3 from __future__ import unicode_literals
12 from .common import InfoExtractor, SearchInfoExtractor
13 from .subtitles import SubtitlesInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
20 compat_urllib_request,
26 get_element_by_attribute,
37 class YoutubeBaseInfoExtractor(InfoExtractor):
38 """Provide base functions for Youtube extractors"""
39 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
40 _TWOFACTOR_URL = 'https://accounts.google.com/SecondFactor'
41 _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
42 _AGE_URL = 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
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):
48 return bool(self._download_webpage(
50 note='Setting language', errnote='unable to set language',
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 _confirm_age(self):
182 'action_confirm': 'Confirm',
184 req = compat_urllib_request.Request(self._AGE_URL,
185 compat_urllib_parse.urlencode(age_form).encode('ascii'))
187 self._download_webpage(
189 note='Confirming age', errnote='Unable to confirm age',
192 def _real_initialize(self):
193 if self._downloader is None:
195 if self._get_login_info()[0] is not None:
196 if not self._set_language():
198 if not self._login():
203 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
204 IE_DESC = 'YouTube.com'
205 _VALID_URL = r"""(?x)^
207 (?:https?://|//) # http(s):// or protocol-independent URL
208 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
209 (?:www\.)?deturl\.com/www\.youtube\.com/|
210 (?:www\.)?pwnyoutube\.com/|
211 (?:www\.)?yourepeat\.com/|
212 tube\.majestyc\.net/|
213 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
214 (?:.*?\#/)? # handle anchor (#/) redirect urls
215 (?: # the various things that can precede the ID:
216 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
217 |(?: # or the v= param in all its forms
218 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
219 (?:\?|\#!?) # the params delimiter ? or # or #!
220 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
224 |youtu\.be/ # just youtu.be/xxxx
225 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
227 )? # all until now is optional -> you can pass the naked ID
228 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
229 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
230 (?(1).+)? # if we found the ID, everything can follow
232 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
234 '5': {'ext': 'flv', 'width': 400, 'height': 240},
235 '6': {'ext': 'flv', 'width': 450, 'height': 270},
236 '13': {'ext': '3gp'},
237 '17': {'ext': '3gp', 'width': 176, 'height': 144},
238 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
239 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
240 '34': {'ext': 'flv', 'width': 640, 'height': 360},
241 '35': {'ext': 'flv', 'width': 854, 'height': 480},
242 '36': {'ext': '3gp', 'width': 320, 'height': 240},
243 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
244 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
245 '43': {'ext': 'webm', 'width': 640, 'height': 360},
246 '44': {'ext': 'webm', 'width': 854, 'height': 480},
247 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
248 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
252 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
253 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
254 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
255 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
256 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
257 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
258 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
260 # Apple HTTP Live Streaming
261 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
262 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
263 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
264 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
265 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
266 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
267 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
270 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
276 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
277 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
278 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
279 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
280 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
283 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
284 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
285 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
288 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
291 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
292 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
293 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
294 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
295 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
301 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
302 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
303 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
304 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
305 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
308 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
309 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
311 # Dash webm audio with opus inside
312 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
313 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
314 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
317 '_rtmp': {'protocol': 'rtmp'},
323 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
327 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
328 'uploader': 'Philipp Hagemeister',
329 'uploader_id': 'phihag',
330 'upload_date': '20121002',
331 '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 .',
332 'categories': ['Science & Technology'],
334 'dislike_count': int,
338 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
339 'note': 'Test generic use_cipher_signature video (#897)',
343 'upload_date': '20120506',
344 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
345 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
346 'uploader': 'Icona Pop',
347 'uploader_id': 'IconaPop',
351 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
352 'note': 'Test VEVO video with age protection (#956)',
356 'upload_date': '20130703',
357 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
358 'description': 'md5:64249768eec3bc4276236606ea996373',
359 'uploader': 'justintimberlakeVEVO',
360 'uploader_id': 'justintimberlakeVEVO',
364 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
365 'note': 'Embed-only video (#1746)',
369 'upload_date': '20120608',
370 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
371 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
372 'uploader': 'SET India',
373 'uploader_id': 'setindia'
377 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
378 'note': '256k DASH audio (format 141) via DASH manifest',
382 'upload_date': '20121002',
383 'uploader_id': '8KVIDEO',
385 'uploader': '8KVIDEO',
386 'title': 'UHDTV TEST 8K VIDEO.mp4'
389 'youtube_include_dash_manifest': True,
393 # DASH manifest with encrypted signature
395 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
399 'title': 'Afrojack - The Spark ft. Spree Wilson',
400 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8',
401 'uploader': 'AfrojackVEVO',
402 'uploader_id': 'AfrojackVEVO',
403 'upload_date': '20131011',
406 'youtube_include_dash_manifest': True,
412 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
416 'upload_date': '20100909',
417 'uploader': 'The Amazing Atheist',
418 'uploader_id': 'TheAmazingAtheist',
419 'title': 'Burning Everyone\'s Koran',
420 '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',
425 def __init__(self, *args, **kwargs):
426 super(YoutubeIE, self).__init__(*args, **kwargs)
427 self._player_cache = {}
429 def report_video_info_webpage_download(self, video_id):
430 """Report attempt to download video info webpage."""
431 self.to_screen('%s: Downloading video info webpage' % video_id)
433 def report_information_extraction(self, video_id):
434 """Report attempt to extract video information."""
435 self.to_screen('%s: Extracting video information' % video_id)
437 def report_unavailable_format(self, video_id, format):
438 """Report extracted video URL."""
439 self.to_screen('%s: Format %s not available' % (video_id, format))
441 def report_rtmp_download(self):
442 """Indicate the download will use the RTMP protocol."""
443 self.to_screen('RTMP download detected')
445 def _signature_cache_id(self, example_sig):
446 """ Return a string representation of a signature """
447 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
449 def _extract_signature_function(self, video_id, player_url, example_sig):
451 r'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
454 raise ExtractorError('Cannot identify player %r' % player_url)
455 player_type = id_m.group('ext')
456 player_id = id_m.group('id')
458 # Read from filesystem cache
459 func_id = '%s_%s_%s' % (
460 player_type, player_id, self._signature_cache_id(example_sig))
461 assert os.path.basename(func_id) == func_id
463 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
464 if cache_spec is not None:
465 return lambda s: ''.join(s[i] for i in cache_spec)
467 if player_type == 'js':
468 code = self._download_webpage(
469 player_url, video_id,
470 note='Downloading %s player %s' % (player_type, player_id),
471 errnote='Download of %s failed' % player_url)
472 res = self._parse_sig_js(code)
473 elif player_type == 'swf':
474 urlh = self._request_webpage(
475 player_url, video_id,
476 note='Downloading %s player %s' % (player_type, player_id),
477 errnote='Download of %s failed' % player_url)
479 res = self._parse_sig_swf(code)
481 assert False, 'Invalid player type %r' % player_type
483 if cache_spec is None:
484 test_string = ''.join(map(compat_chr, range(len(example_sig))))
485 cache_res = res(test_string)
486 cache_spec = [ord(c) for c in cache_res]
488 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
491 def _print_sig_code(self, func, example_sig):
492 def gen_sig_code(idxs):
493 def _genslice(start, end, step):
494 starts = '' if start == 0 else str(start)
495 ends = (':%d' % (end+step)) if end + step >= 0 else ':'
496 steps = '' if step == 1 else (':%d' % step)
497 return 's[%s%s%s]' % (starts, ends, steps)
500 start = '(Never used)' # Quelch pyflakes warnings - start will be
501 # set as soon as step is set
502 for i, prev in zip(idxs[1:], idxs[:-1]):
506 yield _genslice(start, prev, step)
509 if i - prev in [-1, 1]:
518 yield _genslice(start, i, step)
520 test_string = ''.join(map(compat_chr, range(len(example_sig))))
521 cache_res = func(test_string)
522 cache_spec = [ord(c) for c in cache_res]
523 expr_code = ' + '.join(gen_sig_code(cache_spec))
524 signature_id_tuple = '(%s)' % (
525 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
526 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
527 ' return %s\n') % (signature_id_tuple, expr_code)
528 self.to_screen('Extracted signature function:\n' + code)
530 def _parse_sig_js(self, jscode):
531 funcname = self._search_regex(
532 r'\.sig\|\|([a-zA-Z0-9]+)\(', jscode,
533 'Initial JS player signature function name')
535 jsi = JSInterpreter(jscode)
536 initial_function = jsi.extract_function(funcname)
537 return lambda s: initial_function([s])
539 def _parse_sig_swf(self, file_contents):
540 swfi = SWFInterpreter(file_contents)
541 TARGET_CLASSNAME = 'SignatureDecipher'
542 searched_class = swfi.extract_class(TARGET_CLASSNAME)
543 initial_function = swfi.extract_function(searched_class, 'decipher')
544 return lambda s: initial_function([s])
546 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
547 """Turn the encrypted s field into a working signature"""
549 if player_url is None:
550 raise ExtractorError('Cannot decrypt signature without player_url')
552 if player_url.startswith('//'):
553 player_url = 'https:' + player_url
555 player_id = (player_url, self._signature_cache_id(s))
556 if player_id not in self._player_cache:
557 func = self._extract_signature_function(
558 video_id, player_url, s
560 self._player_cache[player_id] = func
561 func = self._player_cache[player_id]
562 if self._downloader.params.get('youtube_print_sig_code'):
563 self._print_sig_code(func, s)
565 except Exception as e:
566 tb = traceback.format_exc()
567 raise ExtractorError(
568 'Signature extraction failed: ' + tb, cause=e)
570 def _get_available_subtitles(self, video_id, webpage):
572 sub_list = self._download_webpage(
573 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
574 video_id, note=False)
575 except ExtractorError as err:
576 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
578 lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
583 if lang in sub_lang_list:
585 params = compat_urllib_parse.urlencode({
588 'fmt': self._downloader.params.get('subtitlesformat', 'srt'),
589 'name': unescapeHTML(l[0]).encode('utf-8'),
591 url = 'https://www.youtube.com/api/timedtext?' + params
592 sub_lang_list[lang] = url
593 if not sub_lang_list:
594 self._downloader.report_warning('video doesn\'t have subtitles')
598 def _get_available_automatic_caption(self, video_id, webpage):
599 """We need the webpage for getting the captions url, pass it as an
600 argument to speed up the process."""
601 sub_format = self._downloader.params.get('subtitlesformat', 'srt')
602 self.to_screen('%s: Looking for automatic captions' % video_id)
603 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
604 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
606 self._downloader.report_warning(err_msg)
608 player_config = json.loads(mobj.group(1))
610 args = player_config[u'args']
611 caption_url = args[u'ttsurl']
612 timestamp = args[u'timestamp']
613 # We get the available subtitles
614 list_params = compat_urllib_parse.urlencode({
619 list_url = caption_url + '&' + list_params
620 caption_list = self._download_xml(list_url, video_id)
621 original_lang_node = caption_list.find('track')
622 if original_lang_node is None or original_lang_node.attrib.get('kind') != 'asr':
623 self._downloader.report_warning('Video doesn\'t have automatic captions')
625 original_lang = original_lang_node.attrib['lang_code']
628 for lang_node in caption_list.findall('target'):
629 sub_lang = lang_node.attrib['lang_code']
630 params = compat_urllib_parse.urlencode({
631 'lang': original_lang,
637 sub_lang_list[sub_lang] = caption_url + '&' + params
639 # An extractor error can be raise by the download process if there are
640 # no automatic captions but there are subtitles
641 except (KeyError, ExtractorError):
642 self._downloader.report_warning(err_msg)
646 def extract_id(cls, url):
647 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
649 raise ExtractorError('Invalid URL: %s' % url)
650 video_id = mobj.group(2)
653 def _extract_from_m3u8(self, manifest_url, video_id):
656 def _get_urls(_manifest):
657 lines = _manifest.split('\n')
658 urls = filter(lambda l: l and not l.startswith('#'),
661 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
662 formats_urls = _get_urls(manifest)
663 for format_url in formats_urls:
664 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
665 url_map[itag] = format_url
668 def _extract_annotations(self, video_id):
669 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
670 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
672 def _real_extract(self, url):
674 'http' if self._downloader.params.get('prefer_insecure', False)
677 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
678 mobj = re.search(self._NEXT_URL_RE, url)
680 url = proto + '://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
681 video_id = self.extract_id(url)
684 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
686 c for c in self._downloader.cookiejar
687 if c.domain == '.youtube.com' and c.name == 'PREF']
688 for pc in pref_cookies:
689 if 'hl=' in pc.value:
690 pc.value = re.sub(r'hl=[^&]+', 'hl=en', pc.value)
695 video_webpage = self._download_webpage(url, video_id)
697 # Attempt to extract SWF player URL
698 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
700 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
705 self.report_video_info_webpage_download(video_id)
706 if re.search(r'player-age-gate-content">', video_webpage) is not None:
708 # We simulate the access to the video from www.youtube.com/v/{video_id}
709 # this can be viewed without login into Youtube
710 data = compat_urllib_parse.urlencode({
711 'video_id': video_id,
712 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
713 'sts': self._search_regex(
714 r'"sts"\s*:\s*(\d+)', video_webpage, 'sts', default=''),
716 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
717 video_info_webpage = self._download_webpage(
718 video_info_url, video_id,
719 note='Refetching age-gated info webpage',
720 errnote='unable to download video info webpage')
721 video_info = compat_parse_qs(video_info_webpage)
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, video_id,
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 # Decide which formats to download
859 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
861 raise ValueError('Could not find vevo ID')
862 json_code = uppercase_escape(mobj.group(1))
863 ytplayer_config = json.loads(json_code)
864 args = ytplayer_config['args']
865 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
866 # this signatures are encrypted
867 if 'url_encoded_fmt_stream_map' not in args:
868 raise ValueError('No stream_map present') # caught below
869 re_signature = re.compile(r'[&,]s=')
870 m_s = re_signature.search(args['url_encoded_fmt_stream_map'])
872 self.to_screen('%s: Encrypted signatures detected.' % video_id)
873 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
874 m_s = re_signature.search(args.get('adaptive_fmts', ''))
876 if 'adaptive_fmts' in video_info:
877 video_info['adaptive_fmts'][0] += ',' + args['adaptive_fmts']
879 video_info['adaptive_fmts'] = [args['adaptive_fmts']]
883 def _map_to_format_list(urlmap):
885 for itag, video_real_url in urlmap.items():
888 'url': video_real_url,
889 'player_url': player_url,
891 if itag in self._formats:
892 dct.update(self._formats[itag])
896 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
897 self.report_rtmp_download()
899 'format_id': '_rtmp',
901 'url': video_info['conn'][0],
902 'player_url': player_url,
904 elif len(video_info.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info.get('adaptive_fmts', [])) >= 1:
905 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
906 if 'rtmpe%3Dyes' in encoded_url_map:
907 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
909 for url_data_str in encoded_url_map.split(','):
910 url_data = compat_parse_qs(url_data_str)
911 if 'itag' not in url_data or 'url' not in url_data:
913 format_id = url_data['itag'][0]
914 url = url_data['url'][0]
916 if 'sig' in url_data:
917 url += '&signature=' + url_data['sig'][0]
918 elif 's' in url_data:
919 encrypted_sig = url_data['s'][0]
922 jsplayer_url_json = self._search_regex(
923 r'"assets":.+?"js":\s*("[^"]+")',
924 video_webpage, 'JS player URL')
925 player_url = json.loads(jsplayer_url_json)
926 if player_url is None:
927 player_url_json = self._search_regex(
928 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
929 video_webpage, 'age gate player URL')
930 player_url = json.loads(player_url_json)
932 if self._downloader.params.get('verbose'):
933 if player_url is None:
934 player_version = 'unknown'
935 player_desc = 'unknown'
937 if player_url.endswith('swf'):
938 player_version = self._search_regex(
939 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
940 'flash player', fatal=False)
941 player_desc = 'flash player %s' % player_version
943 player_version = self._search_regex(
944 r'html5player-([^/]+?)(?:/html5player)?\.js',
946 'html5 player', fatal=False)
947 player_desc = 'html5 player %s' % player_version
949 parts_sizes = self._signature_cache_id(encrypted_sig)
950 self.to_screen('{%s} signature length %s, %s' %
951 (format_id, parts_sizes, player_desc))
953 signature = self._decrypt_signature(
954 encrypted_sig, video_id, player_url, age_gate)
955 url += '&signature=' + signature
956 if 'ratebypass' not in url:
957 url += '&ratebypass=yes'
958 url_map[format_id] = url
959 formats = _map_to_format_list(url_map)
960 elif video_info.get('hlsvp'):
961 manifest_url = video_info['hlsvp'][0]
962 url_map = self._extract_from_m3u8(manifest_url, video_id)
963 formats = _map_to_format_list(url_map)
965 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
967 # Look for the DASH manifest
968 if self._downloader.params.get('youtube_include_dash_manifest', True):
970 # The DASH manifest used needs to be the one from the original video_webpage.
971 # The one found in get_video_info seems to be using different signatures.
972 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
973 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
974 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
976 dash_manifest_url = video_info.get('dashmpd')[0]
978 dash_manifest_url = ytplayer_config['args']['dashmpd']
980 def decrypt_sig(mobj):
982 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
983 return '/signature/%s' % dec_s
984 dash_manifest_url = re.sub(r'/s/([\w\.]+)', decrypt_sig, dash_manifest_url)
985 dash_doc = self._download_xml(
986 dash_manifest_url, video_id,
987 note='Downloading DASH manifest',
988 errnote='Could not download DASH manifest')
989 for r in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
990 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
993 format_id = r.attrib['id']
994 video_url = url_el.text
995 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
997 'format_id': format_id,
999 'width': int_or_none(r.attrib.get('width')),
1000 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
1001 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
1002 'filesize': filesize,
1005 existing_format = next(
1006 fo for fo in formats
1007 if fo['format_id'] == format_id)
1008 except StopIteration:
1009 f.update(self._formats.get(format_id, {}))
1012 existing_format.update(f)
1014 except (ExtractorError, KeyError) as e:
1015 self.report_warning('Skipping DASH manifest: %r' % e, video_id)
1017 self._sort_formats(formats)
1021 'uploader': video_uploader,
1022 'uploader_id': video_uploader_id,
1023 'upload_date': upload_date,
1024 'title': video_title,
1025 'thumbnail': video_thumbnail,
1026 'description': video_description,
1027 'categories': video_categories,
1028 'subtitles': video_subtitles,
1029 'duration': video_duration,
1030 'age_limit': 18 if age_gate else 0,
1031 'annotations': video_annotations,
1032 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1033 'view_count': view_count,
1034 'like_count': like_count,
1035 'dislike_count': dislike_count,
1040 class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
1041 IE_DESC = 'YouTube.com playlists'
1042 _VALID_URL = r"""(?x)(?:
1047 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1048 \? (?:.*?&)*? (?:p|a|list)=
1052 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1053 # Top tracks, they can also include dots
1058 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1060 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1061 _MORE_PAGES_INDICATOR = r'data-link-type="next"'
1062 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1063 IE_NAME = 'youtube:playlist'
1065 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1067 'title': 'ytdl test PL',
1068 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1070 'playlist_count': 3,
1072 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1074 'title': 'YDL_Empty_List',
1076 'playlist_count': 0,
1078 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1079 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1081 'title': '29C3: Not my department',
1083 'playlist_count': 95,
1085 'note': 'issue #673',
1086 'url': 'PLBB231211A4F62143',
1088 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1090 'playlist_mincount': 26,
1092 'note': 'Large playlist',
1093 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1095 'title': 'Uploads from Cauchemar',
1097 'playlist_mincount': 799,
1099 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1101 'title': 'YDL_safe_search',
1103 'playlist_count': 2,
1106 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1107 'playlist_count': 4,
1112 'note': 'Embedded SWF player',
1113 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1114 'playlist_count': 4,
1120 def _real_initialize(self):
1123 def _ids_to_results(self, ids):
1125 self.url_result(vid_id, 'Youtube', video_id=vid_id)
1128 def _extract_mix(self, playlist_id):
1129 # The mixes are generated from a a single video
1130 # the id of the playlist is just 'RD' + video_id
1131 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1132 webpage = self._download_webpage(
1133 url, playlist_id, 'Downloading Youtube mix')
1134 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1136 search_title('playlist-title') or
1137 search_title('title long-title') or
1138 search_title('title'))
1139 title = clean_html(title_span)
1140 ids = orderedSet(re.findall(
1141 r'''(?xs)data-video-username=".*?".*?
1142 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1144 url_results = self._ids_to_results(ids)
1146 return self.playlist_result(url_results, playlist_id, title)
1148 def _real_extract(self, url):
1149 # Extract playlist id
1150 mobj = re.match(self._VALID_URL, url)
1152 raise ExtractorError('Invalid URL: %s' % url)
1153 playlist_id = mobj.group(1) or mobj.group(2)
1155 # Check if it's a video-specific URL
1156 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1157 if 'v' in query_dict:
1158 video_id = query_dict['v'][0]
1159 if self._downloader.params.get('noplaylist'):
1160 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1161 return self.url_result(video_id, 'Youtube', video_id=video_id)
1163 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1165 if playlist_id.startswith('RD'):
1166 # Mixes require a custom extraction process
1167 return self._extract_mix(playlist_id)
1168 if playlist_id.startswith('TL'):
1169 raise ExtractorError('For downloading YouTube.com top lists, use '
1170 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected=True)
1172 url = self._TEMPLATE_URL % playlist_id
1173 page = self._download_webpage(url, playlist_id)
1174 more_widget_html = content_html = page
1176 # Check if the playlist exists or is private
1177 if re.search(r'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page) is not None:
1178 raise ExtractorError(
1179 'The playlist doesn\'t exist or is private, use --username or '
1180 '--netrc to access it.',
1183 # Extract the video ids from the playlist pages
1186 for page_num in itertools.count(1):
1187 matches = re.finditer(self._VIDEO_RE, content_html)
1188 # We remove the duplicates and the link with index 0
1189 # (it's not the first video of the playlist)
1190 new_ids = orderedSet(m.group('id') for m in matches if m.group('index') != '0')
1193 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1197 more = self._download_json(
1198 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
1199 'Downloading page #%s' % page_num,
1200 transform_source=uppercase_escape)
1201 content_html = more['content_html']
1202 more_widget_html = more['load_more_widget_html']
1204 playlist_title = self._html_search_regex(
1205 r'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1208 url_results = self._ids_to_results(ids)
1209 return self.playlist_result(url_results, playlist_id, playlist_title)
1212 class YoutubeTopListIE(YoutubePlaylistIE):
1213 IE_NAME = 'youtube:toplist'
1214 IE_DESC = ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1215 ' (Example: "yttoplist:music:Top Tracks")')
1216 _VALID_URL = r'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1218 'url': 'yttoplist:music:Trending',
1219 'playlist_mincount': 5,
1220 'skip': 'Only works for logged-in users',
1223 def _real_extract(self, url):
1224 mobj = re.match(self._VALID_URL, url)
1225 channel = mobj.group('chann')
1226 title = mobj.group('title')
1227 query = compat_urllib_parse.urlencode({'title': title})
1228 channel_page = self._download_webpage(
1229 'https://www.youtube.com/%s' % channel, title)
1230 link = self._html_search_regex(
1232 <a\s+href="([^"]+)".*?>\s*
1233 <span\s+class="branded-page-module-title-text">\s*
1234 <span[^>]*>.*?%s.*?</span>''' % re.escape(query),
1235 channel_page, 'list')
1236 url = compat_urlparse.urljoin('https://www.youtube.com/', link)
1238 video_re = r'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1240 # sometimes the webpage doesn't contain the videos
1241 # retry until we get them
1242 for i in itertools.count(0):
1243 msg = 'Downloading Youtube mix'
1245 msg += ', retry #%d' % i
1247 webpage = self._download_webpage(url, title, msg)
1248 ids = orderedSet(re.findall(video_re, webpage))
1251 url_results = self._ids_to_results(ids)
1252 return self.playlist_result(url_results, playlist_title=title)
1255 class YoutubeChannelIE(InfoExtractor):
1256 IE_DESC = 'YouTube.com channels'
1257 _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1258 _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
1259 _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'
1260 IE_NAME = 'youtube:channel'
1262 'note': 'paginated channel',
1263 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1264 'playlist_mincount': 91,
1267 def extract_videos_from_page(self, page):
1269 for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
1270 if mobj.group(1) not in ids_in_page:
1271 ids_in_page.append(mobj.group(1))
1274 def _real_extract(self, url):
1275 # Extract channel id
1276 mobj = re.match(self._VALID_URL, url)
1278 raise ExtractorError('Invalid URL: %s' % url)
1280 # Download channel page
1281 channel_id = mobj.group(1)
1283 url = 'https://www.youtube.com/channel/%s/videos' % channel_id
1284 channel_page = self._download_webpage(url, channel_id)
1285 autogenerated = re.search(r'''(?x)
1287 channel-header-autogenerated-label|
1288 yt-channel-title-autogenerated
1289 )[^"]*"''', channel_page) is not None
1292 # The videos are contained in a single page
1293 # the ajax pages can't be used, they are empty
1294 video_ids = self.extract_videos_from_page(channel_page)
1296 # Download all channel pages using the json-based channel_ajax query
1297 for pagenum in itertools.count(1):
1298 url = self._MORE_PAGES_URL % (pagenum, channel_id)
1299 page = self._download_json(
1300 url, channel_id, note='Downloading page #%s' % pagenum,
1301 transform_source=uppercase_escape)
1303 ids_in_page = self.extract_videos_from_page(page['content_html'])
1304 video_ids.extend(ids_in_page)
1306 if self._MORE_PAGES_INDICATOR not in page['load_more_widget_html']:
1309 self._downloader.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
1311 url_entries = [self.url_result(video_id, 'Youtube', video_id=video_id)
1312 for video_id in video_ids]
1313 return self.playlist_result(url_entries, channel_id)
1316 class YoutubeUserIE(InfoExtractor):
1317 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1318 _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1319 _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s'
1320 _GDATA_PAGE_SIZE = 50
1321 _GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1322 IE_NAME = 'youtube:user'
1325 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1326 'playlist_mincount': 320,
1328 'title': 'TheLinuxFoundation',
1331 'url': 'ytuser:phihag',
1332 'only_matching': True,
1336 def suitable(cls, url):
1337 # Don't return True if the url can be extracted with other youtube
1338 # extractor, the regex would is too permissive and it would match.
1339 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1340 if any(ie.suitable(url) for ie in other_ies):
1343 return super(YoutubeUserIE, cls).suitable(url)
1345 def _real_extract(self, url):
1347 mobj = re.match(self._VALID_URL, url)
1349 raise ExtractorError('Invalid URL: %s' % url)
1351 username = mobj.group(1)
1353 # Download video ids using YouTube Data API. Result size per
1354 # query is limited (currently to 50 videos) so we need to query
1355 # page by page until there are no video ids - it means we got
1358 def download_page(pagenum):
1359 start_index = pagenum * self._GDATA_PAGE_SIZE + 1
1361 gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
1362 page = self._download_webpage(
1363 gdata_url, username,
1364 'Downloading video ids from %d to %d' % (
1365 start_index, start_index + self._GDATA_PAGE_SIZE))
1368 response = json.loads(page)
1369 except ValueError as err:
1370 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
1371 if 'entry' not in response['feed']:
1374 # Extract video identifiers
1375 entries = response['feed']['entry']
1376 for entry in entries:
1377 title = entry['title']['$t']
1378 video_id = entry['id']['$t'].split('/')[-1]
1382 'ie_key': 'Youtube',
1386 url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
1388 return self.playlist_result(url_results, playlist_title=username)
1391 class YoutubeSearchIE(SearchInfoExtractor):
1392 IE_DESC = 'YouTube.com searches'
1393 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1395 IE_NAME = 'youtube:search'
1396 _SEARCH_KEY = 'ytsearch'
1398 def _get_n_results(self, query, n):
1399 """Get a specified number of results for a query"""
1406 while (PAGE_SIZE * pagenum) < limit:
1407 result_url = self._API_URL % (
1408 compat_urllib_parse.quote_plus(query.encode('utf-8')),
1409 (PAGE_SIZE * pagenum) + 1)
1410 data_json = self._download_webpage(
1411 result_url, video_id='query "%s"' % query,
1412 note='Downloading page %s' % (pagenum + 1),
1413 errnote='Unable to download API page')
1414 data = json.loads(data_json)
1415 api_response = data['data']
1417 if 'items' not in api_response:
1418 raise ExtractorError(
1419 '[youtube] No video results', expected=True)
1421 new_ids = list(video['id'] for video in api_response['items'])
1422 video_ids += new_ids
1424 limit = min(n, api_response['totalItems'])
1427 if len(video_ids) > n:
1428 video_ids = video_ids[:n]
1429 videos = [self.url_result(video_id, 'Youtube', video_id=video_id)
1430 for video_id in video_ids]
1431 return self.playlist_result(videos, query)
1434 class YoutubeSearchDateIE(YoutubeSearchIE):
1435 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1436 _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1437 _SEARCH_KEY = 'ytsearchdate'
1438 IE_DESC = 'YouTube.com searches, newest videos first'
1441 class YoutubeSearchURLIE(InfoExtractor):
1442 IE_DESC = 'YouTube.com search URLs'
1443 IE_NAME = 'youtube:search_url'
1444 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1446 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1447 'playlist_mincount': 5,
1449 'title': 'youtube-dl test video',
1453 def _real_extract(self, url):
1454 mobj = re.match(self._VALID_URL, url)
1455 query = compat_urllib_parse.unquote_plus(mobj.group('query'))
1457 webpage = self._download_webpage(url, query)
1458 result_code = self._search_regex(
1459 r'(?s)<ol class="item-section"(.*?)</ol>', webpage, 'result HTML')
1461 part_codes = re.findall(
1462 r'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code)
1464 for part_code in part_codes:
1465 part_title = self._html_search_regex(
1466 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1467 part_url_snippet = self._html_search_regex(
1468 r'(?s)href="([^"]+)"', part_code, 'item URL')
1469 part_url = compat_urlparse.urljoin(
1470 'https://www.youtube.com/', part_url_snippet)
1474 'title': part_title,
1478 '_type': 'playlist',
1484 class YoutubeShowIE(InfoExtractor):
1485 IE_DESC = 'YouTube.com (multi-season) shows'
1486 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1487 IE_NAME = 'youtube:show'
1489 'url': 'http://www.youtube.com/show/airdisasters',
1490 'playlist_mincount': 3,
1492 'id': 'airdisasters',
1493 'title': 'Air Disasters',
1497 def _real_extract(self, url):
1498 mobj = re.match(self._VALID_URL, url)
1499 playlist_id = mobj.group('id')
1500 webpage = self._download_webpage(
1501 url, playlist_id, 'Downloading show webpage')
1502 # There's one playlist for each season of the show
1503 m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1504 self.to_screen('%s: Found %s seasons' % (playlist_id, len(m_seasons)))
1507 'https://www.youtube.com' + season.group(1), 'YoutubePlaylist')
1508 for season in m_seasons
1510 title = self._og_search_title(webpage, fatal=False)
1513 '_type': 'playlist',
1520 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1522 Base class for extractors that fetch info from
1523 http://www.youtube.com/feed_ajax
1524 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1526 _LOGIN_REQUIRED = True
1527 # use action_load_personal_feed instead of action_load_system_feed
1528 _PERSONAL_FEED = False
1531 def _FEED_TEMPLATE(self):
1532 action = 'action_load_system_feed'
1533 if self._PERSONAL_FEED:
1534 action = 'action_load_personal_feed'
1535 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1539 return 'youtube:%s' % self._FEED_NAME
1541 def _real_initialize(self):
1544 def _real_extract(self, url):
1547 for i in itertools.count(1):
1548 info = self._download_json(self._FEED_TEMPLATE % paging,
1549 '%s feed' % self._FEED_NAME,
1550 'Downloading page %s' % i)
1551 feed_html = info.get('feed_html') or info.get('content_html')
1552 load_more_widget_html = info.get('load_more_widget_html') or feed_html
1553 m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1554 ids = orderedSet(m.group(1) for m in m_ids)
1555 feed_entries.extend(
1556 self.url_result(video_id, 'Youtube', video_id=video_id)
1557 for video_id in ids)
1559 r'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1560 load_more_widget_html)
1563 paging = mobj.group('paging')
1564 return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1567 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1568 IE_DESC = 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1569 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1570 _FEED_NAME = 'recommended'
1571 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1574 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1575 IE_DESC = 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1576 _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1577 _FEED_NAME = 'watch_later'
1578 _PLAYLIST_TITLE = 'Youtube Watch Later'
1579 _PERSONAL_FEED = True
1582 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1583 IE_DESC = 'Youtube watch history, "ythistory" keyword (requires authentication)'
1584 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1585 _FEED_NAME = 'history'
1586 _PERSONAL_FEED = True
1587 _PLAYLIST_TITLE = 'Youtube Watch History'
1590 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1591 IE_NAME = 'youtube:favorites'
1592 IE_DESC = 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1593 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1594 _LOGIN_REQUIRED = True
1596 def _real_extract(self, url):
1597 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1598 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1599 return self.url_result(playlist_id, 'YoutubePlaylist')
1602 class YoutubeSubscriptionsIE(YoutubePlaylistIE):
1603 IE_NAME = 'youtube:subscriptions'
1604 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1605 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1608 def _real_extract(self, url):
1609 title = 'Youtube Subscriptions'
1610 page = self._download_webpage('https://www.youtube.com/feed/subscriptions', title)
1612 # The extraction process is the same as for playlists, but the regex
1613 # for the video ids doesn't contain an index
1615 more_widget_html = content_html = page
1617 for page_num in itertools.count(1):
1618 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1619 new_ids = orderedSet(matches)
1622 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1626 more = self._download_json(
1627 'https://youtube.com/%s' % mobj.group('more'), title,
1628 'Downloading page #%s' % page_num,
1629 transform_source=uppercase_escape)
1630 content_html = more['content_html']
1631 more_widget_html = more['load_more_widget_html']
1634 '_type': 'playlist',
1636 'entries': self._ids_to_results(ids),
1640 class YoutubeTruncatedURLIE(InfoExtractor):
1641 IE_NAME = 'youtube:truncated_url'
1642 IE_DESC = False # Do not list
1643 _VALID_URL = r'''(?x)
1644 (?:https?://)?[^/]+/watch\?(?:
1646 annotation_id=annotation_[^&]+
1648 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1652 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1653 'only_matching': True,
1655 'url': 'http://www.youtube.com/watch?',
1656 'only_matching': True,
1659 def _real_extract(self, url):
1660 raise ExtractorError(
1661 'Did you forget to quote the URL? Remember that & is a meta '
1662 'character in most shells, so you want to put the URL in quotes, '
1664 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1665 ' or simply youtube-dl BaW_jenozKc .',