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,
28 get_element_by_attribute,
39 class YoutubeBaseInfoExtractor(InfoExtractor):
40 """Provide base functions for Youtube extractors"""
41 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
42 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
43 _NETRC_MACHINE = 'youtube'
44 # If True it will raise an error if no login info is provided
45 _LOGIN_REQUIRED = False
47 def _set_language(self):
49 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
50 # YouTube sets the expire time to about two months
51 expire_time=time.time() + 2 * 30 * 24 * 3600)
55 Attempt to log in to YouTube.
56 True is returned if successful or skipped.
57 False is returned if login failed.
59 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
61 (username, password) = self._get_login_info()
62 # No authentication to be performed
64 if self._LOGIN_REQUIRED:
65 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
68 login_page = self._download_webpage(
69 self._LOGIN_URL, None,
70 note='Downloading login page',
71 errnote='unable to fetch login page', fatal=False)
72 if login_page is False:
75 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
76 login_page, 'Login GALX parameter')
80 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
85 'PersistentCookie': 'yes',
87 'bgresponse': 'js_disabled',
88 'checkConnection': '',
89 'checkedDomains': 'youtube',
101 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
103 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
104 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
106 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
107 login_results = self._download_webpage(
109 note='Logging in', errnote='unable to log in', fatal=False)
110 if login_results is False:
113 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
114 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
117 # TODO add SMS and phone call support - these require making a request and then prompting the user
119 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
120 tfa_code = self._get_tfa_info()
123 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
124 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
127 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
129 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
131 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
132 secTok = match.group(1)
133 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
135 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
136 timeStmp = match.group(1)
139 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
141 'smsUserPin': tfa_code,
142 'smsVerifyPin': 'Verify',
144 'PersistentCookie': 'yes',
145 'checkConnection': '',
146 'checkedDomains': 'youtube',
149 'timeStmp': timeStmp,
150 'service': 'youtube',
153 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
154 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
156 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
157 tfa_results = self._download_webpage(
159 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
161 if tfa_results is False:
164 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
165 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
167 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
168 self._downloader.report_warning('unable to log in - did the page structure change?')
170 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
171 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
174 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
175 self._downloader.report_warning('unable to log in: bad username or password')
179 def _real_initialize(self):
180 if self._downloader is None:
183 if not self._login():
187 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
188 IE_DESC = 'YouTube.com'
189 _VALID_URL = r"""(?x)^
191 (?:https?://|//) # http(s):// or protocol-independent URL
192 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
193 (?:www\.)?deturl\.com/www\.youtube\.com/|
194 (?:www\.)?pwnyoutube\.com/|
195 (?:www\.)?yourepeat\.com/|
196 tube\.majestyc\.net/|
197 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
198 (?:.*?\#/)? # handle anchor (#/) redirect urls
199 (?: # the various things that can precede the ID:
200 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
201 |(?: # or the v= param in all its forms
202 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
203 (?:\?|\#!?) # the params delimiter ? or # or #!
204 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
208 |youtu\.be/ # just youtu.be/xxxx
209 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
211 )? # all until now is optional -> you can pass the naked ID
212 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
213 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
214 (?(1).+)? # if we found the ID, everything can follow
216 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
218 '5': {'ext': 'flv', 'width': 400, 'height': 240},
219 '6': {'ext': 'flv', 'width': 450, 'height': 270},
220 '13': {'ext': '3gp'},
221 '17': {'ext': '3gp', 'width': 176, 'height': 144},
222 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
223 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
224 '34': {'ext': 'flv', 'width': 640, 'height': 360},
225 '35': {'ext': 'flv', 'width': 854, 'height': 480},
226 '36': {'ext': '3gp', 'width': 320, 'height': 240},
227 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
228 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
229 '43': {'ext': 'webm', 'width': 640, 'height': 360},
230 '44': {'ext': 'webm', 'width': 854, 'height': 480},
231 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
232 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
236 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
237 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
238 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
239 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
240 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
241 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
242 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
244 # Apple HTTP Live Streaming
245 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
246 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
247 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
248 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
249 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
250 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
251 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
254 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
264 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
267 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
268 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
269 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
272 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
278 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
279 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
290 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
293 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
294 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
296 # Dash webm audio with opus inside
297 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
298 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
299 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
302 '_rtmp': {'protocol': 'rtmp'},
308 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
312 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
313 'uploader': 'Philipp Hagemeister',
314 'uploader_id': 'phihag',
315 'upload_date': '20121002',
316 '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 .',
317 'categories': ['Science & Technology'],
319 'dislike_count': int,
323 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
324 'note': 'Test generic use_cipher_signature video (#897)',
328 'upload_date': '20120506',
329 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
330 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
331 'uploader': 'Icona Pop',
332 'uploader_id': 'IconaPop',
336 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
337 'note': 'Test VEVO video with age protection (#956)',
341 'upload_date': '20130703',
342 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
343 'description': 'md5:64249768eec3bc4276236606ea996373',
344 'uploader': 'justintimberlakeVEVO',
345 'uploader_id': 'justintimberlakeVEVO',
349 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
350 'note': 'Embed-only video (#1746)',
354 'upload_date': '20120608',
355 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
356 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
357 'uploader': 'SET India',
358 'uploader_id': 'setindia'
362 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
363 'note': '256k DASH audio (format 141) via DASH manifest',
367 'upload_date': '20121002',
368 'uploader_id': '8KVIDEO',
370 'uploader': '8KVIDEO',
371 'title': 'UHDTV TEST 8K VIDEO.mp4'
374 'youtube_include_dash_manifest': True,
378 # DASH manifest with encrypted signature
380 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
384 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
385 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
386 'uploader': 'AfrojackVEVO',
387 'uploader_id': 'AfrojackVEVO',
388 'upload_date': '20131011',
391 'youtube_include_dash_manifest': True,
397 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
401 'upload_date': '20100909',
402 'uploader': 'The Amazing Atheist',
403 'uploader_id': 'TheAmazingAtheist',
404 'title': 'Burning Everyone\'s Koran',
405 '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',
408 # Normal age-gate video (No vevo, embed allowed)
410 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
414 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
415 'description': 'md5:eca57043abae25130f58f655ad9a7771',
416 'uploader': 'The Witcher',
417 'uploader_id': 'WitcherGame',
418 'upload_date': '20140605',
421 # Age-gate video with encrypted signature
423 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
427 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
428 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
429 'uploader': 'LloydVEVO',
430 'uploader_id': 'LloydVEVO',
431 'upload_date': '20110629',
434 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
436 'url': '__2ABJjxzNo',
440 'upload_date': '20100430',
441 'uploader_id': 'deadmau5',
442 'description': 'md5:12c56784b8032162bb936a5f76d55360',
443 'uploader': 'deadmau5',
444 'title': 'Deadmau5 - Some Chords (HD)',
446 'expected_warnings': [
447 'DASH manifest missing',
450 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
452 'url': 'lqQg6PlCWgI',
456 'upload_date': '20120731',
457 'uploader_id': 'olympic',
458 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
459 'uploader': 'Olympics',
460 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
463 'skip_download': 'requires avconv',
468 def __init__(self, *args, **kwargs):
469 super(YoutubeIE, self).__init__(*args, **kwargs)
470 self._player_cache = {}
472 def report_video_info_webpage_download(self, video_id):
473 """Report attempt to download video info webpage."""
474 self.to_screen('%s: Downloading video info webpage' % video_id)
476 def report_information_extraction(self, video_id):
477 """Report attempt to extract video information."""
478 self.to_screen('%s: Extracting video information' % video_id)
480 def report_unavailable_format(self, video_id, format):
481 """Report extracted video URL."""
482 self.to_screen('%s: Format %s not available' % (video_id, format))
484 def report_rtmp_download(self):
485 """Indicate the download will use the RTMP protocol."""
486 self.to_screen('RTMP download detected')
488 def _signature_cache_id(self, example_sig):
489 """ Return a string representation of a signature """
490 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
492 def _extract_signature_function(self, video_id, player_url, example_sig):
494 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
497 raise ExtractorError('Cannot identify player %r' % player_url)
498 player_type = id_m.group('ext')
499 player_id = id_m.group('id')
501 # Read from filesystem cache
502 func_id = '%s_%s_%s' % (
503 player_type, player_id, self._signature_cache_id(example_sig))
504 assert os.path.basename(func_id) == func_id
506 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
507 if cache_spec is not None:
508 return lambda s: ''.join(s[i] for i in cache_spec)
510 if player_type == 'js':
511 code = self._download_webpage(
512 player_url, video_id,
513 note='Downloading %s player %s' % (player_type, player_id),
514 errnote='Download of %s failed' % player_url)
515 res = self._parse_sig_js(code)
516 elif player_type == 'swf':
517 urlh = self._request_webpage(
518 player_url, video_id,
519 note='Downloading %s player %s' % (player_type, player_id),
520 errnote='Download of %s failed' % player_url)
522 res = self._parse_sig_swf(code)
524 assert False, 'Invalid player type %r' % player_type
526 if cache_spec is None:
527 test_string = ''.join(map(compat_chr, range(len(example_sig))))
528 cache_res = res(test_string)
529 cache_spec = [ord(c) for c in cache_res]
531 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
534 def _print_sig_code(self, func, example_sig):
535 def gen_sig_code(idxs):
536 def _genslice(start, end, step):
537 starts = '' if start == 0 else str(start)
538 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
539 steps = '' if step == 1 else (':%d' % step)
540 return 's[%s%s%s]' % (starts, ends, steps)
543 # Quelch pyflakes warnings - start will be set when step is set
544 start = '(Never used)'
545 for i, prev in zip(idxs[1:], idxs[:-1]):
549 yield _genslice(start, prev, step)
552 if i - prev in [-1, 1]:
561 yield _genslice(start, i, step)
563 test_string = ''.join(map(compat_chr, range(len(example_sig))))
564 cache_res = func(test_string)
565 cache_spec = [ord(c) for c in cache_res]
566 expr_code = ' + '.join(gen_sig_code(cache_spec))
567 signature_id_tuple = '(%s)' % (
568 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
569 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
570 ' return %s\n') % (signature_id_tuple, expr_code)
571 self.to_screen('Extracted signature function:\n' + code)
573 def _parse_sig_js(self, jscode):
574 funcname = self._search_regex(
575 r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
576 'Initial JS player signature function name')
578 jsi = JSInterpreter(jscode)
579 initial_function = jsi.extract_function(funcname)
580 return lambda s: initial_function([s])
582 def _parse_sig_swf(self, file_contents):
583 swfi = SWFInterpreter(file_contents)
584 TARGET_CLASSNAME = 'SignatureDecipher'
585 searched_class = swfi.extract_class(TARGET_CLASSNAME)
586 initial_function = swfi.extract_function(searched_class, 'decipher')
587 return lambda s: initial_function([s])
589 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
590 """Turn the encrypted s field into a working signature"""
592 if player_url is None:
593 raise ExtractorError('Cannot decrypt signature without player_url')
595 if player_url.startswith('//'):
596 player_url = 'https:' + player_url
598 player_id = (player_url, self._signature_cache_id(s))
599 if player_id not in self._player_cache:
600 func = self._extract_signature_function(
601 video_id, player_url, s
603 self._player_cache[player_id] = func
604 func = self._player_cache[player_id]
605 if self._downloader.params.get('youtube_print_sig_code'):
606 self._print_sig_code(func, s)
608 except Exception as e:
609 tb = traceback.format_exc()
610 raise ExtractorError(
611 'Signature extraction failed: ' + tb, cause=e)
613 def _get_available_subtitles(self, video_id, webpage):
615 subs_doc = self._download_xml(
616 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
617 video_id, note=False)
618 except ExtractorError as err:
619 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
623 for track in subs_doc.findall('track'):
624 lang = track.attrib['lang_code']
625 if lang in sub_lang_list:
627 params = compat_urllib_parse.urlencode({
630 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
631 'name': track.attrib['name'].encode('utf-8'),
633 url = 'https://www.youtube.com/api/timedtext?' + params
634 sub_lang_list[lang] = url
635 if not sub_lang_list:
636 self._downloader.report_warning('video doesn\'t have subtitles')
640 def _get_available_automatic_caption(self, video_id, webpage):
641 """We need the webpage for getting the captions url, pass it as an
642 argument to speed up the process."""
643 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
644 self.to_screen('%s: Looking for automatic captions' % video_id)
645 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
646 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
648 self._downloader.report_warning(err_msg)
650 player_config = json.loads(mobj.group(1))
652 args = player_config['args']
653 caption_url = args['ttsurl']
654 timestamp = args['timestamp']
655 # We get the available subtitles
656 list_params = compat_urllib_parse.urlencode({
661 list_url = caption_url + '&' + list_params
662 caption_list = self._download_xml(list_url, video_id)
663 original_lang_node = caption_list.find('track')
664 if original_lang_node is None:
665 self._downloader.report_warning('Video doesn\'t have automatic captions')
667 original_lang = original_lang_node.attrib['lang_code']
668 caption_kind = original_lang_node.attrib.get('kind', '')
671 for lang_node in caption_list.findall('target'):
672 sub_lang = lang_node.attrib['lang_code']
673 params = compat_urllib_parse.urlencode({
674 'lang': original_lang,
678 'kind': caption_kind,
680 sub_lang_list[sub_lang] = caption_url + '&' + params
682 # An extractor error can be raise by the download process if there are
683 # no automatic captions but there are subtitles
684 except (KeyError, ExtractorError):
685 self._downloader.report_warning(err_msg)
689 def extract_id(cls, url):
690 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
692 raise ExtractorError('Invalid URL: %s' % url)
693 video_id = mobj.group(2)
696 def _extract_from_m3u8(self, manifest_url, video_id):
699 def _get_urls(_manifest):
700 lines = _manifest.split('\n')
701 urls = filter(lambda l: l and not l.startswith('#'),
704 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
705 formats_urls = _get_urls(manifest)
706 for format_url in formats_urls:
707 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
708 url_map[itag] = format_url
711 def _extract_annotations(self, video_id):
712 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
713 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
715 def _parse_dash_manifest(
716 self, video_id, dash_manifest_url, player_url, age_gate):
717 def decrypt_sig(mobj):
719 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
720 return '/signature/%s' % dec_s
721 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
722 dash_doc = self._download_xml(
723 dash_manifest_url, video_id,
724 note='Downloading DASH manifest',
725 errnote='Could not download DASH manifest')
728 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
729 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
732 format_id = r.attrib['id']
733 video_url = url_el.text
734 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
736 'format_id': format_id,
738 'width': int_or_none(r.attrib.get('width')),
739 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
740 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
741 'filesize': filesize,
742 'fps': int_or_none(r.attrib.get('frameRate')),
745 existing_format = next(
747 if fo['format_id'] == format_id)
748 except StopIteration:
749 f.update(self._formats.get(format_id, {}))
752 existing_format.update(f)
755 def _real_extract(self, url):
757 'http' if self._downloader.params.get('prefer_insecure', False)
760 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
761 mobj = re.search(self._NEXT_URL_RE, url)
763 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
764 video_id = self.extract_id(url)
767 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
768 video_webpage = self._download_webpage(url, video_id)
770 # Attempt to extract SWF player URL
771 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
773 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
778 if re.search(r'player-age-gate-content">', video_webpage) is not None:
780 # We simulate the access to the video from www.youtube.com/v/{video_id}
781 # this can be viewed without login into Youtube
782 url = proto + '://www.youtube.com/embed/%s' % video_id
783 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
784 data = compat_urllib_parse.urlencode({
785 'video_id': video_id,
786 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
787 'sts': self._search_regex(
788 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
790 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
791 video_info_webpage = self._download_webpage(
792 video_info_url, video_id,
793 note='Refetching age-gated info webpage',
794 errnote='unable to download video info webpage')
795 video_info = compat_parse_qs(video_info_webpage)
799 # Try looking directly into the video webpage
800 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
802 raise ValueError('Could not find ytplayer.config') # caught below
803 json_code = uppercase_escape(mobj.group(1))
804 ytplayer_config = json.loads(json_code)
805 args = ytplayer_config['args']
806 # Convert to the same format returned by compat_parse_qs
807 video_info = dict((k, [v]) for k, v in args.items())
808 if 'url_encoded_fmt_stream_map' not in args:
809 raise ValueError('No stream_map present') # caught below
811 # We fallback to the get_video_info pages (used by the embed page)
812 self.report_video_info_webpage_download(video_id)
813 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
815 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
816 % (proto, video_id, el_type))
817 video_info_webpage = self._download_webpage(
819 video_id, note=False,
820 errnote='unable to download video info webpage')
821 video_info = compat_parse_qs(video_info_webpage)
822 if 'token' in video_info:
824 if 'token' not in video_info:
825 if 'reason' in video_info:
826 raise ExtractorError(
827 'YouTube said: %s' % video_info['reason'][0],
828 expected=True, video_id=video_id)
830 raise ExtractorError(
831 '"token" parameter not in video info for unknown reason',
834 if 'view_count' in video_info:
835 view_count = int(video_info['view_count'][0])
839 # Check for "rental" videos
840 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
841 raise ExtractorError('"rental" videos not supported')
843 # Start extracting information
844 self.report_information_extraction(video_id)
847 if 'author' not in video_info:
848 raise ExtractorError('Unable to extract uploader name')
849 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
852 video_uploader_id = None
853 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
855 video_uploader_id = mobj.group(1)
857 self._downloader.report_warning('unable to extract uploader nickname')
860 if 'title' in video_info:
861 video_title = video_info['title'][0]
863 self._downloader.report_warning('Unable to extract video title')
867 # We try first to get a high quality image:
868 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
869 video_webpage, re.DOTALL)
870 if m_thumb is not None:
871 video_thumbnail = m_thumb.group(1)
872 elif 'thumbnail_url' not in video_info:
873 self._downloader.report_warning('unable to extract video thumbnail')
874 video_thumbnail = None
875 else: # don't panic if we can't find it
876 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
880 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
883 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
886 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
887 upload_date = unified_strdate(upload_date)
889 m_cat_container = self._search_regex(
890 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
891 video_webpage, 'categories', default=None)
893 category = self._html_search_regex(
894 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
896 video_categories = None if category is None else [category]
898 video_categories = None
901 video_description = get_element_by_id("eow-description", video_webpage)
902 if video_description:
903 video_description = re.sub(r'''(?x)
905 (?:[a-zA-Z-]+="[^"]+"\s+)*?
907 (?:[a-zA-Z-]+="[^"]+"\s+)*?
908 class="yt-uix-redirect-link"\s*>
911 ''', r'\1', video_description)
912 video_description = clean_html(video_description)
914 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
916 video_description = unescapeHTML(fd_mobj.group(1))
918 video_description = ''
920 def _extract_count(count_name):
921 count = self._search_regex(
922 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
923 video_webpage, count_name, default=None)
924 if count is not None:
925 return int(count.replace(',', ''))
927 like_count = _extract_count('like')
928 dislike_count = _extract_count('dislike')
931 video_subtitles = self.extract_subtitles(video_id, video_webpage)
933 if self._downloader.params.get('listsubtitles', False):
934 self._list_available_subtitles(video_id, video_webpage)
937 if 'length_seconds' not in video_info:
938 self._downloader.report_warning('unable to extract video duration')
939 video_duration = None
941 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
944 video_annotations = None
945 if self._downloader.params.get('writeannotations', False):
946 video_annotations = self._extract_annotations(video_id)
948 def _map_to_format_list(urlmap):
950 for itag, video_real_url in urlmap.items():
953 'url': video_real_url,
954 'player_url': player_url,
956 if itag in self._formats:
957 dct.update(self._formats[itag])
961 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
962 self.report_rtmp_download()
964 'format_id': '_rtmp',
966 'url': video_info['conn'][0],
967 'player_url': player_url,
969 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
970 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
971 if 'rtmpe%3Dyes' in encoded_url_map:
972 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
974 for url_data_str in encoded_url_map.split(','):
975 url_data = compat_parse_qs(url_data_str)
976 if 'itag' not in url_data or 'url' not in url_data:
978 format_id = url_data['itag'][0]
979 url = url_data['url'][0]
981 if 'sig' in url_data:
982 url += '&signature=' + url_data['sig'][0]
983 elif 's' in url_data:
984 encrypted_sig = url_data['s'][0]
986 jsplayer_url_json = self._search_regex(
987 r'"assets":.+?"js":\s*("[^"]+")',
988 embed_webpage if age_gate else video_webpage, 'JS player URL')
989 player_url = json.loads(jsplayer_url_json)
990 if player_url is None:
991 player_url_json = self._search_regex(
992 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
993 video_webpage, 'age gate player URL')
994 player_url = json.loads(player_url_json)
996 if self._downloader.params.get('verbose'):
997 if player_url is None:
998 player_version = 'unknown'
999 player_desc = 'unknown'
1001 if player_url.endswith('swf'):
1002 player_version = self._search_regex(
1003 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1004 'flash player', fatal=False)
1005 player_desc = 'flash player %s' % player_version
1007 player_version = self._search_regex(
1008 r'html5player-([^/]+?)(?:/html5player)?\.js',
1010 'html5 player', fatal=False)
1011 player_desc = 'html5 player %s' % player_version
1013 parts_sizes = self._signature_cache_id(encrypted_sig)
1014 self.to_screen('{%s} signature length %s, %s' %
1015 (format_id, parts_sizes, player_desc))
1017 signature = self._decrypt_signature(
1018 encrypted_sig, video_id, player_url, age_gate)
1019 url += '&signature=' + signature
1020 if 'ratebypass' not in url:
1021 url += '&ratebypass=yes'
1022 url_map[format_id] = url
1023 formats = _map_to_format_list(url_map)
1024 elif video_info.get('hlsvp'):
1025 manifest_url = video_info['hlsvp'][0]
1026 url_map = self._extract_from_m3u8(manifest_url, video_id)
1027 formats = _map_to_format_list(url_map)
1029 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1031 # Look for the DASH manifest
1032 if self._downloader.params.get('youtube_include_dash_manifest', True):
1033 dash_mpd = video_info.get('dashmpd')
1035 dash_manifest_url = dash_mpd[0]
1037 dash_formats = self._parse_dash_manifest(
1038 video_id, dash_manifest_url, player_url, age_gate)
1039 except (ExtractorError, KeyError) as e:
1040 self.report_warning(
1041 'Skipping DASH manifest: %r' % e, video_id)
1043 formats.extend(dash_formats)
1045 self._sort_formats(formats)
1049 'uploader': video_uploader,
1050 'uploader_id': video_uploader_id,
1051 'upload_date': upload_date,
1052 'title': video_title,
1053 'thumbnail': video_thumbnail,
1054 'description': video_description,
1055 'categories': video_categories,
1056 'subtitles': video_subtitles,
1057 'duration': video_duration,
1058 'age_limit': 18 if age_gate else 0,
1059 'annotations': video_annotations,
1060 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1061 'view_count': view_count,
1062 'like_count': like_count,
1063 'dislike_count': dislike_count,
1068 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1069 IE_DESC = 'YouTube.com playlists'
1070 _VALID_URL = r"""(?x)(?:
1075 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1076 \? (?:.*?&)*? (?:p|a|list)=
1080 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1081 # Top tracks, they can also include dots
1086 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1088 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1089 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1090 IE_NAME = 'youtube:playlist'
1092 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1094 'title': 'ytdl test PL',
1095 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1097 'playlist_count': 3,
1099 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1101 'title': 'YDL_Empty_List',
1103 'playlist_count': 0,
1105 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1106 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1108 'title': '29C3: Not my department',
1110 'playlist_count': 95,
1112 'note': 'issue #673',
1113 'url': 'PLBB231211A4F62143',
1115 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1117 'playlist_mincount': 26,
1119 'note': 'Large playlist',
1120 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1122 'title': 'Uploads from Cauchemar',
1124 'playlist_mincount': 799,
1126 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1128 'title': 'YDL_safe_search',
1130 'playlist_count': 2,
1133 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1134 'playlist_count': 4,
1139 'note': 'Embedded SWF player',
1140 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1141 'playlist_count': 4,
1146 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1147 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1149 'title': 'Uploads from Interstellar Movie',
1151 'playlist_mincout': 21,
1154 def _real_initialize(self):
1157 def _ids_to_results(self, ids):
1159 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1162 def _extract_mix(self, playlist_id):
1163 # The mixes are generated from a a single video
1164 # the id of the playlist is just 'RD' + video_id
1165 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1166 webpage = self._download_webpage(
1167 url, playlist_id, 'Downloading Youtube mix')
1168 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1170 search_title('playlist-title') or
1171 search_title('title long-title') or
1172 search_title('title'))
1173 title = clean_html(title_span)
1174 ids = orderedSet(re.findall(
1175 r'''(?xs)data-video-username=".*?".*?
1176 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1178 url_results = self._ids_to_results(ids)
1180 return self.playlist_result(url_results, playlist_id, title)
1182 def _real_extract(self, url):
1183 # Extract playlist id
1184 mobj = re.match(self._VALID_URL, url)
1186 raise ExtractorError('Invalid URL: %s' % url)
1187 playlist_id = mobj.group(1) or mobj.group(2)
1189 # Check if it's a video-specific URL
1190 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1191 if 'v' in query_dict:
1192 video_id = query_dict['v'][0]
1193 if self._downloader.params.get('noplaylist'):
1194 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1195 return self.url_result(video_id, 'Youtube', video_id=video_id)
1197 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1199 if playlist_id.startswith('RD'):
1200 # Mixes require a custom extraction process
1201 return self._extract_mix(playlist_id)
1202 if playlist_id.startswith('TL'):
1203 raise ExtractorError('For downloading YouTube.com top lists, use '
1204 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1206 url = self._TEMPLATE_URL % playlist_id
1207 page = self._download_webpage(url, playlist_id)
1208 more_widget_html = content_html = page
1210 # Check if the playlist exists or is private
1211 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1212 raise ExtractorError(
1213 'The playlist doesn\'t exist or is private, use --username or '
1214 '--netrc to access it.',
1217 # Extract the video ids from the playlist pages
1220 for page_num in itertools.count(1):
1221 matches = re.finditer(self._VIDEO_RE, content_html)
1222 # We remove the duplicates and the link with index 0
1223 # (it's not the first video of the playlist)
1224 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1227 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1231 more = self._download_json(
1232 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1233 'Downloading page #%s' % page_num,
1234 transform_source=uppercase_escape)
1235 content_html = more['content_html']
1236 if not content_html.strip():
1237 # Some webpages show a "Load more" button but they don't
1240 more_widget_html = more['load_more_widget_html']
1242 playlist_title = self._html_search_regex(
1243 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1246 url_results = self._ids_to_results(ids)
1247 return self.playlist_result(url_results, playlist_id, playlist_title)
1250 class YoutubeTopListIE(YoutubePlaylistIE):
1251 IE_NAME = 'youtube:toplist'
1252 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1253 ' (Example: "yttoplist:music:Top Tracks")')
1254 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1256 'url': 'yttoplist:music:Trending',
1257 'playlist_mincount': 5,
1258 'skip': 'Only works for logged-in users',
1261 def _real_extract(self, url):
1262 mobj = re.match(self._VALID_URL, url)
1263 channel = mobj.group('chann')
1264 title = mobj.group('title')
1265 query = compat_urllib_parse.urlencode({'title': title})
1266 channel_page = self._download_webpage(
1267 'https://www.youtube.com/%s' % channel, title)
1268 link = self._html_search_regex(
1270 <a\s+href="([^"]+)".*?>\s*
1271 <span\s+class="branded-page-module-title-text">\s*
1272 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1273 channel_page, 'list')
1274 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1276 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1278 # sometimes the webpage doesn't contain the videos
1279 # retry until we get them
1280 for i in itertools.count(0):
1281 msg = 'Downloading Youtube mix'
1283 msg += ', retry #%d' % i
1285 webpage = self._download_webpage(url, title, msg)
1286 ids = orderedSet(re.findall(video_re, webpage))
1289 url_results = self._ids_to_results(ids)
1290 return self.playlist_result(url_results, playlist_title=title)
1293 class YoutubeChannelIE(InfoExtractor):
1294 IE_DESC = 'YouTube.com channels'
1295 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1296 IE_NAME = 'youtube:channel'
1298 'note': 'paginated channel',
1299 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1300 'playlist_mincount': 91,
1303 def extract_videos_from_page(self, page):
1305 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1306 if mobj.group(1) not in ids_in_page:
1307 ids_in_page.append(mobj.group(1))
1310 def _real_extract(self, url):
1311 channel_id = self._match_id(url)
1314 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1315 channel_page = self._download_webpage(url, channel_id)
1316 autogenerated = re.search(r'''(?x)
1318 channel-header-autogenerated-label|
1319 yt-channel-title-autogenerated
1320 )[^"]*"''', channel_page) is not None
1323 # The videos are contained in a single page
1324 # the ajax pages can't be used, they are empty
1325 video_ids = self.extract_videos_from_page(channel_page)
1327 self.url_result(video_id, 'Youtube', video_id=video_id)
1328 for video_id in video_ids]
1329 return self.playlist_result(entries, channel_id)
1332 more_widget_html = content_html = channel_page
1333 for pagenum in itertools.count(1):
1335 ids_in_page = self.extract_videos_from_page(content_html)
1336 for video_id in ids_in_page:
1337 yield self.url_result(
1338 video_id, 'Youtube', video_id=video_id)
1341 r'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1346 more = self._download_json(
1347 'https://youtube.com/%s' % mobj.group('more'), channel_id,
1348 'Downloading page #%s' % (pagenum + 1),
1349 transform_source=uppercase_escape)
1350 content_html = more['content_html']
1351 more_widget_html = more['load_more_widget_html']
1353 return self.playlist_result(_entries(), channel_id)
1356 class YoutubeUserIE(InfoExtractor):
1357 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1358 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1359 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1360 _GDATA_PAGE_SIZE = 50
1361 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1362 IE_NAME = 'youtube:user'
1365 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1366 'playlist_mincount': 320,
1368 'title': 'TheLinuxFoundation',
1371 'url': 'ytuser:phihag',
1372 'only_matching': True,
1376 def suitable(cls, url):
1377 # Don't return True if the url can be extracted with other youtube
1378 # extractor, the regex would is too permissive and it would match.
1379 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1380 if any(ie.suitable(url) for ie in other_ies):
1383 return super(YoutubeUserIE, cls).suitable(url)
1385 def _real_extract(self, url):
1386 username = self._match_id(url)
1388 # Download video ids using YouTube Data API. Result size per
1389 # query is limited (currently to 50 videos) so we need to query
1390 # page by page until there are no video ids - it means we got
1393 def download_page(pagenum):
1394 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1396 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1397 page = self._download_webpage(
1398 gdata_url, username,
1399 'Downloading video ids from %d to %d' % (
1400 start_index, start_index + self._GDATA_PAGE_SIZE))
1403 response = json.loads(page)
1404 except ValueError as err:
1405 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1406 if 'entry' not in response['feed']:
1409 # Extract video identifiers
1410 entries = response['feed']['entry']
1411 for entry in entries:
1412 title = entry['title']['$t']
1413 video_id = entry['id']['$t'].split('/')[-1]
1417 'ie_key': 'Youtube',
1421 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1423 return self.playlist_result(url_results, playlist_title=username)
1426 class YoutubeSearchIE(SearchInfoExtractor):
1427 IE_DESC = 'YouTube.com searches'
1428 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1430 IE_NAME = 'youtube:search'
1431 _SEARCH_KEY = 'ytsearch'
1433 def _get_n_results(self, query, n):
1434 """Get a specified number of results for a query"""
1441 while (PAGE_SIZE * pagenum) < limit:
1442 result_url = self._API_URL % (
1443 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1444 (PAGE_SIZE * pagenum) + 1)
1445 data_json = self._download_webpage(
1446 result_url, video_id='query "%s"' % query,
1447 note='Downloading page %s' % (pagenum + 1),
1448 errnote='Unable to download API page')
1449 data = json.loads(data_json)
1450 api_response = data['data']
1452 if 'items' not in api_response:
1453 raise ExtractorError(
1454 '[youtube] No video results', expected=True)
1456 new_ids = list(video['id'] for video in api_response['items'])
1457 video_ids += new_ids
1459 limit = min(n, api_response['totalItems'])
1462 if len(video_ids) > n:
1463 video_ids = video_ids[:n]
1464 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1465 for video_id in video_ids]
1466 return self.playlist_result(videos, query)
1469 class YoutubeSearchDateIE(YoutubeSearchIE):
1470 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1471 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1472 _SEARCH_KEY = 'ytsearchdate'
1473 IE_DESC = 'YouTube.com searches, newest videos first'
1476 class YoutubeSearchURLIE(InfoExtractor):
1477 IE_DESC = 'YouTube.com search URLs'
1478 IE_NAME = 'youtube:search_url'
1479 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1481 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1482 'playlist_mincount': 5,
1484 'title': 'youtube-dl test video',
1488 def _real_extract(self, url):
1489 mobj = re.match(self._VALID_URL, url)
1490 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1492 webpage = self._download_webpage(url, query)
1493 result_code = self._search_regex(
1494 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1496 part_codes = re.findall(
1497 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1499 for part_code in part_codes:
1500 part_title = self._html_search_regex(
1501 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1502 part_url_snippet = self._html_search_regex(
1503 r'(?s)href="([^"]+)"', part_code, 'item URL')
1504 part_url = compat_urlparse.urljoin(
1505 'https://www.youtube.com/', part_url_snippet)
1509 'title': part_title,
1513 '_type': 'playlist',
1519 class YoutubeShowIE(InfoExtractor):
1520 IE_DESC = 'YouTube.com (multi-season) shows'
1521 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1522 IE_NAME = 'youtube:show'
1524 'url': 'http://www.youtube.com/show/airdisasters',
1525 'playlist_mincount': 3,
1527 'id': 'airdisasters',
1528 'title': 'Air Disasters',
1532 def _real_extract(self, url):
1533 mobj = re.match(self._VALID_URL, url)
1534 playlist_id = mobj.group('id')
1535 webpage = self._download_webpage(
1536 url, playlist_id, 'Downloading show webpage')
1537 # There's one playlist for each season of the show
1538 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1539 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1542 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1543 for season in m_seasons
1545 title = self._og_search_title(webpage, fatal=False)
1548 '_type': 'playlist',
1555 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1557 Base class for extractors that fetch info from
1558 http://www.youtube.com/feed_ajax
1559 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1561 _LOGIN_REQUIRED = True
1562 # use action_load_personal_feed instead of action_load_system_feed
1563 _PERSONAL_FEED = False
1566 def _FEED_TEMPLATE(self):
1567 action = 'action_load_system_feed'
1568 if self._PERSONAL_FEED:
1569 action = 'action_load_personal_feed'
1570 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1574 return 'youtube:%s' % self._FEED_NAME
1576 def _real_initialize(self):
1579 def _real_extract(self, url):
1582 for i in itertools.count(1):
1583 info = self._download_json(
1584 self._FEED_TEMPLATE % paging,
1585 '%s feed' % self._FEED_NAME,
1586 'Downloading page %s' % i,
1587 transform_source=uppercase_escape)
1588 feed_html = info.get('feed_html') or info.get('content_html')
1589 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1590 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1591 ids = orderedSet(m.group(1) for m in m_ids)
1592 feed_entries.extend(
1593 self.url_result(video_id, 'Youtube', video_id=video_id)
1594 for video_id in ids)
1596 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1597 load_more_widget_html)
1600 paging = mobj.group('paging')
1601 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1604 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1605 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1606 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1607 _FEED_NAME = 'recommended'
1608 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1611 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1612 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1613 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1614 _FEED_NAME = 'watch_later'
1615 _PLAYLIST_TITLE = 'Youtube Watch Later'
1616 _PERSONAL_FEED = True
1619 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1620 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1621 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1622 _FEED_NAME = 'history'
1623 _PERSONAL_FEED = True
1624 _PLAYLIST_TITLE = 'Youtube Watch History'
1627 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1628 IE_NAME = 'youtube:favorites'
1629 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1630 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1631 _LOGIN_REQUIRED = True
1633 def _real_extract(self, url):
1634 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1635 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1636 return self.url_result(playlist_id, 'YoutubePlaylist')
1639 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1640 IE_NAME = 'youtube:subscriptions'
1641 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1642 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1645 def _real_extract(self, url):
1646 title = 'Youtube Subscriptions'
1647 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1649 # The extraction process is the same as for playlists, but the regex
1650 # for the video ids doesn't contain an index
1652 more_widget_html = content_html = page
1654 for page_num in itertools.count(1):
1655 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1656 new_ids = orderedSet(matches)
1659 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1663 more = self._download_json(
1664 'https://youtube.com/%s' % mobj.group('more'), title,
1665 'Downloading page #%s' % page_num,
1666 transform_source=uppercase_escape)
1667 content_html = more['content_html']
1668 more_widget_html = more['load_more_widget_html']
1671 '_type': 'playlist',
1673 'entries': self._ids_to_results(ids),
1677 class YoutubeTruncatedURLIE(InfoExtractor):
1678 IE_NAME = 'youtube:truncated_url'
1679 IE_DESC = False # Do not list
1680 _VALID_URL = r'''(?x)
1681 (?:https?://)?[^/]+/watch\?(?:
1683 annotation_id=annotation_[^&]+
1685 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1689 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1690 'only_matching': True,
1692 'url': 'http://www.youtube.com/watch?',
1693 'only_matching': True,
1696 def _real_extract(self, url):
1697 raise ExtractorError(
1698 'Did you forget to quote the URL? Remember that & is a meta '
1699 'character in most shells, so you want to put the URL in quotes, '
1701 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1702 ' or simply youtube-dl BaW_jenozKc .',
1706 class YoutubeTruncatedIDIE(InfoExtractor):
1707 IE_NAME = 'youtube:truncated_id'
1708 IE_DESC = False # Do not list
1709 _VALID_URL = r'https?://(?:www\.)youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1712 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1713 'only_matching': True,
1716 def _real_extract(self, url):
1717 video_id = self._match_id(url)
1718 raise ExtractorError(
1719 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),