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
21 compat_urllib_request,
27 get_element_by_attribute,
38 class YoutubeBaseInfoExtractor(InfoExtractor):
39 """Provide base functions for Youtube extractors"""
40 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
41 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
42 _NETRC_MACHINE = 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED = False
46 def _set_language(self):
47 self._set_cookie('.youtube.com', 'PREF', 'f1=50000000&hl=en',
48 # YouTube sets the expire time to about two months
49 expire_time=time.time() + 60*24*3600)
53 Attempt to log in to YouTube.
54 True is returned if successful or skipped.
55 False is returned if login failed.
57 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
59 (username, password) = self._get_login_info()
60 # No authentication to be performed
62 if self._LOGIN_REQUIRED:
63 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
66 login_page = self._download_webpage(
67 self._LOGIN_URL, None,
68 note='Downloading login page',
69 errnote='unable to fetch login page', fatal=False)
70 if login_page is False:
73 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
74 login_page, 'Login GALX parameter')
78 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
83 'PersistentCookie': 'yes',
85 'bgresponse': 'js_disabled',
86 'checkConnection': '',
87 'checkedDomains': 'youtube',
99 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
101 login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in login_form_strs.items())
102 login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
104 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
105 login_results = self._download_webpage(
107 note='Logging in', errnote='unable to log in', fatal=False)
108 if login_results is False:
111 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
112 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
115 # TODO add SMS and phone call support - these require making a request and then prompting the user
117 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', login_results) is not None:
118 tfa_code = self._get_tfa_info()
121 self._downloader.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
122 self._downloader.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
125 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
127 match = re.search(r'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
129 self._downloader.report_warning('Failed to get secTok - did the page structure change?')
130 secTok = match.group(1)
131 match = re.search(r'id="timeStmp"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
133 self._downloader.report_warning('Failed to get timeStmp - did the page structure change?')
134 timeStmp = match.group(1)
137 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
139 'smsUserPin': tfa_code,
140 'smsVerifyPin': 'Verify',
142 'PersistentCookie': 'yes',
143 'checkConnection': '',
144 'checkedDomains': 'youtube',
147 'timeStmp': timeStmp,
148 'service': 'youtube',
151 tfa_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k, v in tfa_form_strs.items())
152 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii')
154 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
155 tfa_results = self._download_webpage(
157 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
159 if tfa_results is False:
162 if re.search(r'(?i)<form[^>]* id="gaia_secondfactorform"', tfa_results) is not None:
163 self._downloader.report_warning('Two-factor code expired. Please try again, or use a one-use backup code instead.')
165 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
166 self._downloader.report_warning('unable to log in - did the page structure change?')
168 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
169 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
172 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
173 self._downloader.report_warning('unable to log in: bad username or password')
177 def _real_initialize(self):
178 if self._downloader is None:
181 if not self._login():
185 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
186 IE_DESC = 'YouTube.com'
187 _VALID_URL = r"""(?x)^
189 (?:https?://|//) # http(s):// or protocol-independent URL
190 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
191 (?:www\.)?deturl\.com/www\.youtube\.com/|
192 (?:www\.)?pwnyoutube\.com/|
193 (?:www\.)?yourepeat\.com/|
194 tube\.majestyc\.net/|
195 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
196 (?:.*?\#/)? # handle anchor (#/) redirect urls
197 (?: # the various things that can precede the ID:
198 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
199 |(?: # or the v= param in all its forms
200 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
201 (?:\?|\#!?) # the params delimiter ? or # or #!
202 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
206 |youtu\.be/ # just youtu.be/xxxx
207 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
209 )? # all until now is optional -> you can pass the naked ID
210 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
211 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
212 (?(1).+)? # if we found the ID, everything can follow
214 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
216 '5': {'ext': 'flv', 'width': 400, 'height': 240},
217 '6': {'ext': 'flv', 'width': 450, 'height': 270},
218 '13': {'ext': '3gp'},
219 '17': {'ext': '3gp', 'width': 176, 'height': 144},
220 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
221 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
222 '34': {'ext': 'flv', 'width': 640, 'height': 360},
223 '35': {'ext': 'flv', 'width': 854, 'height': 480},
224 '36': {'ext': '3gp', 'width': 320, 'height': 240},
225 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
226 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
227 '43': {'ext': 'webm', 'width': 640, 'height': 360},
228 '44': {'ext': 'webm', 'width': 854, 'height': 480},
229 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
230 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
234 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
235 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
236 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
237 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
238 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
239 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
240 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
242 # Apple HTTP Live Streaming
243 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
244 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
245 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
246 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
247 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
248 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
249 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
252 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
253 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
254 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
261 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
262 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
265 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
266 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
267 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
270 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
271 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
272 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
277 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
278 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
279 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
287 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
288 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
291 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
292 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
294 # Dash webm audio with opus inside
295 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
296 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
297 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
300 '_rtmp': {'protocol': 'rtmp'},
306 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
310 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
311 'uploader': 'Philipp Hagemeister',
312 'uploader_id': 'phihag',
313 'upload_date': '20121002',
314 '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 .',
315 'categories': ['Science & Technology'],
317 'dislike_count': int,
321 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
322 'note': 'Test generic use_cipher_signature video (#897)',
326 'upload_date': '20120506',
327 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
328 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
329 'uploader': 'Icona Pop',
330 'uploader_id': 'IconaPop',
334 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
335 'note': 'Test VEVO video with age protection (#956)',
339 'upload_date': '20130703',
340 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
341 'description': 'md5:64249768eec3bc4276236606ea996373',
342 'uploader': 'justintimberlakeVEVO',
343 'uploader_id': 'justintimberlakeVEVO',
347 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
348 'note': 'Embed-only video (#1746)',
352 'upload_date': '20120608',
353 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
354 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
355 'uploader': 'SET India',
356 'uploader_id': 'setindia'
360 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
361 'note': '256k DASH audio (format 141) via DASH manifest',
365 'upload_date': '20121002',
366 'uploader_id': '8KVIDEO',
368 'uploader': '8KVIDEO',
369 'title': 'UHDTV TEST 8K VIDEO.mp4'
372 'youtube_include_dash_manifest': True,
376 # DASH manifest with encrypted signature
378 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
382 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
383 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
384 'uploader': 'AfrojackVEVO',
385 'uploader_id': 'AfrojackVEVO',
386 'upload_date': '20131011',
389 'youtube_include_dash_manifest': True,
395 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
399 'upload_date': '20100909',
400 'uploader': 'The Amazing Atheist',
401 'uploader_id': 'TheAmazingAtheist',
402 'title': 'Burning Everyone\'s Koran',
403 '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',
406 # Normal age-gate video (No vevo, embed allowed)
408 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
412 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
413 'description': 'md5:eca57043abae25130f58f655ad9a7771',
414 'uploader': 'The Witcher',
415 'uploader_id': 'WitcherGame',
416 'upload_date': '20140605',
421 def __init__(self, *args, **kwargs):
422 super(YoutubeIE, self).__init__(*args, **kwargs)
423 self._player_cache = {}
425 def report_video_info_webpage_download(self, video_id):
426 """Report attempt to download video info webpage."""
427 self.to_screen('%s: Downloading video info webpage' % video_id)
429 def report_information_extraction(self, video_id):
430 """Report attempt to extract video information."""
431 self.to_screen('%s: Extracting video information' % video_id)
433 def report_unavailable_format(self, video_id, format):
434 """Report extracted video URL."""
435 self.to_screen('%s: Format %s not available' % (video_id, format))
437 def report_rtmp_download(self):
438 """Indicate the download will use the RTMP protocol."""
439 self.to_screen('RTMP download detected')
441 def _signature_cache_id(self, example_sig):
442 """ Return a string representation of a signature """
443 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
445 def _extract_signature_function(self, video_id, player_url, example_sig):
447 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
450 raise ExtractorError('Cannot identify player %r' % player_url)
451 player_type = id_m.group('ext')
452 player_id = id_m.group('id')
454 # Read from filesystem cache
455 func_id = '%s_%s_%s' % (
456 player_type, player_id, self._signature_cache_id(example_sig))
457 assert os.path.basename(func_id) == func_id
459 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
460 if cache_spec is not None:
461 return lambda s: ''.join(s[i] for i in cache_spec)
463 if player_type == 'js':
464 code = self._download_webpage(
465 player_url, video_id,
466 note='Downloading %s player %s' % (player_type, player_id),
467 errnote='Download of %s failed' % player_url)
468 res = self._parse_sig_js(code)
469 elif player_type == 'swf':
470 urlh = self._request_webpage(
471 player_url, video_id,
472 note='Downloading %s player %s' % (player_type, player_id),
473 errnote='Download of %s failed' % player_url)
475 res = self._parse_sig_swf(code)
477 assert False, 'Invalid player type %r' % player_type
479 if cache_spec is None:
480 test_string = ''.join(map(compat_chr, range(len(example_sig))))
481 cache_res = res(test_string)
482 cache_spec = [ord(c) for c in cache_res]
484 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
487 def _print_sig_code(self, func, example_sig):
488 def gen_sig_code(idxs):
489 def _genslice(start, end, step):
490 starts = '' if start == 0 else str(start)
491 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
492 steps = '' if step == 1 else (':%d' % step)
493 return 's[%s%s%s]' % (starts, ends, steps)
496 start = '(Never used)' # Quelch pyflakes warnings - start will be
497 # set as soon as step is set
498 for i, prev in zip(idxs[1:], idxs[:-1]):
502 yield _genslice(start, prev, step)
505 if i - prev in [-1, 1]:
514 yield _genslice(start, i, step)
516 test_string = ''.join(map(compat_chr, range(len(example_sig))))
517 cache_res = func(test_string)
518 cache_spec = [ord(c) for c in cache_res]
519 expr_code = ' + '.join(gen_sig_code(cache_spec))
520 signature_id_tuple = '(%s)' % (
521 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
522 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
523 ' return %s\n') % (signature_id_tuple, expr_code)
524 self.to_screen('Extracted signature function:\n' + code)
526 def _parse_sig_js(self, jscode):
527 funcname = self._search_regex(
528 r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
529 'Initial JS player signature function name')
531 jsi = JSInterpreter(jscode)
532 initial_function = jsi.extract_function(funcname)
533 return lambda s: initial_function([s])
535 def _parse_sig_swf(self, file_contents):
536 swfi = SWFInterpreter(file_contents)
537 TARGET_CLASSNAME = 'SignatureDecipher'
538 searched_class = swfi.extract_class(TARGET_CLASSNAME)
539 initial_function = swfi.extract_function(searched_class, 'decipher')
540 return lambda s: initial_function([s])
542 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
543 """Turn the encrypted s field into a working signature"""
545 if player_url is None:
546 raise ExtractorError('Cannot decrypt signature without player_url')
548 if player_url.startswith('//'):
549 player_url = 'https:' + player_url
551 player_id = (player_url, self._signature_cache_id(s))
552 if player_id not in self._player_cache:
553 func = self._extract_signature_function(
554 video_id, player_url, s
556 self._player_cache[player_id] = func
557 func = self._player_cache[player_id]
558 if self._downloader.params.get('youtube_print_sig_code'):
559 self._print_sig_code(func, s)
561 except Exception as e:
562 tb = traceback.format_exc()
563 raise ExtractorError(
564 'Signature extraction failed: ' + tb, cause=e)
566 def _get_available_subtitles(self, video_id, webpage):
568 sub_list = self._download_webpage(
569 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
570 video_id, note=False)
571 except ExtractorError as err:
572 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
574 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
579 if lang in sub_lang_list:
581 params = compat_urllib_parse.urlencode({
584 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
585 'name': unescapeHTML(l[0]).encode('utf-8'),
587 url = 'https://www.youtube.com/api/timedtext?' + params
588 sub_lang_list[lang] = url
589 if not sub_lang_list:
590 self._downloader.report_warning('video doesn\'t have subtitles')
594 def _get_available_automatic_caption(self, video_id, webpage):
595 """We need the webpage for getting the captions url, pass it as an
596 argument to speed up the process."""
597 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
598 self.to_screen('%s: Looking for automatic captions' % video_id)
599 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
600 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
602 self._downloader.report_warning(err_msg)
604 player_config = json.loads(mobj.group(1))
606 args = player_config['args']
607 caption_url = args['ttsurl']
608 timestamp = args['timestamp']
609 # We get the available subtitles
610 list_params = compat_urllib_parse.urlencode({
615 list_url = caption_url + '&' + list_params
616 caption_list = self._download_xml(list_url, video_id)
617 original_lang_node = caption_list.find('track')
618 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr':
619 self._downloader.report_warning('Video doesn\'t have automatic captions')
621 original_lang = original_lang_node.attrib['lang_code']
624 for lang_node in caption_list.findall('target'):
625 sub_lang = lang_node.attrib['lang_code']
626 params = compat_urllib_parse.urlencode({
627 'lang': original_lang,
633 sub_lang_list[sub_lang] = caption_url + '&' + params
635 # An extractor error can be raise by the download process if there are
636 # no automatic captions but there are subtitles
637 except (KeyError, ExtractorError):
638 self._downloader.report_warning(err_msg)
642 def extract_id(cls, url):
643 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
645 raise ExtractorError('Invalid URL: %s' % url)
646 video_id = mobj.group(2)
649 def _extract_from_m3u8(self, manifest_url, video_id):
652 def _get_urls(_manifest):
653 lines = _manifest.split('\n')
654 urls = filter(lambda l: l and not l.startswith('#'),
657 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
658 formats_urls = _get_urls(manifest)
659 for format_url in formats_urls:
660 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
661 url_map[itag] = format_url
664 def _extract_annotations(self, video_id):
665 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
666 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
668 def _real_extract(self, url):
670 'http' if self._downloader.params.get('prefer_insecure', False)
673 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
674 mobj = re.search(self._NEXT_URL_RE, url)
676 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
677 video_id = self.extract_id(url)
680 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
681 video_webpage = self._download_webpage(url, video_id)
683 # Attempt to extract SWF player URL
684 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
686 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
691 if re.search(r'player-age-gate-content">', video_webpage) is not None:
693 # We simulate the access to the video from www.youtube.com/v/{video_id}
694 # this can be viewed without login into Youtube
695 data = compat_urllib_parse.urlencode({
696 'video_id': video_id,
697 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
698 'sts': self._search_regex(
699 r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
701 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
702 video_info_webpage = self._download_webpage(
703 video_info_url, video_id,
704 note='Refetching age-gated info webpage',
705 errnote='unable to download video info webpage')
706 video_info = compat_parse_qs(video_info_webpage)
710 # Try looking directly into the video webpage
711 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
713 raise ValueError('Could not find ytplayer.config') # caught below
714 json_code = uppercase_escape(mobj.group(1))
715 ytplayer_config = json.loads(json_code)
716 args = ytplayer_config['args']
717 # Convert to the same format returned by compat_parse_qs
718 video_info = dict((k, [v]) for k, v in args.items())
719 if 'url_encoded_fmt_stream_map' not in args:
720 raise ValueError('No stream_map present') # caught below
722 # We fallback to the get_video_info pages (used by the embed page)
723 self.report_video_info_webpage_download(video_id)
724 for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
725 video_info_url = (proto + '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
726 % (video_id, el_type))
727 video_info_webpage = self._download_webpage(video_info_url,
728 video_id, note=False,
729 errnote='unable to download video info webpage')
730 video_info = compat_parse_qs(video_info_webpage)
731 if 'token' in video_info:
733 if 'token' not in video_info:
734 if 'reason' in video_info:
735 raise ExtractorError(
736 'YouTube said: %s' % video_info['reason'][0],
737 expected=True, video_id=video_id)
739 raise ExtractorError(
740 '"token" parameter not in video info for unknown reason',
743 if 'view_count' in video_info:
744 view_count = int(video_info['view_count'][0])
748 # Check for "rental" videos
749 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
750 raise ExtractorError('"rental" videos not supported')
752 # Start extracting information
753 self.report_information_extraction(video_id)
756 if 'author' not in video_info:
757 raise ExtractorError('Unable to extract uploader name')
758 video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
761 video_uploader_id = None
762 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
764 video_uploader_id = mobj.group(1)
766 self._downloader.report_warning('unable to extract uploader nickname')
769 if 'title' in video_info:
770 video_title = video_info['title'][0]
772 self._downloader.report_warning('Unable to extract video title')
776 # We try first to get a high quality image:
777 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
778 video_webpage, re.DOTALL)
779 if m_thumb is not None:
780 video_thumbnail = m_thumb.group(1)
781 elif 'thumbnail_url' not in video_info:
782 self._downloader.report_warning('unable to extract video thumbnail')
783 video_thumbnail = None
784 else: # don't panic if we can't find it
785 video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
789 mobj = re.search(r'(?s)id="eow-date.*?>(.*?)</span>', video_webpage)
792 r'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
795 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
796 upload_date = unified_strdate(upload_date)
798 m_cat_container = self._search_regex(
799 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
800 video_webpage, 'categories', fatal=False)
802 category = self._html_search_regex(
803 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
805 video_categories = None if category is None else [category]
807 video_categories = None
810 video_description = get_element_by_id("eow-description", video_webpage)
811 if video_description:
812 video_description = re.sub(r'''(?x)
814 (?:[a-zA-Z-]+="[^"]+"\s+)*?
816 (?:[a-zA-Z-]+="[^"]+"\s+)*?
817 class="yt-uix-redirect-link"\s*>
820 ''', r'\1', video_description)
821 video_description = clean_html(video_description)
823 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
825 video_description = unescapeHTML(fd_mobj.group(1))
827 video_description = ''
829 def _extract_count(count_name):
830 count = self._search_regex(
831 r'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re.escape(count_name),
832 video_webpage, count_name, default=None)
833 if count is not None:
834 return int(count.replace(',', ''))
836 like_count = _extract_count('like')
837 dislike_count = _extract_count('dislike')
840 video_subtitles = self.extract_subtitles(video_id, video_webpage)
842 if self._downloader.params.get('listsubtitles', False):
843 self._list_available_subtitles(video_id, video_webpage)
846 if 'length_seconds' not in video_info:
847 self._downloader.report_warning('unable to extract video duration')
848 video_duration = None
850 video_duration = int(compat_urllib_parse.unquote_plus(video_info['length_seconds'][0]))
853 video_annotations = None
854 if self._downloader.params.get('writeannotations', False):
855 video_annotations = self._extract_annotations(video_id)
857 def _map_to_format_list(urlmap):
859 for itag, video_real_url in urlmap.items():
862 'url': video_real_url,
863 'player_url': player_url,
865 if itag in self._formats:
866 dct.update(self._formats[itag])
870 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
871 self.report_rtmp_download()
873 'format_id': '_rtmp',
875 'url': video_info['conn'][0],
876 'player_url': player_url,
878 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
879 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
880 if 'rtmpe%3Dyes' in encoded_url_map:
881 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
883 for url_data_str in encoded_url_map.split(','):
884 url_data = compat_parse_qs(url_data_str)
885 if 'itag' not in url_data or 'url' not in url_data:
887 format_id = url_data['itag'][0]
888 url = url_data['url'][0]
890 if 'sig' in url_data:
891 url += '&signature=' + url_data['sig'][0]
892 elif 's' in url_data:
893 encrypted_sig = url_data['s'][0]
896 jsplayer_url_json = self._search_regex(
897 r'"assets":.+?"js":\s*("[^"]+")',
898 video_webpage, 'JS player URL')
899 player_url = json.loads(jsplayer_url_json)
900 if player_url is None:
901 player_url_json = self._search_regex(
902 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
903 video_webpage, 'age gate player URL')
904 player_url = json.loads(player_url_json)
906 if self._downloader.params.get('verbose'):
907 if player_url is None:
908 player_version = 'unknown'
909 player_desc = 'unknown'
911 if player_url.endswith('swf'):
912 player_version = self._search_regex(
913 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
914 'flash player', fatal=False)
915 player_desc = 'flash player %s' % player_version
917 player_version = self._search_regex(
918 r'html5player-([^/]+?)(?:/html5player)?\.js',
920 'html5 player', fatal=False)
921 player_desc = 'html5 player %s' % player_version
923 parts_sizes = self._signature_cache_id(encrypted_sig)
924 self.to_screen('{%s} signature length %s, %s' %
925 (format_id, parts_sizes, player_desc))
927 signature = self._decrypt_signature(
928 encrypted_sig, video_id, player_url, age_gate)
929 url += '&signature=' + signature
930 if 'ratebypass' not in url:
931 url += '&ratebypass=yes'
932 url_map[format_id] = url
933 formats = _map_to_format_list(url_map)
934 elif video_info.get('hlsvp'):
935 manifest_url = video_info['hlsvp'][0]
936 url_map = self._extract_from_m3u8(manifest_url, video_id)
937 formats = _map_to_format_list(url_map)
939 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
941 # Look for the DASH manifest
942 if self._downloader.params.get('youtube_include_dash_manifest', True):
944 # The DASH manifest used needs to be the one from the original video_webpage.
945 # The one found in get_video_info seems to be using different signatures.
946 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
947 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
948 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
949 dash_manifest_url = video_info.get('dashmpd')[0]
951 def decrypt_sig(mobj):
953 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
954 return '/signature/%s' % dec_s
955 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
956 dash_doc = self._download_xml(
957 dash_manifest_url, video_id,
958 note='Downloading DASH manifest',
959 errnote='Could not download DASH manifest')
960 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
961 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
964 format_id = r.attrib['id']
965 video_url = url_el.text
966 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
968 'format_id': format_id,
970 'width': int_or_none(r.attrib.get('width')),
971 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
972 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
973 'filesize': filesize,
976 existing_format = next(
978 if fo['format_id'] == format_id)
979 except StopIteration:
980 f.update(self._formats.get(format_id, {}))
983 existing_format.update(f)
985 except (ExtractorError, KeyError) as e:
986 self.report_warning('Skipping DASH manifest: %r' % e, video_id)
988 self._sort_formats(formats)
992 'uploader': video_uploader,
993 'uploader_id': video_uploader_id,
994 'upload_date': upload_date,
995 'title': video_title,
996 'thumbnail': video_thumbnail,
997 'description': video_description,
998 'categories': video_categories,
999 'subtitles': video_subtitles,
1000 'duration': video_duration,
1001 'age_limit': 18 if age_gate else 0,
1002 'annotations': video_annotations,
1003 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1004 'view_count': view_count,
1005 'like_count': like_count,
1006 'dislike_count': dislike_count,
1011 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1012 IE_DESC = 'YouTube.com playlists'
1013 _VALID_URL = r"""(?x)(?:
1018 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1019 \? (?:.*?&)*? (?:p|a|list)=
1023 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1024 # Top tracks, they can also include dots
1029 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1031 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1032 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
1033 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1034 IE_NAME = 'youtube:playlist'
1036 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1038 'title': 'ytdl test PL',
1039 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1041 'playlist_count': 3,
1043 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1045 'title': 'YDL_Empty_List',
1047 'playlist_count': 0,
1049 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1050 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1052 'title': '29C3: Not my department',
1054 'playlist_count': 95,
1056 'note': 'issue #673',
1057 'url': 'PLBB231211A4F62143',
1059 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1061 'playlist_mincount': 26,
1063 'note': 'Large playlist',
1064 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1066 'title': 'Uploads from Cauchemar',
1068 'playlist_mincount': 799,
1070 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1072 'title': 'YDL_safe_search',
1074 'playlist_count': 2,
1077 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1078 'playlist_count': 4,
1083 'note': 'Embedded SWF player',
1084 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1085 'playlist_count': 4,
1091 def _real_initialize(self):
1094 def _ids_to_results(self, ids):
1096 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1099 def _extract_mix(self, playlist_id):
1100 # The mixes are generated from a a single video
1101 # the id of the playlist is just 'RD' + video_id
1102 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1103 webpage = self._download_webpage(
1104 url, playlist_id, 'Downloading Youtube mix')
1105 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1107 search_title('playlist-title') or
1108 search_title('title long-title') or
1109 search_title('title'))
1110 title = clean_html(title_span)
1111 ids = orderedSet(re.findall(
1112 r'''(?xs)data-video-username=".*?".*?
1113 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1115 url_results = self._ids_to_results(ids)
1117 return self.playlist_result(url_results, playlist_id, title)
1119 def _real_extract(self, url):
1120 # Extract playlist id
1121 mobj = re.match(self._VALID_URL, url)
1123 raise ExtractorError('Invalid URL: %s' % url)
1124 playlist_id = mobj.group(1) or mobj.group(2)
1126 # Check if it's a video-specific URL
1127 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1128 if 'v' in query_dict:
1129 video_id = query_dict['v'][0]
1130 if self._downloader.params.get('noplaylist'):
1131 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1132 return self.url_result(video_id, 'Youtube', video_id=video_id)
1134 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1136 if playlist_id.startswith('RD'):
1137 # Mixes require a custom extraction process
1138 return self._extract_mix(playlist_id)
1139 if playlist_id.startswith('TL'):
1140 raise ExtractorError('For downloading YouTube.com top lists, use '
1141 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1143 url = self._TEMPLATE_URL % playlist_id
1144 page = self._download_webpage(url, playlist_id)
1145 more_widget_html = content_html = page
1147 # Check if the playlist exists or is private
1148 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1149 raise ExtractorError(
1150 'The playlist doesn\'t exist or is private, use --username or '
1151 '--netrc to access it.',
1154 # Extract the video ids from the playlist pages
1157 for page_num in itertools.count(1):
1158 matches = re.finditer(self._VIDEO_RE, content_html)
1159 # We remove the duplicates and the link with index 0
1160 # (it's not the first video of the playlist)
1161 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1164 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1168 more = self._download_json(
1169 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1170 'Downloading page #%s' % page_num,
1171 transform_source=uppercase_escape)
1172 content_html = more['content_html']
1173 more_widget_html = more['load_more_widget_html']
1175 playlist_title = self._html_search_regex(
1176 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1179 url_results = self._ids_to_results(ids)
1180 return self.playlist_result(url_results, playlist_id, playlist_title)
1183 class YoutubeTopListIE(YoutubePlaylistIE):
1184 IE_NAME = 'youtube:toplist'
1185 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1186 ' (Example: "yttoplist:music:Top Tracks")')
1187 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1189 'url': 'yttoplist:music:Trending',
1190 'playlist_mincount': 5,
1191 'skip': 'Only works for logged-in users',
1194 def _real_extract(self, url):
1195 mobj = re.match(self._VALID_URL, url)
1196 channel = mobj.group('chann')
1197 title = mobj.group('title')
1198 query = compat_urllib_parse.urlencode({'title': title})
1199 channel_page = self._download_webpage(
1200 'https://www.youtube.com/%s' % channel, title)
1201 link = self._html_search_regex(
1203 <a\s+href="([^"]+)".*?>\s*
1204 <span\s+class="branded-page-module-title-text">\s*
1205 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1206 channel_page, 'list')
1207 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1209 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1211 # sometimes the webpage doesn't contain the videos
1212 # retry until we get them
1213 for i in itertools.count(0):
1214 msg = 'Downloading Youtube mix'
1216 msg += ', retry #%d' % i
1218 webpage = self._download_webpage(url, title, msg)
1219 ids = orderedSet(re.findall(video_re, webpage))
1222 url_results = self._ids_to_results(ids)
1223 return self.playlist_result(url_results, playlist_title=title)
1226 class YoutubeChannelIE(InfoExtractor):
1227 IE_DESC = 'YouTube.com channels'
1228 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1229 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1230 _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'
1231 IE_NAME = 'youtube:channel'
1233 'note': 'paginated channel',
1234 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1235 'playlist_mincount': 91,
1238 def extract_videos_from_page(self, page):
1240 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1241 if mobj.group(1) not in ids_in_page:
1242 ids_in_page.append(mobj.group(1))
1245 def _real_extract(self, url):
1246 # Extract channel id
1247 mobj = re.match(self._VALID_URL, url)
1249 raise ExtractorError('Invalid URL: %s' % url)
1251 # Download channel page
1252 channel_id = mobj.group(1)
1254 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1255 channel_page = self._download_webpage(url, channel_id)
1256 autogenerated = re.search(r'''(?x)
1258 channel-header-autogenerated-label|
1259 yt-channel-title-autogenerated
1260 )[^"]*"''', channel_page) is not None
1263 # The videos are contained in a single page
1264 # the ajax pages can't be used, they are empty
1265 video_ids = self.extract_videos_from_page(channel_page)
1267 # Download all channel pages using the json-based channel_ajax query
1268 for pagenum in itertools.count(1):
1269 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1270 page = self._download_json(
1271 url, channel_id, note='Downloading page #%s' % pagenum,
1272 transform_source=uppercase_escape)
1274 ids_in_page = self.extract_videos_from_page(page['content_html'])
1275 video_ids.extend(ids_in_page)
1277 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1280 self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1282 url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
1283 for video_id in video_ids]
1284 return self.playlist_result(url_entries, channel_id)
1287 class YoutubeUserIE(InfoExtractor):
1288 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1289 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1290 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1291 _GDATA_PAGE_SIZE = 50
1292 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1293 IE_NAME = 'youtube:user'
1296 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1297 'playlist_mincount': 320,
1299 'title': 'TheLinuxFoundation',
1302 'url': 'ytuser:phihag',
1303 'only_matching': True,
1307 def suitable(cls, url):
1308 # Don't return True if the url can be extracted with other youtube
1309 # extractor, the regex would is too permissive and it would match.
1310 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1311 if any(ie.suitable(url) for ie in other_ies):
1314 return super(YoutubeUserIE, cls).suitable(url)
1316 def _real_extract(self, url):
1318 mobj = re.match(self._VALID_URL, url)
1320 raise ExtractorError('Invalid URL: %s' % url)
1322 username = mobj.group(1)
1324 # Download video ids using YouTube Data API. Result size per
1325 # query is limited (currently to 50 videos) so we need to query
1326 # page by page until there are no video ids - it means we got
1329 def download_page(pagenum):
1330 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1332 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1333 page = self._download_webpage(
1334 gdata_url, username,
1335 'Downloading video ids from %d to %d' % (
1336 start_index, start_index + self._GDATA_PAGE_SIZE))
1339 response = json.loads(page)
1340 except ValueError as err:
1341 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1342 if 'entry' not in response['feed']:
1345 # Extract video identifiers
1346 entries = response['feed']['entry']
1347 for entry in entries:
1348 title = entry['title']['$t']
1349 video_id = entry['id']['$t'].split('/')[-1]
1353 'ie_key': 'Youtube',
1357 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1359 return self.playlist_result(url_results, playlist_title=username)
1362 class YoutubeSearchIE(SearchInfoExtractor):
1363 IE_DESC = 'YouTube.com searches'
1364 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1366 IE_NAME = 'youtube:search'
1367 _SEARCH_KEY = 'ytsearch'
1369 def _get_n_results(self, query, n):
1370 """Get a specified number of results for a query"""
1377 while (PAGE_SIZE * pagenum) < limit:
1378 result_url = self._API_URL % (
1379 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1380 (PAGE_SIZE * pagenum) + 1)
1381 data_json = self._download_webpage(
1382 result_url, video_id='query "%s"' % query,
1383 note='Downloading page %s' % (pagenum + 1),
1384 errnote='Unable to download API page')
1385 data = json.loads(data_json)
1386 api_response = data['data']
1388 if 'items' not in api_response:
1389 raise ExtractorError(
1390 '[youtube] No video results', expected=True)
1392 new_ids = list(video['id'] for video in api_response['items'])
1393 video_ids += new_ids
1395 limit = min(n, api_response['totalItems'])
1398 if len(video_ids) > n:
1399 video_ids = video_ids[:n]
1400 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1401 for video_id in video_ids]
1402 return self.playlist_result(videos, query)
1405 class YoutubeSearchDateIE(YoutubeSearchIE):
1406 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1407 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1408 _SEARCH_KEY = 'ytsearchdate'
1409 IE_DESC = 'YouTube.com searches, newest videos first'
1412 class YoutubeSearchURLIE(InfoExtractor):
1413 IE_DESC = 'YouTube.com search URLs'
1414 IE_NAME = 'youtube:search_url'
1415 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1417 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1418 'playlist_mincount': 5,
1420 'title': 'youtube-dl test video',
1424 def _real_extract(self, url):
1425 mobj = re.match(self._VALID_URL, url)
1426 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1428 webpage = self._download_webpage(url, query)
1429 result_code = self._search_regex(
1430 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1432 part_codes = re.findall(
1433 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1435 for part_code in part_codes:
1436 part_title = self._html_search_regex(
1437 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1438 part_url_snippet = self._html_search_regex(
1439 r'(?s)href="([^"]+)"', part_code, 'item URL')
1440 part_url = compat_urlparse.urljoin(
1441 'https://www.youtube.com/', part_url_snippet)
1445 'title': part_title,
1449 '_type': 'playlist',
1455 class YoutubeShowIE(InfoExtractor):
1456 IE_DESC = 'YouTube.com (multi-season) shows'
1457 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1458 IE_NAME = 'youtube:show'
1460 'url': 'http://www.youtube.com/show/airdisasters',
1461 'playlist_mincount': 3,
1463 'id': 'airdisasters',
1464 'title': 'Air Disasters',
1468 def _real_extract(self, url):
1469 mobj = re.match(self._VALID_URL, url)
1470 playlist_id = mobj.group('id')
1471 webpage = self._download_webpage(
1472 url, playlist_id, 'Downloading show webpage')
1473 # There's one playlist for each season of the show
1474 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1475 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1478 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1479 for season in m_seasons
1481 title = self._og_search_title(webpage, fatal=False)
1484 '_type': 'playlist',
1491 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1493 Base class for extractors that fetch info from
1494 http://www.youtube.com/feed_ajax
1495 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1497 _LOGIN_REQUIRED = True
1498 # use action_load_personal_feed instead of action_load_system_feed
1499 _PERSONAL_FEED = False
1502 def _FEED_TEMPLATE(self):
1503 action = 'action_load_system_feed'
1504 if self._PERSONAL_FEED:
1505 action = 'action_load_personal_feed'
1506 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1510 return 'youtube:%s' % self._FEED_NAME
1512 def _real_initialize(self):
1515 def _real_extract(self, url):
1518 for i in itertools.count(1):
1519 info = self._download_json(self._FEED_TEMPLATE % paging,
1520 '%s feed' % self._FEED_NAME,
1521 'Downloading page %s' % i)
1522 feed_html = info.get('feed_html') or info.get('content_html')
1523 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1524 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1525 ids = orderedSet(m.group(1) for m in m_ids)
1526 feed_entries.extend(
1527 self.url_result(video_id, 'Youtube', video_id=video_id)
1528 for video_id in ids)
1530 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1531 load_more_widget_html)
1534 paging = mobj.group('paging')
1535 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1538 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1539 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1540 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1541 _FEED_NAME = 'recommended'
1542 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1545 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1546 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1547 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1548 _FEED_NAME = 'watch_later'
1549 _PLAYLIST_TITLE = 'Youtube Watch Later'
1550 _PERSONAL_FEED = True
1553 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1554 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1555 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1556 _FEED_NAME = 'history'
1557 _PERSONAL_FEED = True
1558 _PLAYLIST_TITLE = 'Youtube Watch History'
1561 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1562 IE_NAME = 'youtube:favorites'
1563 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1564 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1565 _LOGIN_REQUIRED = True
1567 def _real_extract(self, url):
1568 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1569 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1570 return self.url_result(playlist_id, 'YoutubePlaylist')
1573 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1574 IE_NAME = 'youtube:subscriptions'
1575 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1576 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1579 def _real_extract(self, url):
1580 title = 'Youtube Subscriptions'
1581 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1583 # The extraction process is the same as for playlists, but the regex
1584 # for the video ids doesn't contain an index
1586 more_widget_html = content_html = page
1588 for page_num in itertools.count(1):
1589 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1590 new_ids = orderedSet(matches)
1593 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1597 more = self._download_json(
1598 'https://youtube.com/%s' % mobj.group('more'), title,
1599 'Downloading page #%s' % page_num,
1600 transform_source=uppercase_escape)
1601 content_html = more['content_html']
1602 more_widget_html = more['load_more_widget_html']
1605 '_type': 'playlist',
1607 'entries': self._ids_to_results(ids),
1611 class YoutubeTruncatedURLIE(InfoExtractor):
1612 IE_NAME = 'youtube:truncated_url'
1613 IE_DESC = False # Do not list
1614 _VALID_URL = r'''(?x)
1615 (?:https?://)?[^/]+/watch\?(?:
1617 annotation_id=annotation_[^&]+
1619 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1623 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1624 'only_matching': True,
1626 'url': 'http://www.youtube.com/watch?',
1627 'only_matching': True,
1630 def _real_extract(self, url):
1631 raise ExtractorError(
1632 'Did you forget to quote the URL? Remember that & is a meta '
1633 'character in most shells, so you want to put the URL in quotes, '
1635 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1636 ' or simply youtube-dl BaW_jenozKc .',