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 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
423 'url': '__2ABJjxzNo',
427 'upload_date': '20100430',
428 'uploader_id': 'deadmau5',
429 'description': 'md5:12c56784b8032162bb936a5f76d55360',
430 'uploader': 'deadmau5',
431 'title': 'Deadmau5 - Some Chords (HD)',
433 'expected_warnings': [
434 'DASH manifest missing',
437 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
439 'url': 'lqQg6PlCWgI',
443 'upload_date': '20120731',
444 'uploader_id': 'olympic',
445 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
446 'uploader': 'Olympics',
447 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
450 'skip_download': 'requires avconv',
455 def __init__(self, *args, **kwargs):
456 super(YoutubeIE, self).__init__(*args, **kwargs)
457 self._player_cache = {}
459 def report_video_info_webpage_download(self, video_id):
460 """Report attempt to download video info webpage."""
461 self.to_screen('%s: Downloading video info webpage' % video_id)
463 def report_information_extraction(self, video_id):
464 """Report attempt to extract video information."""
465 self.to_screen('%s: Extracting video information' % video_id)
467 def report_unavailable_format(self, video_id, format):
468 """Report extracted video URL."""
469 self.to_screen('%s: Format %s not available' % (video_id, format))
471 def report_rtmp_download(self):
472 """Indicate the download will use the RTMP protocol."""
473 self.to_screen('RTMP download detected')
475 def _signature_cache_id(self, example_sig):
476 """ Return a string representation of a signature """
477 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
479 def _extract_signature_function(self, video_id, player_url, example_sig):
481 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
484 raise ExtractorError('Cannot identify player %r' % player_url)
485 player_type = id_m.group('ext')
486 player_id = id_m.group('id')
488 # Read from filesystem cache
489 func_id = '%s_%s_%s' % (
490 player_type, player_id, self._signature_cache_id(example_sig))
491 assert os.path.basename(func_id) == func_id
493 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
494 if cache_spec is not None:
495 return lambda s: ''.join(s[i] for i in cache_spec)
497 if player_type == 'js':
498 code = self._download_webpage(
499 player_url, video_id,
500 note='Downloading %s player %s' % (player_type, player_id),
501 errnote='Download of %s failed' % player_url)
502 res = self._parse_sig_js(code)
503 elif player_type == 'swf':
504 urlh = self._request_webpage(
505 player_url, video_id,
506 note='Downloading %s player %s' % (player_type, player_id),
507 errnote='Download of %s failed' % player_url)
509 res = self._parse_sig_swf(code)
511 assert False, 'Invalid player type %r' % player_type
513 if cache_spec is None:
514 test_string = ''.join(map(compat_chr, range(len(example_sig))))
515 cache_res = res(test_string)
516 cache_spec = [ord(c) for c in cache_res]
518 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
521 def _print_sig_code(self, func, example_sig):
522 def gen_sig_code(idxs):
523 def _genslice(start, end, step):
524 starts = '' if start == 0 else str(start)
525 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
526 steps = '' if step == 1 else (':%d' % step)
527 return 's[%s%s%s]' % (starts, ends, steps)
530 start = '(Never used)' # Quelch pyflakes warnings - start will be
531 # set as soon as step is set
532 for i, prev in zip(idxs[1:], idxs[:-1]):
536 yield _genslice(start, prev, step)
539 if i - prev in [-1, 1]:
548 yield _genslice(start, i, step)
550 test_string = ''.join(map(compat_chr, range(len(example_sig))))
551 cache_res = func(test_string)
552 cache_spec = [ord(c) for c in cache_res]
553 expr_code = ' + '.join(gen_sig_code(cache_spec))
554 signature_id_tuple = '(%s)' % (
555 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
556 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
557 ' return %s\n') % (signature_id_tuple, expr_code)
558 self.to_screen('Extracted signature function:\n' + code)
560 def _parse_sig_js(self, jscode):
561 funcname = self._search_regex(
562 r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
563 'Initial JS player signature function name')
565 jsi = JSInterpreter(jscode)
566 initial_function = jsi.extract_function(funcname)
567 return lambda s: initial_function([s])
569 def _parse_sig_swf(self, file_contents):
570 swfi = SWFInterpreter(file_contents)
571 TARGET_CLASSNAME = 'SignatureDecipher'
572 searched_class = swfi.extract_class(TARGET_CLASSNAME)
573 initial_function = swfi.extract_function(searched_class, 'decipher')
574 return lambda s: initial_function([s])
576 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
577 """Turn the encrypted s field into a working signature"""
579 if player_url is None:
580 raise ExtractorError('Cannot decrypt signature without player_url')
582 if player_url.startswith('//'):
583 player_url = 'https:' + player_url
585 player_id = (player_url, self._signature_cache_id(s))
586 if player_id not in self._player_cache:
587 func = self._extract_signature_function(
588 video_id, player_url, s
590 self._player_cache[player_id] = func
591 func = self._player_cache[player_id]
592 if self._downloader.params.get('youtube_print_sig_code'):
593 self._print_sig_code(func, s)
595 except Exception as e:
596 tb = traceback.format_exc()
597 raise ExtractorError(
598 'Signature extraction failed: ' + tb, cause=e)
600 def _get_available_subtitles(self, video_id, webpage):
602 sub_list = self._download_webpage(
603 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
604 video_id, note=False)
605 except ExtractorError as err:
606 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
608 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
613 if lang in sub_lang_list:
615 params = compat_urllib_parse.urlencode({
618 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
619 'name': unescapeHTML(l[0]).encode('utf-8'),
621 url = 'https://www.youtube.com/api/timedtext?' + params
622 sub_lang_list[lang] = url
623 if not sub_lang_list:
624 self._downloader.report_warning('video doesn\'t have subtitles')
628 def _get_available_automatic_caption(self, video_id, webpage):
629 """We need the webpage for getting the captions url, pass it as an
630 argument to speed up the process."""
631 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
632 self.to_screen('%s: Looking for automatic captions' % video_id)
633 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
634 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
636 self._downloader.report_warning(err_msg)
638 player_config = json.loads(mobj.group(1))
640 args = player_config['args']
641 caption_url = args['ttsurl']
642 timestamp = args['timestamp']
643 # We get the available subtitles
644 list_params = compat_urllib_parse.urlencode({
649 list_url = caption_url + '&' + list_params
650 caption_list = self._download_xml(list_url, video_id)
651 original_lang_node = caption_list.find('track')
652 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr':
653 self._downloader.report_warning('Video doesn\'t have automatic captions')
655 original_lang = original_lang_node.attrib['lang_code']
658 for lang_node in caption_list.findall('target'):
659 sub_lang = lang_node.attrib['lang_code']
660 params = compat_urllib_parse.urlencode({
661 'lang': original_lang,
667 sub_lang_list[sub_lang] = caption_url + '&' + params
669 # An extractor error can be raise by the download process if there are
670 # no automatic captions but there are subtitles
671 except (KeyError, ExtractorError):
672 self._downloader.report_warning(err_msg)
676 def extract_id(cls, url):
677 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
679 raise ExtractorError('Invalid URL: %s' % url)
680 video_id = mobj.group(2)
683 def _extract_from_m3u8(self, manifest_url, video_id):
686 def _get_urls(_manifest):
687 lines = _manifest.split('\n')
688 urls = filter(lambda l: l and not l.startswith('#'),
691 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
692 formats_urls = _get_urls(manifest)
693 for format_url in formats_urls:
694 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
695 url_map[itag] = format_url
698 def _extract_annotations(self, video_id):
699 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
700 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
702 def _parse_dash_manifest(
703 self, video_id, dash_manifest_url, player_url, age_gate):
704 def decrypt_sig(mobj):
706 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
707 return '/signature/%s' % dec_s
708 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
709 dash_doc = self._download_xml(
710 dash_manifest_url, video_id,
711 note='Downloading DASH manifest',
712 errnote='Could not download DASH manifest')
715 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
716 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
719 format_id = r.attrib['id']
720 video_url = url_el.text
721 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
723 'format_id': format_id,
725 'width': int_or_none(r.attrib.get('width')),
726 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
727 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
728 'filesize': filesize,
729 'fps': int_or_none(r.attrib.get('frameRate')),
732 existing_format = next(
734 if fo['format_id'] == format_id)
735 except StopIteration:
736 f.update(self._formats.get(format_id, {}))
739 existing_format.update(f)
742 def _real_extract(self, url):
744 'http' if self._downloader.params.get('prefer_insecure', False)
747 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
748 mobj = re.search(self._NEXT_URL_RE, url)
750 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
751 video_id = self.extract_id(url)
754 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
755 video_webpage = self._download_webpage(url, video_id)
757 # Attempt to extract SWF player URL
758 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
760 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
765 if re.search(r'player-age-gate-content">', video_webpage) is not None:
767 # We simulate the access to the video from www.youtube.com/v/{video_id}
768 # this can be viewed without login into Youtube
769 data = compat_urllib_parse.urlencode({
770 'video_id': video_id,
771 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
772 'sts': self._search_regex(
773 r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
775 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
776 video_info_webpage = self._download_webpage(
777 video_info_url, video_id,
778 note='Refetching age-gated info webpage',
779 errnote='unable to download video info webpage')
780 video_info = compat_parse_qs(video_info_webpage)
784 # Try looking directly into the video webpage
785 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
787 raise ValueError('Could not find ytplayer.config') # caught below
788 json_code = uppercase_escape(mobj.group(1))
789 ytplayer_config = json.loads(json_code)
790 args = ytplayer_config['args']
791 # Convert to the same format returned by compat_parse_qs
792 video_info = dict((k, [v]) for k, v in args.items())
793 if 'url_encoded_fmt_stream_map' not in args:
794 raise ValueError('No stream_map present') # caught below
796 # We fallback to the get_video_info pages (used by the embed page)
797 self.report_video_info_webpage_download(video_id)
798 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
800 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
801 % (proto, video_id, el_type))
802 video_info_webpage = self._download_webpage(
804 video_id, note=False,
805 errnote='unable to download video info webpage')
806 video_info = compat_parse_qs(video_info_webpage)
807 if 'token' in video_info:
809 if 'token' not in video_info:
810 if 'reason' in video_info:
811 raise ExtractorError(
812 'YouTube said: %s' % video_info['reason'][0],
813 expected=True, video_id=video_id)
815 raise ExtractorError(
816 '"token" parameter not in video info for unknown reason',
819 if 'view_count' in video_info:
820 view_count = int(video_info['view_count'][0])
824 # Check for "rental" videos
825 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
826 raise ExtractorError('"rental" videos not supported')
828 # Start extracting information
829 self.report_information_extraction(video_id)
832 if 'author' not in video_info:
833 raise ExtractorError('Unable to extract uploader name')
834 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
837 video_uploader_id = None
838 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
840 video_uploader_id = mobj.group(1)
842 self._downloader.report_warning('unable to extract uploader nickname')
845 if 'title' in video_info:
846 video_title = video_info['title'][0]
848 self._downloader.report_warning('Unable to extract video title')
852 # We try first to get a high quality image:
853 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
854 video_webpage, re.DOTALL)
855 if m_thumb is not None:
856 video_thumbnail = m_thumb.group(1)
857 elif 'thumbnail_url' not in video_info:
858 self._downloader.report_warning('unable to extract video thumbnail')
859 video_thumbnail = None
860 else: # don't panic if we can't find it
861 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
865 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
868 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
871 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
872 upload_date = unified_strdate(upload_date)
874 m_cat_container = self._search_regex(
875 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
876 video_webpage, 'categories', default=None)
878 category = self._html_search_regex(
879 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
881 video_categories = None if category is None else [category]
883 video_categories = None
886 video_description = get_element_by_id("eow-description", video_webpage)
887 if video_description:
888 video_description = re.sub(r'''(?x)
890 (?:[a-zA-Z-]+="[^"]+"\s+)*?
892 (?:[a-zA-Z-]+="[^"]+"\s+)*?
893 class="yt-uix-redirect-link"\s*>
896 ''', r'\1', video_description)
897 video_description = clean_html(video_description)
899 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
901 video_description = unescapeHTML(fd_mobj.group(1))
903 video_description = ''
905 def _extract_count(count_name):
906 count = self._search_regex(
907 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
908 video_webpage, count_name, default=None)
909 if count is not None:
910 return int(count.replace(',', ''))
912 like_count = _extract_count('like')
913 dislike_count = _extract_count('dislike')
916 video_subtitles = self.extract_subtitles(video_id, video_webpage)
918 if self._downloader.params.get('listsubtitles', False):
919 self._list_available_subtitles(video_id, video_webpage)
922 if 'length_seconds' not in video_info:
923 self._downloader.report_warning('unable to extract video duration')
924 video_duration = None
926 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
929 video_annotations = None
930 if self._downloader.params.get('writeannotations', False):
931 video_annotations = self._extract_annotations(video_id)
933 def _map_to_format_list(urlmap):
935 for itag, video_real_url in urlmap.items():
938 'url': video_real_url,
939 'player_url': player_url,
941 if itag in self._formats:
942 dct.update(self._formats[itag])
946 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
947 self.report_rtmp_download()
949 'format_id': '_rtmp',
951 'url': video_info['conn'][0],
952 'player_url': player_url,
954 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
955 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
956 if 'rtmpe%3Dyes' in encoded_url_map:
957 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
959 for url_data_str in encoded_url_map.split(','):
960 url_data = compat_parse_qs(url_data_str)
961 if 'itag' not in url_data or 'url' not in url_data:
963 format_id = url_data['itag'][0]
964 url = url_data['url'][0]
966 if 'sig' in url_data:
967 url += '&signature=' + url_data['sig'][0]
968 elif 's' in url_data:
969 encrypted_sig = url_data['s'][0]
972 jsplayer_url_json = self._search_regex(
973 r'"assets":.+?"js":\s*("[^"]+")',
974 video_webpage, 'JS player URL')
975 player_url = json.loads(jsplayer_url_json)
976 if player_url is None:
977 player_url_json = self._search_regex(
978 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
979 video_webpage, 'age gate player URL')
980 player_url = json.loads(player_url_json)
982 if self._downloader.params.get('verbose'):
983 if player_url is None:
984 player_version = 'unknown'
985 player_desc = 'unknown'
987 if player_url.endswith('swf'):
988 player_version = self._search_regex(
989 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
990 'flash player', fatal=False)
991 player_desc = 'flash player %s' % player_version
993 player_version = self._search_regex(
994 r'html5player-([^/]+?)(?:/html5player)?\.js',
996 'html5 player', fatal=False)
997 player_desc = 'html5 player %s' % player_version
999 parts_sizes = self._signature_cache_id(encrypted_sig)
1000 self.to_screen('{%s} signature length %s, %s' %
1001 (format_id, parts_sizes, player_desc))
1003 signature = self._decrypt_signature(
1004 encrypted_sig, video_id, player_url, age_gate)
1005 url += '&signature=' + signature
1006 if 'ratebypass' not in url:
1007 url += '&ratebypass=yes'
1008 url_map[format_id] = url
1009 formats = _map_to_format_list(url_map)
1010 elif video_info.get('hlsvp'):
1011 manifest_url = video_info['hlsvp'][0]
1012 url_map = self._extract_from_m3u8(manifest_url, video_id)
1013 formats = _map_to_format_list(url_map)
1015 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1017 # Look for the DASH manifest
1018 if self._downloader.params.get('youtube_include_dash_manifest', True):
1019 dash_mpd = video_info.get('dashmpd')
1021 dash_manifest_url = dash_mpd[0]
1023 dash_formats = self._parse_dash_manifest(
1024 video_id, dash_manifest_url, player_url, age_gate)
1025 except (ExtractorError, KeyError) as e:
1026 self.report_warning(
1027 'Skipping DASH manifest: %r' % e, video_id)
1029 formats.extend(dash_formats)
1031 self._sort_formats(formats)
1035 'uploader': video_uploader,
1036 'uploader_id': video_uploader_id,
1037 'upload_date': upload_date,
1038 'title': video_title,
1039 'thumbnail': video_thumbnail,
1040 'description': video_description,
1041 'categories': video_categories,
1042 'subtitles': video_subtitles,
1043 'duration': video_duration,
1044 'age_limit': 18 if age_gate else 0,
1045 'annotations': video_annotations,
1046 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1047 'view_count': view_count,
1048 'like_count': like_count,
1049 'dislike_count': dislike_count,
1054 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1055 IE_DESC = 'YouTube.com playlists'
1056 _VALID_URL = r"""(?x)(?:
1061 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1062 \? (?:.*?&)*? (?:p|a|list)=
1066 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1067 # Top tracks, they can also include dots
1072 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1074 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1075 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
1076 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1077 IE_NAME = 'youtube:playlist'
1079 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1081 'title': 'ytdl test PL',
1082 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1084 'playlist_count': 3,
1086 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1088 'title': 'YDL_Empty_List',
1090 'playlist_count': 0,
1092 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1093 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1095 'title': '29C3: Not my department',
1097 'playlist_count': 95,
1099 'note': 'issue #673',
1100 'url': 'PLBB231211A4F62143',
1102 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1104 'playlist_mincount': 26,
1106 'note': 'Large playlist',
1107 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1109 'title': 'Uploads from Cauchemar',
1111 'playlist_mincount': 799,
1113 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1115 'title': 'YDL_safe_search',
1117 'playlist_count': 2,
1120 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1121 'playlist_count': 4,
1126 'note': 'Embedded SWF player',
1127 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1128 'playlist_count': 4,
1134 def _real_initialize(self):
1137 def _ids_to_results(self, ids):
1139 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1142 def _extract_mix(self, playlist_id):
1143 # The mixes are generated from a a single video
1144 # the id of the playlist is just 'RD' + video_id
1145 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1146 webpage = self._download_webpage(
1147 url, playlist_id, 'Downloading Youtube mix')
1148 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1150 search_title('playlist-title') or
1151 search_title('title long-title') or
1152 search_title('title'))
1153 title = clean_html(title_span)
1154 ids = orderedSet(re.findall(
1155 r'''(?xs)data-video-username=".*?".*?
1156 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1158 url_results = self._ids_to_results(ids)
1160 return self.playlist_result(url_results, playlist_id, title)
1162 def _real_extract(self, url):
1163 # Extract playlist id
1164 mobj = re.match(self._VALID_URL, url)
1166 raise ExtractorError('Invalid URL: %s' % url)
1167 playlist_id = mobj.group(1) or mobj.group(2)
1169 # Check if it's a video-specific URL
1170 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1171 if 'v' in query_dict:
1172 video_id = query_dict['v'][0]
1173 if self._downloader.params.get('noplaylist'):
1174 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1175 return self.url_result(video_id, 'Youtube', video_id=video_id)
1177 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1179 if playlist_id.startswith('RD'):
1180 # Mixes require a custom extraction process
1181 return self._extract_mix(playlist_id)
1182 if playlist_id.startswith('TL'):
1183 raise ExtractorError('For downloading YouTube.com top lists, use '
1184 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1186 url = self._TEMPLATE_URL % playlist_id
1187 page = self._download_webpage(url, playlist_id)
1188 more_widget_html = content_html = page
1190 # Check if the playlist exists or is private
1191 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1192 raise ExtractorError(
1193 'The playlist doesn\'t exist or is private, use --username or '
1194 '--netrc to access it.',
1197 # Extract the video ids from the playlist pages
1200 for page_num in itertools.count(1):
1201 matches = re.finditer(self._VIDEO_RE, content_html)
1202 # We remove the duplicates and the link with index 0
1203 # (it's not the first video of the playlist)
1204 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1207 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1211 more = self._download_json(
1212 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1213 'Downloading page #%s' % page_num,
1214 transform_source=uppercase_escape)
1215 content_html = more['content_html']
1216 more_widget_html = more['load_more_widget_html']
1218 playlist_title = self._html_search_regex(
1219 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1222 url_results = self._ids_to_results(ids)
1223 return self.playlist_result(url_results, playlist_id, playlist_title)
1226 class YoutubeTopListIE(YoutubePlaylistIE):
1227 IE_NAME = 'youtube:toplist'
1228 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1229 ' (Example: "yttoplist:music:Top Tracks")')
1230 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1232 'url': 'yttoplist:music:Trending',
1233 'playlist_mincount': 5,
1234 'skip': 'Only works for logged-in users',
1237 def _real_extract(self, url):
1238 mobj = re.match(self._VALID_URL, url)
1239 channel = mobj.group('chann')
1240 title = mobj.group('title')
1241 query = compat_urllib_parse.urlencode({'title': title})
1242 channel_page = self._download_webpage(
1243 'https://www.youtube.com/%s' % channel, title)
1244 link = self._html_search_regex(
1246 <a\s+href="([^"]+)".*?>\s*
1247 <span\s+class="branded-page-module-title-text">\s*
1248 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1249 channel_page, 'list')
1250 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1252 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1254 # sometimes the webpage doesn't contain the videos
1255 # retry until we get them
1256 for i in itertools.count(0):
1257 msg = 'Downloading Youtube mix'
1259 msg += ', retry #%d' % i
1261 webpage = self._download_webpage(url, title, msg)
1262 ids = orderedSet(re.findall(video_re, webpage))
1265 url_results = self._ids_to_results(ids)
1266 return self.playlist_result(url_results, playlist_title=title)
1269 class YoutubeChannelIE(InfoExtractor):
1270 IE_DESC = 'YouTube.com channels'
1271 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1272 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1273 _MORE_PAGES_URL = 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
1274 IE_NAME = 'youtube:channel'
1276 'note': 'paginated channel',
1277 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1278 'playlist_mincount': 91,
1281 def extract_videos_from_page(self, page):
1283 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1284 if mobj.group(1) not in ids_in_page:
1285 ids_in_page.append(mobj.group(1))
1288 def _real_extract(self, url):
1289 channel_id = self._match_id(url)
1292 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1293 channel_page = self._download_webpage(url, channel_id)
1294 autogenerated = re.search(r'''(?x)
1296 channel-header-autogenerated-label|
1297 yt-channel-title-autogenerated
1298 )[^"]*"''', channel_page) is not None
1301 # The videos are contained in a single page
1302 # the ajax pages can't be used, they are empty
1303 video_ids = self.extract_videos_from_page(channel_page)
1305 self.url_result(video_id, 'Youtube', video_id=video_id)
1306 for video_id in video_ids]
1307 return self.playlist_result(entries, channel_id)
1310 for pagenum in itertools.count(1):
1311 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1312 page = self._download_json(
1313 url, channel_id, note='Downloading page #%s' % pagenum,
1314 transform_source=uppercase_escape)
1316 ids_in_page = self.extract_videos_from_page(page['content_html'])
1317 for video_id in ids_in_page:
1318 yield self.url_result(
1319 video_id, 'Youtube', video_id=video_id)
1321 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1324 return self.playlist_result(_entries(), channel_id)
1327 class YoutubeUserIE(InfoExtractor):
1328 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1329 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1330 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1331 _GDATA_PAGE_SIZE = 50
1332 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1333 IE_NAME = 'youtube:user'
1336 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1337 'playlist_mincount': 320,
1339 'title': 'TheLinuxFoundation',
1342 'url': 'ytuser:phihag',
1343 'only_matching': True,
1347 def suitable(cls, url):
1348 # Don't return True if the url can be extracted with other youtube
1349 # extractor, the regex would is too permissive and it would match.
1350 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1351 if any(ie.suitable(url) for ie in other_ies):
1354 return super(YoutubeUserIE, cls).suitable(url)
1356 def _real_extract(self, url):
1357 username = self._match_id(url)
1359 # Download video ids using YouTube Data API. Result size per
1360 # query is limited (currently to 50 videos) so we need to query
1361 # page by page until there are no video ids - it means we got
1364 def download_page(pagenum):
1365 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1367 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1368 page = self._download_webpage(
1369 gdata_url, username,
1370 'Downloading video ids from %d to %d' % (
1371 start_index, start_index + self._GDATA_PAGE_SIZE))
1374 response = json.loads(page)
1375 except ValueError as err:
1376 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1377 if 'entry' not in response['feed']:
1380 # Extract video identifiers
1381 entries = response['feed']['entry']
1382 for entry in entries:
1383 title = entry['title']['$t']
1384 video_id = entry['id']['$t'].split('/')[-1]
1388 'ie_key': 'Youtube',
1392 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1394 return self.playlist_result(url_results, playlist_title=username)
1397 class YoutubeSearchIE(SearchInfoExtractor):
1398 IE_DESC = 'YouTube.com searches'
1399 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1401 IE_NAME = 'youtube:search'
1402 _SEARCH_KEY = 'ytsearch'
1404 def _get_n_results(self, query, n):
1405 """Get a specified number of results for a query"""
1412 while (PAGE_SIZE * pagenum) < limit:
1413 result_url = self._API_URL % (
1414 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1415 (PAGE_SIZE * pagenum) + 1)
1416 data_json = self._download_webpage(
1417 result_url, video_id='query "%s"' % query,
1418 note='Downloading page %s' % (pagenum + 1),
1419 errnote='Unable to download API page')
1420 data = json.loads(data_json)
1421 api_response = data['data']
1423 if 'items' not in api_response:
1424 raise ExtractorError(
1425 '[youtube] No video results', expected=True)
1427 new_ids = list(video['id'] for video in api_response['items'])
1428 video_ids += new_ids
1430 limit = min(n, api_response['totalItems'])
1433 if len(video_ids) > n:
1434 video_ids = video_ids[:n]
1435 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1436 for video_id in video_ids]
1437 return self.playlist_result(videos, query)
1440 class YoutubeSearchDateIE(YoutubeSearchIE):
1441 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1442 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1443 _SEARCH_KEY = 'ytsearchdate'
1444 IE_DESC = 'YouTube.com searches, newest videos first'
1447 class YoutubeSearchURLIE(InfoExtractor):
1448 IE_DESC = 'YouTube.com search URLs'
1449 IE_NAME = 'youtube:search_url'
1450 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1452 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1453 'playlist_mincount': 5,
1455 'title': 'youtube-dl test video',
1459 def _real_extract(self, url):
1460 mobj = re.match(self._VALID_URL, url)
1461 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1463 webpage = self._download_webpage(url, query)
1464 result_code = self._search_regex(
1465 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1467 part_codes = re.findall(
1468 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1470 for part_code in part_codes:
1471 part_title = self._html_search_regex(
1472 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1473 part_url_snippet = self._html_search_regex(
1474 r'(?s)href="([^"]+)"', part_code, 'item URL')
1475 part_url = compat_urlparse.urljoin(
1476 'https://www.youtube.com/', part_url_snippet)
1480 'title': part_title,
1484 '_type': 'playlist',
1490 class YoutubeShowIE(InfoExtractor):
1491 IE_DESC = 'YouTube.com (multi-season) shows'
1492 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1493 IE_NAME = 'youtube:show'
1495 'url': 'http://www.youtube.com/show/airdisasters',
1496 'playlist_mincount': 3,
1498 'id': 'airdisasters',
1499 'title': 'Air Disasters',
1503 def _real_extract(self, url):
1504 mobj = re.match(self._VALID_URL, url)
1505 playlist_id = mobj.group('id')
1506 webpage = self._download_webpage(
1507 url, playlist_id, 'Downloading show webpage')
1508 # There's one playlist for each season of the show
1509 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1510 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1513 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1514 for season in m_seasons
1516 title = self._og_search_title(webpage, fatal=False)
1519 '_type': 'playlist',
1526 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1528 Base class for extractors that fetch info from
1529 http://www.youtube.com/feed_ajax
1530 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1532 _LOGIN_REQUIRED = True
1533 # use action_load_personal_feed instead of action_load_system_feed
1534 _PERSONAL_FEED = False
1537 def _FEED_TEMPLATE(self):
1538 action = 'action_load_system_feed'
1539 if self._PERSONAL_FEED:
1540 action = 'action_load_personal_feed'
1541 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1545 return 'youtube:%s' % self._FEED_NAME
1547 def _real_initialize(self):
1550 def _real_extract(self, url):
1553 for i in itertools.count(1):
1554 info = self._download_json(self._FEED_TEMPLATE % paging,
1555 '%s feed' % self._FEED_NAME,
1556 'Downloading page %s' % i)
1557 feed_html = info.get('feed_html') or info.get('content_html')
1558 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1559 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1560 ids = orderedSet(m.group(1) for m in m_ids)
1561 feed_entries.extend(
1562 self.url_result(video_id, 'Youtube', video_id=video_id)
1563 for video_id in ids)
1565 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1566 load_more_widget_html)
1569 paging = mobj.group('paging')
1570 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1573 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1574 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1575 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1576 _FEED_NAME = 'recommended'
1577 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1580 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1581 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1582 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1583 _FEED_NAME = 'watch_later'
1584 _PLAYLIST_TITLE = 'Youtube Watch Later'
1585 _PERSONAL_FEED = True
1588 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1589 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1590 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1591 _FEED_NAME = 'history'
1592 _PERSONAL_FEED = True
1593 _PLAYLIST_TITLE = 'Youtube Watch History'
1596 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1597 IE_NAME = 'youtube:favorites'
1598 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1599 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1600 _LOGIN_REQUIRED = True
1602 def _real_extract(self, url):
1603 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1604 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1605 return self.url_result(playlist_id, 'YoutubePlaylist')
1608 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1609 IE_NAME = 'youtube:subscriptions'
1610 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1611 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1614 def _real_extract(self, url):
1615 title = 'Youtube Subscriptions'
1616 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1618 # The extraction process is the same as for playlists, but the regex
1619 # for the video ids doesn't contain an index
1621 more_widget_html = content_html = page
1623 for page_num in itertools.count(1):
1624 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1625 new_ids = orderedSet(matches)
1628 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1632 more = self._download_json(
1633 'https://youtube.com/%s' % mobj.group('more'), title,
1634 'Downloading page #%s' % page_num,
1635 transform_source=uppercase_escape)
1636 content_html = more['content_html']
1637 more_widget_html = more['load_more_widget_html']
1640 '_type': 'playlist',
1642 'entries': self._ids_to_results(ids),
1646 class YoutubeTruncatedURLIE(InfoExtractor):
1647 IE_NAME = 'youtube:truncated_url'
1648 IE_DESC = False # Do not list
1649 _VALID_URL = r'''(?x)
1650 (?:https?://)?[^/]+/watch\?(?:
1652 annotation_id=annotation_[^&]+
1654 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1658 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1659 'only_matching': True,
1661 'url': 'http://www.youtube.com/watch?',
1662 'only_matching': True,
1665 def _real_extract(self, url):
1666 raise ExtractorError(
1667 'Did you forget to quote the URL? Remember that & is a meta '
1668 'character in most shells, so you want to put the URL in quotes, '
1670 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1671 ' or simply youtube-dl BaW_jenozKc .',