549 res = self._parse_sig_js(code)
550 elif player_type == 'swf':
551 urlh = self._request_webpage(
552 player_url, video_id,
553 note='Downloading %s player %s' % (player_type, player_id),
554 errnote='Download of %s failed' % player_url)
556 res = self._parse_sig_swf(code)
558 assert False, 'Invalid player type %r' % player_type
560 if cache_spec is None:
561 test_string = ''.join(map(compat_chr, range(len(example_sig))))
562 cache_res = res(test_string)
563 cache_spec = [ord(c) for c in cache_res]
565 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
568 def _print_sig_code(self, func, example_sig):
569 def gen_sig_code(idxs):
570 def _genslice(start, end, step):
571 starts = '' if start == 0 else str(start)
572 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
573 steps = '' if step == 1 else (':%d' % step)
574 return 's[%s%s%s]' % (starts, ends, steps)
577 # Quelch pyflakes warnings - start will be set when step is set
578 start = '(Never used)'
579 for i, prev in zip(idxs[1:], idxs[:-1]):
583 yield _genslice(start, prev, step)
586 if i - prev in [-1, 1]:
595 yield _genslice(start, i, step)
597 test_string = ''.join(map(compat_chr, range(len(example_sig))))
598 cache_res = func(test_string)
599 cache_spec = [ord(c) for c in cache_res]
600 expr_code = ' + '.join(gen_sig_code(cache_spec))
601 signature_id_tuple = '(%s)' % (
602 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
603 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
604 ' return %s\n') % (signature_id_tuple, expr_code)
605 self.to_screen('Extracted signature function:\n' + code)
607 def _parse_sig_js(self, jscode):
608 funcname = self._search_regex(
609 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
610 'Initial JS player signature function name')
612 jsi = JSInterpreter(jscode)
613 initial_function = jsi.extract_function(funcname)
614 return lambda s: initial_function([s])
616 def _parse_sig_swf(self, file_contents):
617 swfi = SWFInterpreter(file_contents)
618 TARGET_CLASSNAME = 'SignatureDecipher'
619 searched_class = swfi.extract_class(TARGET_CLASSNAME)
620 initial_function = swfi.extract_function(searched_class, 'decipher')
621 return lambda s: initial_function([s])
623 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
624 """Turn the encrypted s field into a working signature"""
626 if player_url is None:
627 raise ExtractorError('Cannot decrypt signature without player_url')
629 if player_url.startswith('//'):
630 player_url = 'https:' + player_url
632 player_id = (player_url, self._signature_cache_id(s))
633 if player_id not in self._player_cache:
634 func = self._extract_signature_function(
635 video_id, player_url, s
637 self._player_cache[player_id] = func
638 func = self._player_cache[player_id]
639 if self._downloader.params.get('youtube_print_sig_code'):
640 self._print_sig_code(func, s)
642 except Exception as e:
643 tb = traceback.format_exc()
644 raise ExtractorError(
645 'Signature extraction failed: ' + tb, cause=e)
647 def _get_available_subtitles(self, video_id, webpage):
649 subs_doc = self._download_xml(
650 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
651 video_id, note=False)
652 except ExtractorError as err:
653 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
657 for track in subs_doc.findall('track'):
658 lang = track.attrib['lang_code']
659 if lang in sub_lang_list:
661 params = compat_urllib_parse.urlencode({
664 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
665 'name': track.attrib['name'].encode('utf-8'),
667 url = 'https://www.youtube.com/api/timedtext?' + params
668 sub_lang_list[lang] = url
669 if not sub_lang_list:
670 self._downloader.report_warning('video doesn\'t have subtitles')
674 def _get_available_automatic_caption(self, video_id, webpage):
675 """We need the webpage for getting the captions url, pass it as an
676 argument to speed up the process."""
677 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
678 self.to_screen('%s: Looking for automatic captions' % video_id)
679 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
680 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
682 self._downloader.report_warning(err_msg)
684 player_config = json.loads(mobj.group(1))
686 args = player_config['args']
687 caption_url = args['ttsurl']
688 timestamp = args['timestamp']
689 # We get the available subtitles
690 list_params = compat_urllib_parse.urlencode({
695 list_url = caption_url + '&' + list_params
696 caption_list = self._download_xml(list_url, video_id)
697 original_lang_node = caption_list.find('track')
698 if original_lang_node is None:
699 self._downloader.report_warning('Video doesn\'t have automatic captions')
701 original_lang = original_lang_node.attrib['lang_code']
702 caption_kind = original_lang_node.attrib.get('kind', '')
705 for lang_node in caption_list.findall('target'):
706 sub_lang = lang_node.attrib['lang_code']
707 params = compat_urllib_parse.urlencode({
708 'lang': original_lang,
712 'kind': caption_kind,
714 sub_lang_list[sub_lang] = caption_url + '&' + params
716 # An extractor error can be raise by the download process if there are
717 # no automatic captions but there are subtitles
718 except (KeyError, ExtractorError):
719 self._downloader.report_warning(err_msg)
723 def extract_id(cls, url):
724 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
726 raise ExtractorError('Invalid URL: %s' % url)
727 video_id = mobj.group(2)
730 def _extract_from_m3u8(self, manifest_url, video_id):
733 def _get_urls(_manifest):
734 lines = _manifest.split('\n')
735 urls = filter(lambda l: l and not l.startswith('#'),
738 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
739 formats_urls = _get_urls(manifest)
740 for format_url in formats_urls:
741 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
742 url_map[itag] = format_url
745 def _extract_annotations(self, video_id):
746 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
747 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
749 def _parse_dash_manifest(
750 self, video_id, dash_manifest_url, player_url, age_gate):
751 def decrypt_sig(mobj):
753 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
754 return '/signature/%s' % dec_s
755 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
756 dash_doc = self._download_xml(
757 dash_manifest_url, video_id,
758 note='Downloading DASH manifest',
759 errnote='Could not download DASH manifest')
762 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
763 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
766 format_id = r.attrib['id']
767 video_url = url_el.text
768 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
770 'format_id': format_id,
772 'width': int_or_none(r.attrib.get('width')),
773 'height': int_or_none(r.attrib.get('height')),
774 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
775 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
776 'filesize': filesize,
777 'fps': int_or_none(r.attrib.get('frameRate')),
780 existing_format = next(
782 if fo['format_id'] == format_id)
783 except StopIteration:
784 full_info = self._formats.get(format_id, {}).copy()
786 formats.append(full_info)
788 existing_format.update(f)
791 def _real_extract(self, url):
793 'http' if self._downloader.params.get('prefer_insecure', False)
796 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
797 mobj = re.search(self._NEXT_URL_RE, url)
799 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
800 video_id = self.extract_id(url)
803 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
804 video_webpage = self._download_webpage(url, video_id)
806 # Attempt to extract SWF player URL
807 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
809 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
815 if re.search(r'player-age-gate-content">', video_webpage) is not None:
817 # We simulate the access to the video from www.youtube.com/v/{video_id}
818 # this can be viewed without login into Youtube
819 url = proto + '://www.youtube.com/embed/%s' % video_id
820 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
821 data = compat_urllib_parse.urlencode({
822 'video_id': video_id,
823 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
824 'sts': self._search_regex(
825 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
827 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
828 video_info_webpage = self._download_webpage(
829 video_info_url, video_id,
830 note='Refetching age-gated info webpage',
831 errnote='unable to download video info webpage')
832 video_info = compat_parse_qs(video_info_webpage)
836 # Try looking directly into the video webpage
837 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
839 raise ValueError('Could not find ytplayer.config') # caught below
840 json_code = uppercase_escape(mobj.group(1))
841 ytplayer_config = json.loads(json_code)
842 args = ytplayer_config['args']
843 # Convert to the same format returned by compat_parse_qs
844 video_info = dict((k, [v]) for k, v in args.items())
845 if 'url_encoded_fmt_stream_map' not in args:
846 raise ValueError('No stream_map present') # caught below
848 # We fallback to the get_video_info pages (used by the embed page)
849 self.report_video_info_webpage_download(video_id)
850 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
852 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
853 % (proto, video_id, el_type))
854 video_info_webpage = self._download_webpage(
856 video_id, note=False,
857 errnote='unable to download video info webpage')
858 video_info = compat_parse_qs(video_info_webpage)
859 if 'token' in video_info:
861 if 'token' not in video_info:
862 if 'reason' in video_info:
863 raise ExtractorError(
864 'YouTube said: %s' % video_info['reason'][0],
865 expected=True, video_id=video_id)
867 raise ExtractorError(
868 '"token" parameter not in video info for unknown reason',
871 if 'view_count' in video_info:
872 view_count = int(video_info['view_count'][0])
876 # Check for "rental" videos
877 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
878 raise ExtractorError('"rental" videos not supported')
880 # Start extracting information
881 self.report_information_extraction(video_id)
884 if 'author' not in video_info:
885 raise ExtractorError('Unable to extract uploader name')
886 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
889 video_uploader_id = None
890 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
892 video_uploader_id = mobj.group(1)
894 self._downloader.report_warning('unable to extract uploader nickname')
897 if 'title' in video_info:
898 video_title = video_info['title'][0]
900 self._downloader.report_warning('Unable to extract video title')
904 # We try first to get a high quality image:
905 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
906 video_webpage, re.DOTALL)
907 if m_thumb is not None:
908 video_thumbnail = m_thumb.group(1)
909 elif 'thumbnail_url' not in video_info:
910 self._downloader.report_warning('unable to extract video thumbnail')
911 video_thumbnail = None
912 else: # don't panic if we can't find it
913 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
917 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
920 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
923 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
924 upload_date = unified_strdate(upload_date)
926 m_cat_container = self._search_regex(
927 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
928 video_webpage, 'categories', default=None)
930 category = self._html_search_regex(
931 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
933 video_categories = None if category is None else [category]
935 video_categories = None
938 video_description = get_element_by_id("eow-description", video_webpage)
939 if video_description:
940 video_description = re.sub(r'''(?x)
942 (?:[a-zA-Z-]+="[^"]+"\s+)*?
944 (?:[a-zA-Z-]+="[^"]+"\s+)*?
945 class="yt-uix-redirect-link"\s*>
948 ''', r'\1', video_description)
949 video_description = clean_html(video_description)
951 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
953 video_description = unescapeHTML(fd_mobj.group(1))
955 video_description = ''
957 def _extract_count(count_name):
958 count = self._search_regex(
959 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
960 video_webpage, count_name, default=None)
961 if count is not None:
962 return int(count.replace(',', ''))
964 like_count = _extract_count('like')
965 dislike_count = _extract_count('dislike')
968 video_subtitles = self.extract_subtitles(video_id, video_webpage)
970 if self._downloader.params.get('listsubtitles', False):
971 self._list_available_subtitles(video_id, video_webpage)
974 if 'length_seconds' not in video_info:
975 self._downloader.report_warning('unable to extract video duration')
976 video_duration = None
978 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
981 video_annotations = None
982 if self._downloader.params.get('writeannotations', False):
983 video_annotations = self._extract_annotations(video_id)
985 def _map_to_format_list(urlmap):
987 for itag, video_real_url in urlmap.items():
990 'url': video_real_url,
991 'player_url': player_url,
993 if itag in self._formats:
994 dct.update(self._formats[itag])
998 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
999 self.report_rtmp_download()
1001 'format_id': '_rtmp',
1003 'url': video_info['conn'][0],
1004 'player_url': player_url,
1006 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1007 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1008 if 'rtmpe%3Dyes' in encoded_url_map:
1009 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1011 for url_data_str in encoded_url_map.split(','):
1012 url_data = compat_parse_qs(url_data_str)
1013 if 'itag' not in url_data or 'url' not in url_data:
1015 format_id = url_data['itag'][0]
1016 url = url_data['url'][0]
1018 if 'sig' in url_data:
1019 url += '&signature=' + url_data['sig'][0]
1020 elif 's' in url_data:
1021 encrypted_sig = url_data['s'][0]
1022 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1024 jsplayer_url_json = self._search_regex(
1026 embed_webpage if age_gate else video_webpage,
1027 'JS player URL (1)', default=None)
1028 if not jsplayer_url_json and not age_gate:
1029 # We need the embed website after all
1030 if embed_webpage is None:
1031 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1032 embed_webpage = self._download_webpage(
1033 embed_url, video_id, 'Downloading embed webpage')
1034 jsplayer_url_json = self._search_regex(
1035 ASSETS_RE, embed_webpage, 'JS player URL')
1037 player_url = json.loads(jsplayer_url_json)
1038 if player_url is None:
1039 player_url_json = self._search_regex(
1040 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1041 video_webpage, 'age gate player URL')
1042 player_url = json.loads(player_url_json)
1044 if self._downloader.params.get('verbose'):
1045 if player_url is None:
1046 player_version = 'unknown'
1047 player_desc = 'unknown'
1049 if player_url.endswith('swf'):
1050 player_version = self._search_regex(
1051 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1052 'flash player', fatal=False)
1053 player_desc = 'flash player %s' % player_version
1055 player_version = self._search_regex(
1056 r'html5player-([^/]+?)(?:/html5player)?\.js',
1058 'html5 player', fatal=False)
1059 player_desc = 'html5 player %s' % player_version
1061 parts_sizes = self._signature_cache_id(encrypted_sig)
1062 self.to_screen('{%s} signature length %s, %s' %
1063 (format_id, parts_sizes, player_desc))
1065 signature = self._decrypt_signature(
1066 encrypted_sig, video_id, player_url, age_gate)
1067 url += '&signature=' + signature
1068 if 'ratebypass' not in url:
1069 url += '&ratebypass=yes'
1070 url_map[format_id] = url
1071 formats = _map_to_format_list(url_map)
1072 elif video_info.get('hlsvp'):
1073 manifest_url = video_info['hlsvp'][0]
1074 url_map = self._extract_from_m3u8(manifest_url, video_id)
1075 formats = _map_to_format_list(url_map)
1077 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1079 # Look for the DASH manifest
1080 if self._downloader.params.get('youtube_include_dash_manifest', True):
1081 dash_mpd = video_info.get('dashmpd')
1083 dash_manifest_url = dash_mpd[0]
1085 dash_formats = self._parse_dash_manifest(
1086 video_id, dash_manifest_url, player_url, age_gate)
1087 except (ExtractorError, KeyError) as e:
1088 self.report_warning(
1089 'Skipping DASH manifest: %r' % e, video_id)
1091 # Hide the formats we found through non-DASH
1092 dash_keys = set(df['format_id'] for df in dash_formats)
1094 if f['format_id'] in dash_keys:
1095 f['format_id'] = 'nondash-%s' % f['format_id']
1096 f['preference'] = f.get('preference', 0) - 10000
1097 formats.extend(dash_formats)
1099 # Check for malformed aspect ratio
1100 stretched_m = re.search(
1101 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1104 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1106 if f.get('vcodec') != 'none':
1107 f['stretched_ratio'] = ratio
1109 self._sort_formats(formats)
1113 'uploader': video_uploader,
1114 'uploader_id': video_uploader_id,
1115 'upload_date': upload_date,
1116 'title': video_title,
1117 'thumbnail': video_thumbnail,
1118 'description': video_description,
1119 'categories': video_categories,
1120 'subtitles': video_subtitles,
1121 'duration': video_duration,
1122 'age_limit': 18 if age_gate else 0,
1123 'annotations': video_annotations,
1124 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1125 'view_count': view_count,
1126 'like_count': like_count,
1127 'dislike_count': dislike_count,
1128 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1133 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1134 IE_DESC = 'YouTube.com playlists'
1135 _VALID_URL = r"""(?x)(?:
1140 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1141 \? (?:.*?&)*? (?:p|a|list)=
1145 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1146 # Top tracks, they can also include dots
1151 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1153 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1154 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1155 IE_NAME = 'youtube:playlist'
1157 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1159 'title': 'ytdl test PL',
1160 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1162 'playlist_count': 3,
1164 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1166 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1167 'title': 'YDL_Empty_List',
1169 'playlist_count': 0,
1171 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1172 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1174 'title': '29C3: Not my department',
1175 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1177 'playlist_count': 95,
1179 'note': 'issue #673',
1180 'url': 'PLBB231211A4F62143',
1182 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1183 'id': 'PLBB231211A4F62143',
1185 'playlist_mincount': 26,
1187 'note': 'Large playlist',
1188 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1190 'title': 'Uploads from Cauchemar',
1191 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1193 'playlist_mincount': 799,
1195 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1197 'title': 'YDL_safe_search',
1198 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1200 'playlist_count': 2,
1203 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1204 'playlist_count': 4,
1207 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1210 'note': 'Embedded SWF player',
1211 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1212 'playlist_count': 4,
1215 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1218 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1219 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1221 'title': 'Uploads from Interstellar Movie',
1222 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1224 'playlist_mincout': 21,
1227 def _real_initialize(self):
1230 def _ids_to_results(self, ids):
1232 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1235 def _extract_mix(self, playlist_id):
1236 # The mixes are generated from a a single video
1237 # the id of the playlist is just 'RD' + video_id
1238 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1239 webpage = self._download_webpage(
1240 url, playlist_id, 'Downloading Youtube mix')
1241 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1243 search_title('playlist-title') or
1244 search_title('title long-title') or
1245 search_title('title'))
1246 title = clean_html(title_span)
1247 ids = orderedSet(re.findall(
1248 r'''(?xs)data-video-username=".*?".*?
1249 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1251 url_results = self._ids_to_results(ids)
1253 return self.playlist_result(url_results, playlist_id, title)
1255 def _real_extract(self, url):
1256 # Extract playlist id
1257 mobj = re.match(self._VALID_URL, url)
1259 raise ExtractorError('Invalid URL: %s' % url)
1260 playlist_id = mobj.group(1) or mobj.group(2)
1262 # Check if it's a video-specific URL
1263 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1264 if 'v' in query_dict:
1265 video_id = query_dict['v'][0]
1266 if self._downloader.params.get('noplaylist'):
1267 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1268 return self.url_result(video_id, 'Youtube', video_id=video_id)
1270 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1272 if playlist_id.startswith('RD'):
1273 # Mixes require a custom extraction process
1274 return self._extract_mix(playlist_id)
1276 url = self._TEMPLATE_URL % playlist_id
1277 page = self._download_webpage(url, playlist_id)
1278 more_widget_html = content_html = page
1280 # Check if the playlist exists or is private
1281 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1282 raise ExtractorError(
1283 'The playlist doesn\'t exist or is private, use --username or '
1284 '--netrc to access it.',
1287 # Extract the video ids from the playlist pages
1290 for page_num in itertools.count(1):
1291 matches = re.finditer(self._VIDEO_RE, content_html)
1292 # We remove the duplicates and the link with index 0
1293 # (it's not the first video of the playlist)
1294 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1297 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1301 more = self._download_json(
1302 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1303 'Downloading page #%s' % page_num,
1304 transform_source=uppercase_escape)
1305 content_html = more['content_html']
1306 if not content_html.strip():
1307 # Some webpages show a "Load more" button but they don't
1310 more_widget_html = more['load_more_widget_html']
1312 playlist_title = self._html_search_regex(
1313 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1316 url_results = self._ids_to_results(ids)
1317 return self.playlist_result(url_results, playlist_id, playlist_title)
1320 class YoutubeChannelIE(InfoExtractor):
1321 IE_DESC = 'YouTube.com channels'
1322 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1323 IE_NAME = 'youtube:channel'
1325 'note': 'paginated channel',
1326 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1327 'playlist_mincount': 91,
1329 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1333 def extract_videos_from_page(self, page):
1335 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1336 if mobj.group(1) not in ids_in_page:
1337 ids_in_page.append(mobj.group(1))
1340 def _real_extract(self, url):
1341 channel_id = self._match_id(url)
1344 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1345 channel_page = self._download_webpage(url, channel_id)
1346 autogenerated = re.search(r'''(?x)
1348 channel-header-autogenerated-label|
1349 yt-channel-title-autogenerated
1350 )[^"]*"''', channel_page) is not None
1353 # The videos are contained in a single page
1354 # the ajax pages can't be used, they are empty
1355 video_ids = self.extract_videos_from_page(channel_page)
1357 self.url_result(video_id, 'Youtube', video_id=video_id)
1358 for video_id in video_ids]
1359 return self.playlist_result(entries, channel_id)
1362 more_widget_html = content_html = channel_page
1363 for pagenum in itertools.count(1):
1365 ids_in_page = self.extract_videos_from_page(content_html)
1366 for video_id in ids_in_page:
1367 yield self.url_result(
1368 video_id, 'Youtube', video_id=video_id)
1371 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1376 more = self._download_json(
1377 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1378 'Downloading page #%s' % (pagenum + 1),
1379 transform_source=uppercase_escape)
1380 content_html = more['content_html']
1381 more_widget_html = more['load_more_widget_html']
1383 return self.playlist_result(_entries(), channel_id)
1386 class YoutubeUserIE(InfoExtractor):
1387 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1388 _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_-]+)'
1389 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1390 _GDATA_PAGE_SIZE = 50
1391 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1392 IE_NAME = 'youtube:user'
1395 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1396 'playlist_mincount': 320,
1398 'title': 'TheLinuxFoundation',
1401 'url': 'ytuser:phihag',
1402 'only_matching': True,
1406 def suitable(cls, url):
1407 # Don't return True if the url can be extracted with other youtube
1408 # extractor, the regex would is too permissive and it would match.
1409 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1410 if any(ie.suitable(url) for ie in other_ies):
1413 return super(YoutubeUserIE, cls).suitable(url)
1415 def _real_extract(self, url):
1416 username = self._match_id(url)
1418 # Download video ids using YouTube Data API. Result size per
1419 # query is limited (currently to 50 videos) so we need to query
1420 # page by page until there are no video ids - it means we got
1423 def download_page(pagenum):
1424 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1426 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1427 page = self._download_webpage(
1428 gdata_url, username,
1429 'Downloading video ids from %d to %d' % (
1430 start_index, start_index + self._GDATA_PAGE_SIZE))
1433 response = json.loads(page)
1434 except ValueError as err:
1435 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1436 if 'entry' not in response['feed']:
1439 # Extract video identifiers
1440 entries = response['feed']['entry']
1441 for entry in entries:
1442 title = entry['title']['$t']
1443 video_id = entry['id']['$t'].split('/')[-1]
1447 'ie_key': 'Youtube',
1451 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1453 return self.playlist_result(url_results, playlist_title=username)
1456 class YoutubeSearchIE(SearchInfoExtractor):
1457 IE_DESC = 'YouTube.com searches'
1458 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1460 IE_NAME = 'youtube:search'
1461 _SEARCH_KEY = 'ytsearch'
1463 def _get_n_results(self, query, n):
1464 """Get a specified number of results for a query"""
1471 while (PAGE_SIZE * pagenum) < limit:
1472 result_url = self._API_URL % (
1473 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1474 (PAGE_SIZE * pagenum) + 1)
1475 data_json = self._download_webpage(
1476 result_url, video_id='query "%s"' % query,
1477 note='Downloading page %s' % (pagenum + 1),
1478 errnote='Unable to download API page')
1479 data = json.loads(data_json)
1480 api_response = data['data']
1482 if 'items' not in api_response:
1483 raise ExtractorError(
1484 '[youtube] No video results', expected=True)
1486 new_ids = list(video['id'] for video in api_response['items'])
1487 video_ids += new_ids
1489 limit = min(n, api_response['totalItems'])
1492 if len(video_ids) > n:
1493 video_ids = video_ids[:n]
1494 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1495 for video_id in video_ids]
1496 return self.playlist_result(videos, query)
1499 class YoutubeSearchDateIE(YoutubeSearchIE):
1500 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1501 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1502 _SEARCH_KEY = 'ytsearchdate'
1503 IE_DESC = 'YouTube.com searches, newest videos first'
1506 class YoutubeSearchURLIE(InfoExtractor):
1507 IE_DESC = 'YouTube.com search URLs'
1508 IE_NAME = 'youtube:search_url'
1509 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1511 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1512 'playlist_mincount': 5,
1514 'title': 'youtube-dl test video',
1518 def _real_extract(self, url):
1519 mobj = re.match(self._VALID_URL, url)
1520 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1522 webpage = self._download_webpage(url, query)
1523 result_code = self._search_regex(
1524 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1526 part_codes = re.findall(
1527 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1529 for part_code in part_codes:
1530 part_title = self._html_search_regex(
1531 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1532 part_url_snippet = self._html_search_regex(
1533 r'(?s)href="([^"]+)"', part_code, 'item URL')
1534 part_url = compat_urlparse.urljoin(
1535 'https://www.youtube.com/', part_url_snippet)
1539 'title': part_title,
1543 '_type': 'playlist',
1549 class YoutubeShowIE(InfoExtractor):
1550 IE_DESC = 'YouTube.com (multi-season) shows'
1551 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1552 IE_NAME = 'youtube:show'
1554 'url': 'http://www.youtube.com/show/airdisasters',
1555 'playlist_mincount': 3,
1557 'id': 'airdisasters',
1558 'title': 'Air Disasters',
1562 def _real_extract(self, url):
1563 mobj = re.match(self._VALID_URL, url)
1564 playlist_id = mobj.group('id')
1565 webpage = self._download_webpage(
1566 url, playlist_id, 'Downloading show webpage')
1567 # There's one playlist for each season of the show
1568 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1569 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1572 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1573 for season in m_seasons
1575 title = self._og_search_title(webpage, fatal=False)
1578 '_type': 'playlist',
1585 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1587 Base class for extractors that fetch info from
1588 http://www.youtube.com/feed_ajax
1589 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1591 _LOGIN_REQUIRED = True
1592 # use action_load_personal_feed instead of action_load_system_feed
1593 _PERSONAL_FEED = False
1596 def _FEED_TEMPLATE(self):
1597 action = 'action_load_system_feed'
1598 if self._PERSONAL_FEED:
1599 action = 'action_load_personal_feed'
1600 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1604 return 'youtube:%s' % self._FEED_NAME
1606 def _real_initialize(self):
1609 def _real_extract(self, url):
1612 for i in itertools.count(1):
1613 info = self._download_json(
1614 self._FEED_TEMPLATE % paging,
1615 '%s feed' % self._FEED_NAME,
1616 'Downloading page %s' % i,
1617 transform_source=uppercase_escape)
1618 feed_html = info.get('feed_html') or info.get('content_html')
1619 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1620 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1621 ids = orderedSet(m.group(1) for m in m_ids)
1622 feed_entries.extend(
1623 self.url_result(video_id, 'Youtube', video_id=video_id)
1624 for video_id in ids)
1626 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1627 load_more_widget_html)
1630 paging = mobj.group('paging')
1631 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1634 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1635 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1636 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1637 _FEED_NAME = 'recommended'
1638 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1641 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1642 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1643 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1644 _FEED_NAME = 'watch_later'
1645 _PLAYLIST_TITLE = 'Youtube Watch Later'
1646 _PERSONAL_FEED = True
1649 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1650 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1651 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1652 _FEED_NAME = 'history'
1653 _PERSONAL_FEED = True
1654 _PLAYLIST_TITLE = 'Youtube Watch History'
1657 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1658 IE_NAME = 'youtube:favorites'
1659 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1660 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1661 _LOGIN_REQUIRED = True
1663 def _real_extract(self, url):
1664 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1665 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1666 return self.url_result(playlist_id, 'YoutubePlaylist')
1669 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1670 IE_NAME = 'youtube:subscriptions'
1671 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1672 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1675 def _real_extract(self, url):
1676 title = 'Youtube Subscriptions'
1677 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1679 # The extraction process is the same as for playlists, but the regex
1680 # for the video ids doesn't contain an index
1682 more_widget_html = content_html = page
1684 for page_num in itertools.count(1):
1685 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1686 new_ids = orderedSet(matches)
1689 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1693 more = self._download_json(
1694 'https://youtube.com/%s' % mobj.group('more'), title,
1695 'Downloading page #%s' % page_num,
1696 transform_source=uppercase_escape)
1697 content_html = more['content_html']
1698 more_widget_html = more['load_more_widget_html']
1701 '_type': 'playlist',
1703 'entries': self._ids_to_results(ids),
1707 class YoutubeTruncatedURLIE(InfoExtractor):
1708 IE_NAME = 'youtube:truncated_url'
1709 IE_DESC = False # Do not list
1710 _VALID_URL = r'''(?x)
1712 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1715 annotation_id=annotation_[^&]+|
1720 attribution_link\?a=[^&]+
1726 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1727 'only_matching': True,
1729 'url': 'http://www.youtube.com/watch?',
1730 'only_matching': True,
1732 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1733 'only_matching': True,
1735 'url': 'https://www.youtube.com/watch?feature=foo',
1736 'only_matching': True,
1738 'url': 'https://www.youtube.com/watch?hl=en-GB',
1739 'only_matching': True,
1742 def _real_extract(self, url):
1743 raise ExtractorError(
1744 'Did you forget to quote the URL? Remember that & is a meta '
1745 'character in most shells, so you want to put the URL in quotes, '
1747 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1748 ' or simply youtube-dl BaW_jenozKc .',
1752 class YoutubeTruncatedIDIE(InfoExtractor):
1753 IE_NAME = 'youtube:truncated_id'
1754 IE_DESC = False # Do not list
1755 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1758 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1759 'only_matching': True,
1762 def _real_extract(self, url):
1763 video_id = self._match_id(url)
1764 raise ExtractorError(
1765 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),