Merge branch 'master' into subtitles_rework
[youtube-dl] / youtube_dl / extractor / youtube.py
1 # coding: utf-8
2
3 import json
4 import netrc
5 import re
6 import socket
7 import itertools
8
9 from .common import InfoExtractor, SearchInfoExtractor
10 from .subtitles import SubtitlesIE
11 from ..utils import (
12     compat_http_client,
13     compat_parse_qs,
14     compat_urllib_error,
15     compat_urllib_parse,
16     compat_urllib_request,
17     compat_str,
18
19     clean_html,
20     get_element_by_id,
21     ExtractorError,
22     unescapeHTML,
23     unified_strdate,
24     orderedSet,
25 )
26
27 class YoutubeBaseInfoExtractor(InfoExtractor):
28     """Provide base functions for Youtube extractors"""
29     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
30     _LANG_URL = r'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
31     _AGE_URL = 'http://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
32     _NETRC_MACHINE = 'youtube'
33     # If True it will raise an error if no login info is provided
34     _LOGIN_REQUIRED = False
35
36     def report_lang(self):
37         """Report attempt to set language."""
38         self.to_screen(u'Setting language')
39
40     def _set_language(self):
41         request = compat_urllib_request.Request(self._LANG_URL)
42         try:
43             self.report_lang()
44             compat_urllib_request.urlopen(request).read()
45         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
46             self._downloader.report_warning(u'unable to set language: %s' % compat_str(err))
47             return False
48         return True
49
50     def _login(self):
51         (username, password) = self._get_login_info()
52         # No authentication to be performed
53         if username is None:
54             if self._LOGIN_REQUIRED:
55                 raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
56             return False
57
58         request = compat_urllib_request.Request(self._LOGIN_URL)
59         try:
60             login_page = compat_urllib_request.urlopen(request).read().decode('utf-8')
61         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
62             self._downloader.report_warning(u'unable to fetch login page: %s' % compat_str(err))
63             return False
64
65         galx = None
66         dsh = None
67         match = re.search(re.compile(r'<input.+?name="GALX".+?value="(.+?)"', re.DOTALL), login_page)
68         if match:
69           galx = match.group(1)
70         match = re.search(re.compile(r'<input.+?name="dsh".+?value="(.+?)"', re.DOTALL), login_page)
71         if match:
72           dsh = match.group(1)
73
74         # Log in
75         login_form_strs = {
76                 u'continue': u'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
77                 u'Email': username,
78                 u'GALX': galx,
79                 u'Passwd': password,
80                 u'PersistentCookie': u'yes',
81                 u'_utf8': u'霱',
82                 u'bgresponse': u'js_disabled',
83                 u'checkConnection': u'',
84                 u'checkedDomains': u'youtube',
85                 u'dnConn': u'',
86                 u'dsh': dsh,
87                 u'pstMsg': u'0',
88                 u'rmShown': u'1',
89                 u'secTok': u'',
90                 u'signIn': u'Sign in',
91                 u'timeStmp': u'',
92                 u'service': u'youtube',
93                 u'uilel': u'3',
94                 u'hl': u'en_US',
95         }
96         # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
97         # chokes on unicode
98         login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
99         login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
100         request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
101         try:
102             self.report_login()
103             login_results = compat_urllib_request.urlopen(request).read().decode('utf-8')
104             if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
105                 self._downloader.report_warning(u'unable to log in: bad username or password')
106                 return False
107         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
108             self._downloader.report_warning(u'unable to log in: %s' % compat_str(err))
109             return False
110         return True
111
112     def _confirm_age(self):
113         age_form = {
114                 'next_url':     '/',
115                 'action_confirm':   'Confirm',
116                 }
117         request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
118         try:
119             self.report_age_confirmation()
120             compat_urllib_request.urlopen(request).read().decode('utf-8')
121         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
122             raise ExtractorError(u'Unable to confirm age: %s' % compat_str(err))
123         return True
124
125     def _real_initialize(self):
126         if self._downloader is None:
127             return
128         if not self._set_language():
129             return
130         if not self._login():
131             return
132         self._confirm_age()
133
134 class YoutubeSubtitlesIE(SubtitlesIE):
135
136     def _get_available_subtitles(self, video_id):
137         request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id)
138         try:
139             sub_list = compat_urllib_request.urlopen(request).read().decode('utf-8')
140         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
141             self._downloader.report_warning(u'unable to download video subtitles: %s' % compat_str(err))
142             return {}
143         lang_list = re.findall(r'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list)
144
145         sub_lang_list = {}
146         for l in lang_list:
147             lang = l[1]
148             params = compat_urllib_parse.urlencode({
149                 'lang': lang,
150                 'v': video_id,
151                 'fmt': self._downloader.params.get('subtitlesformat'),
152             })
153             url = u'http://www.youtube.com/api/timedtext?' + params
154             sub_lang_list[lang] = url
155         if not sub_lang_list:
156             self._downloader.report_warning(u'video doesn\'t have subtitles')
157             return {}
158         return sub_lang_list
159
160     def _request_automatic_caption(self, video_id, webpage):
161         """We need the webpage for getting the captions url, pass it as an
162            argument to speed up the process."""
163         sub_lang = (self._downloader.params.get('subtitleslangs') or ['en'])[0]
164         sub_format = self._downloader.params.get('subtitlesformat')
165         self.to_screen(u'%s: Looking for automatic captions' % video_id)
166         mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
167         err_msg = u'Couldn\'t find automatic captions for "%s"' % sub_lang
168         if mobj is None:
169             self._downloader.report_warning(err_msg)
170             return {}
171         player_config = json.loads(mobj.group(1))
172         try:
173             args = player_config[u'args']
174             caption_url = args[u'ttsurl']
175             timestamp = args[u'timestamp']
176             params = compat_urllib_parse.urlencode({
177                 'lang': 'en',
178                 'tlang': sub_lang,
179                 'fmt': sub_format,
180                 'ts': timestamp,
181                 'kind': 'asr',
182             })
183             subtitles_url = caption_url + '&' + params
184             sub = self._download_webpage(subtitles_url, video_id, u'Downloading automatic captions')
185             return {sub_lang: sub}
186         # An extractor error can be raise by the download process if there are
187         # no automatic captions but there are subtitles
188         except (KeyError, ExtractorError):
189             self._downloader.report_warning(err_msg)
190             return {}
191
192 class YoutubeIE(YoutubeSubtitlesIE, YoutubeBaseInfoExtractor):
193     IE_DESC = u'YouTube.com'
194     _VALID_URL = r"""^
195                      (
196                          (?:https?://)?                                       # http(s):// (optional)
197                          (?:(?:(?:(?:\w+\.)?youtube(?:-nocookie)?\.com/|
198                             tube\.majestyc\.net/)                             # the various hostnames, with wildcard subdomains
199                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
200                          (?:                                                  # the various things that can precede the ID:
201                              (?:(?:v|embed|e)/)                               # v/ or embed/ or e/
202                              |(?:                                             # or the v= param in all its forms
203                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?)?    # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
204                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
205                                  (?:.*?&)?                                    # any other preceding param (like /?s=tuff&v=xxxx)
206                                  v=
207                              )
208                          ))
209                          |youtu\.be/                                          # just youtu.be/xxxx
210                          )
211                      )?                                                       # all until now is optional -> you can pass the naked ID
212                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
213                      (?(1).+)?                                                # if we found the ID, everything can follow
214                      $"""
215     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
216     # Listed in order of quality
217     _available_formats = ['38', '37', '46', '22', '45', '35', '44', '34', '18', '43', '6', '5', '36', '17', '13',
218                           # Apple HTTP Live Streaming
219                           '96', '95', '94', '93', '92', '132', '151',
220                           # 3D
221                           '85', '84', '102', '83', '101', '82', '100',
222                           # Dash video
223                           '138', '137', '248', '136', '247', '135', '246',
224                           '245', '244', '134', '243', '133', '242', '160',
225                           # Dash audio
226                           '141', '172', '140', '171', '139',
227                           ]
228     _available_formats_prefer_free = ['38', '46', '37', '45', '22', '44', '35', '43', '34', '18', '6', '5', '36', '17', '13',
229                                       # Apple HTTP Live Streaming
230                                       '96', '95', '94', '93', '92', '132', '151',
231                                       # 3D
232                                       '85', '102', '84', '101', '83', '100', '82',
233                                       # Dash video
234                                       '138', '248', '137', '247', '136', '246', '245',
235                                       '244', '135', '243', '134', '242', '133', '160',
236                                       # Dash audio
237                                       '172', '141', '171', '140', '139',
238                                       ]
239     _video_formats_map = {
240         'flv': ['35', '34', '6', '5'],
241         '3gp': ['36', '17', '13'],
242         'mp4': ['38', '37', '22', '18'],
243         'webm': ['46', '45', '44', '43'],
244     }
245     _video_extensions = {
246         '13': '3gp',
247         '17': '3gp',
248         '18': 'mp4',
249         '22': 'mp4',
250         '36': '3gp',
251         '37': 'mp4',
252         '38': 'mp4',
253         '43': 'webm',
254         '44': 'webm',
255         '45': 'webm',
256         '46': 'webm',
257
258         # 3d videos
259         '82': 'mp4',
260         '83': 'mp4',
261         '84': 'mp4',
262         '85': 'mp4',
263         '100': 'webm',
264         '101': 'webm',
265         '102': 'webm',
266
267         # Apple HTTP Live Streaming
268         '92': 'mp4',
269         '93': 'mp4',
270         '94': 'mp4',
271         '95': 'mp4',
272         '96': 'mp4',
273         '132': 'mp4',
274         '151': 'mp4',
275
276         # Dash mp4
277         '133': 'mp4',
278         '134': 'mp4',
279         '135': 'mp4',
280         '136': 'mp4',
281         '137': 'mp4',
282         '138': 'mp4',
283         '139': 'mp4',
284         '140': 'mp4',
285         '141': 'mp4',
286         '160': 'mp4',
287
288         # Dash webm
289         '171': 'webm',
290         '172': 'webm',
291         '242': 'webm',
292         '243': 'webm',
293         '244': 'webm',
294         '245': 'webm',
295         '246': 'webm',
296         '247': 'webm',
297         '248': 'webm',
298     }
299     _video_dimensions = {
300         '5': '240x400',
301         '6': '???',
302         '13': '???',
303         '17': '144x176',
304         '18': '360x640',
305         '22': '720x1280',
306         '34': '360x640',
307         '35': '480x854',
308         '36': '240x320',
309         '37': '1080x1920',
310         '38': '3072x4096',
311         '43': '360x640',
312         '44': '480x854',
313         '45': '720x1280',
314         '46': '1080x1920',
315         '82': '360p',
316         '83': '480p',
317         '84': '720p',
318         '85': '1080p',
319         '92': '240p',
320         '93': '360p',
321         '94': '480p',
322         '95': '720p',
323         '96': '1080p',
324         '100': '360p',
325         '101': '480p',
326         '102': '720p',
327         '132': '240p',
328         '151': '72p',
329         '133': '240p',
330         '134': '360p',
331         '135': '480p',
332         '136': '720p',
333         '137': '1080p',
334         '138': '>1080p',
335         '139': '48k',
336         '140': '128k',
337         '141': '256k',
338         '160': '192p',
339         '171': '128k',
340         '172': '256k',
341         '242': '240p',
342         '243': '360p',
343         '244': '480p',
344         '245': '480p',
345         '246': '480p',
346         '247': '720p',
347         '248': '1080p',
348     }
349     _special_itags = {
350         '82': '3D',
351         '83': '3D',
352         '84': '3D',
353         '85': '3D',
354         '100': '3D',
355         '101': '3D',
356         '102': '3D',
357         '133': 'DASH Video',
358         '134': 'DASH Video',
359         '135': 'DASH Video',
360         '136': 'DASH Video',
361         '137': 'DASH Video',
362         '138': 'DASH Video',
363         '139': 'DASH Audio',
364         '140': 'DASH Audio',
365         '141': 'DASH Audio',
366         '160': 'DASH Video',
367         '171': 'DASH Audio',
368         '172': 'DASH Audio',
369         '242': 'DASH Video',
370         '243': 'DASH Video',
371         '244': 'DASH Video',
372         '245': 'DASH Video',
373         '246': 'DASH Video',
374         '247': 'DASH Video',
375         '248': 'DASH Video',
376     }
377
378     IE_NAME = u'youtube'
379     _TESTS = [
380         {
381             u"url":  u"http://www.youtube.com/watch?v=BaW_jenozKc",
382             u"file":  u"BaW_jenozKc.mp4",
383             u"info_dict": {
384                 u"title": u"youtube-dl test video \"'/\\ä↭𝕐",
385                 u"uploader": u"Philipp Hagemeister",
386                 u"uploader_id": u"phihag",
387                 u"upload_date": u"20121002",
388                 u"description": u"test chars:  \"'/\\ä↭𝕐\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de ."
389             }
390         },
391         {
392             u"url":  u"http://www.youtube.com/watch?v=1ltcDfZMA3U",
393             u"file":  u"1ltcDfZMA3U.flv",
394             u"note": u"Test VEVO video (#897)",
395             u"info_dict": {
396                 u"upload_date": u"20070518",
397                 u"title": u"Maps - It Will Find You",
398                 u"description": u"Music video by Maps performing It Will Find You.",
399                 u"uploader": u"MuteUSA",
400                 u"uploader_id": u"MuteUSA"
401             }
402         },
403         {
404             u"url":  u"http://www.youtube.com/watch?v=UxxajLWwzqY",
405             u"file":  u"UxxajLWwzqY.mp4",
406             u"note": u"Test generic use_cipher_signature video (#897)",
407             u"info_dict": {
408                 u"upload_date": u"20120506",
409                 u"title": u"Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]",
410                 u"description": u"md5:3e2666e0a55044490499ea45fe9037b7",
411                 u"uploader": u"Icona Pop",
412                 u"uploader_id": u"IconaPop"
413             }
414         },
415         {
416             u"url":  u"https://www.youtube.com/watch?v=07FYdnEawAQ",
417             u"file":  u"07FYdnEawAQ.mp4",
418             u"note": u"Test VEVO video with age protection (#956)",
419             u"info_dict": {
420                 u"upload_date": u"20130703",
421                 u"title": u"Justin Timberlake - Tunnel Vision (Explicit)",
422                 u"description": u"md5:64249768eec3bc4276236606ea996373",
423                 u"uploader": u"justintimberlakeVEVO",
424                 u"uploader_id": u"justintimberlakeVEVO"
425             }
426         },
427         {
428             u'url': u'https://www.youtube.com/watch?v=TGi3HqYrWHE',
429             u'file': u'TGi3HqYrWHE.mp4',
430             u'note': u'm3u8 video',
431             u'info_dict': {
432                 u'title': u'Triathlon - Men - London 2012 Olympic Games',
433                 u'description': u'- Men -  TR02 - Triathlon - 07 August 2012 - London 2012 Olympic Games',
434                 u'uploader': u'olympic',
435                 u'upload_date': u'20120807',
436                 u'uploader_id': u'olympic',
437             },
438             u'params': {
439                 u'skip_download': True,
440             },
441         },
442     ]
443
444
445     @classmethod
446     def suitable(cls, url):
447         """Receives a URL and returns True if suitable for this IE."""
448         if YoutubePlaylistIE.suitable(url): return False
449         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
450
451     def report_video_webpage_download(self, video_id):
452         """Report attempt to download video webpage."""
453         self.to_screen(u'%s: Downloading video webpage' % video_id)
454
455     def report_video_info_webpage_download(self, video_id):
456         """Report attempt to download video info webpage."""
457         self.to_screen(u'%s: Downloading video info webpage' % video_id)
458
459     def report_information_extraction(self, video_id):
460         """Report attempt to extract video information."""
461         self.to_screen(u'%s: Extracting video information' % video_id)
462
463     def report_unavailable_format(self, video_id, format):
464         """Report extracted video URL."""
465         self.to_screen(u'%s: Format %s not available' % (video_id, format))
466
467     def report_rtmp_download(self):
468         """Indicate the download will use the RTMP protocol."""
469         self.to_screen(u'RTMP download detected')
470
471     def _decrypt_signature(self, s):
472         """Turn the encrypted s field into a working signature"""
473
474         if len(s) == 92:
475             return s[25] + s[3:25] + s[0] + s[26:42] + s[79] + s[43:79] + s[91] + s[80:83]
476         elif len(s) == 90:
477             return s[25] + s[3:25] + s[2] + s[26:40] + s[77] + s[41:77] + s[89] + s[78:81]
478         elif len(s) == 89:
479             return s[84:78:-1] + s[87] + s[77:60:-1] + s[0] + s[59:3:-1]
480         elif len(s) == 88:
481             return s[7:28] + s[87] + s[29:45] + s[55] + s[46:55] + s[2] + s[56:87] + s[28]
482         elif len(s) == 87:
483             return s[6:27] + s[4] + s[28:39] + s[27] + s[40:59] + s[2] + s[60:]
484         elif len(s) == 86:
485             return s[5:34] + s[0] + s[35:38] + s[3] + s[39:45] + s[38] + s[46:53] + s[73] + s[54:73] + s[85] + s[74:85] + s[53]
486         elif len(s) == 85:
487             return s[40] + s[82:43:-1] + s[22] + s[42:40:-1] + s[83] + s[39:22:-1] + s[0] + s[21:2:-1]
488         elif len(s) == 84:
489             return s[81:36:-1] + s[0] + s[35:2:-1]
490         elif len(s) == 83:
491             return s[81:64:-1] + s[82] + s[63:52:-1] + s[45] + s[51:45:-1] + s[1] + s[44:1:-1] + s[0]
492         elif len(s) == 82:
493             return s[1:19] + s[0] + s[20:68] + s[19] + s[69:82]
494         elif len(s) == 81:
495             return s[56] + s[79:56:-1] + s[41] + s[55:41:-1] + s[80] + s[40:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9]
496         elif len(s) == 80:
497             return s[1:19] + s[0] + s[20:68] + s[19] + s[69:80]
498         elif len(s) == 79:
499             return s[54] + s[77:54:-1] + s[39] + s[53:39:-1] + s[78] + s[38:34:-1] + s[0] + s[33:29:-1] + s[34] + s[28:9:-1] + s[29] + s[8:0:-1] + s[9]
500
501         else:
502             raise ExtractorError(u'Unable to decrypt signature, key length %d not supported; retrying might work' % (len(s)))
503
504     def _decrypt_signature_age_gate(self, s):
505         # The videos with age protection use another player, so the algorithms
506         # can be different.
507         if len(s) == 86:
508             return s[2:63] + s[82] + s[64:82] + s[63]
509         else:
510             # Fallback to the other algortihms
511             return self._decrypt_signature(s)
512
513     def _print_formats(self, formats):
514         print('Available formats:')
515         for x in formats:
516             print('%s\t:\t%s\t[%s]%s' %(x, self._video_extensions.get(x, 'flv'),
517                                         self._video_dimensions.get(x, '???'),
518                                         ' ('+self._special_itags[x]+')' if x in self._special_itags else ''))
519
520     def _extract_id(self, url):
521         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
522         if mobj is None:
523             raise ExtractorError(u'Invalid URL: %s' % url)
524         video_id = mobj.group(2)
525         return video_id
526
527     def _get_video_url_list(self, url_map):
528         """
529         Transform a dictionary in the format {itag:url} to a list of (itag, url)
530         with the requested formats.
531         """
532         req_format = self._downloader.params.get('format', None)
533         format_limit = self._downloader.params.get('format_limit', None)
534         available_formats = self._available_formats_prefer_free if self._downloader.params.get('prefer_free_formats', False) else self._available_formats
535         if format_limit is not None and format_limit in available_formats:
536             format_list = available_formats[available_formats.index(format_limit):]
537         else:
538             format_list = available_formats
539         existing_formats = [x for x in format_list if x in url_map]
540         if len(existing_formats) == 0:
541             raise ExtractorError(u'no known formats available for video')
542         if self._downloader.params.get('listformats', None):
543             self._print_formats(existing_formats)
544             return
545         if req_format is None or req_format == 'best':
546             video_url_list = [(existing_formats[0], url_map[existing_formats[0]])] # Best quality
547         elif req_format == 'worst':
548             video_url_list = [(existing_formats[-1], url_map[existing_formats[-1]])] # worst quality
549         elif req_format in ('-1', 'all'):
550             video_url_list = [(f, url_map[f]) for f in existing_formats] # All formats
551         else:
552             # Specific formats. We pick the first in a slash-delimeted sequence.
553             # Format can be specified as itag or 'mp4' or 'flv' etc. We pick the highest quality
554             # available in the specified format. For example,
555             # if '1/2/3/4' is requested and '2' and '4' are available, we pick '2'.
556             # if '1/mp4/3/4' is requested and '1' and '5' (is a mp4) are available, we pick '1'.
557             # if '1/mp4/3/4' is requested and '4' and '5' (is a mp4) are available, we pick '5'.
558             req_formats = req_format.split('/')
559             video_url_list = None
560             for rf in req_formats:
561                 if rf in url_map:
562                     video_url_list = [(rf, url_map[rf])]
563                     break
564                 if rf in self._video_formats_map:
565                     for srf in self._video_formats_map[rf]:
566                         if srf in url_map:
567                             video_url_list = [(srf, url_map[srf])]
568                             break
569                     else:
570                         continue
571                     break
572             if video_url_list is None:
573                 raise ExtractorError(u'requested format not available')
574         return video_url_list
575
576     def _extract_from_m3u8(self, manifest_url, video_id):
577         url_map = {}
578         def _get_urls(_manifest):
579             lines = _manifest.split('\n')
580             urls = filter(lambda l: l and not l.startswith('#'),
581                             lines)
582             return urls
583         manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
584         formats_urls = _get_urls(manifest)
585         for format_url in formats_urls:
586             itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
587             url_map[itag] = format_url
588         return url_map
589
590     def _real_extract(self, url):
591         if re.match(r'(?:https?://)?[^/]+/watch\?feature=[a-z_]+$', url):
592             self._downloader.report_warning(u'Did you forget to quote the URL? Remember that & is a meta-character in most shells, so you want to put the URL in quotes, like  youtube-dl \'http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc\' (or simply  youtube-dl BaW_jenozKc  ).')
593
594         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
595         mobj = re.search(self._NEXT_URL_RE, url)
596         if mobj:
597             url = 'https://www.youtube.com/' + compat_urllib_parse.unquote(mobj.group(1)).lstrip('/')
598         video_id = self._extract_id(url)
599
600         # Get video webpage
601         self.report_video_webpage_download(video_id)
602         url = 'https://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
603         request = compat_urllib_request.Request(url)
604         try:
605             video_webpage_bytes = compat_urllib_request.urlopen(request).read()
606         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
607             raise ExtractorError(u'Unable to download video webpage: %s' % compat_str(err))
608
609         video_webpage = video_webpage_bytes.decode('utf-8', 'ignore')
610
611         # Attempt to extract SWF player URL
612         mobj = re.search(r'swfConfig.*?"(http:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
613         if mobj is not None:
614             player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
615         else:
616             player_url = None
617
618         # Get video info
619         self.report_video_info_webpage_download(video_id)
620         if re.search(r'player-age-gate-content">', video_webpage) is not None:
621             self.report_age_confirmation()
622             age_gate = True
623             # We simulate the access to the video from www.youtube.com/v/{video_id}
624             # this can be viewed without login into Youtube
625             data = compat_urllib_parse.urlencode({'video_id': video_id,
626                                                   'el': 'embedded',
627                                                   'gl': 'US',
628                                                   'hl': 'en',
629                                                   'eurl': 'https://youtube.googleapis.com/v/' + video_id,
630                                                   'asv': 3,
631                                                   'sts':'1588',
632                                                   })
633             video_info_url = 'https://www.youtube.com/get_video_info?' + data
634             video_info_webpage = self._download_webpage(video_info_url, video_id,
635                                     note=False,
636                                     errnote='unable to download video info webpage')
637             video_info = compat_parse_qs(video_info_webpage)
638         else:
639             age_gate = False
640             for el_type in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
641                 video_info_url = ('https://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
642                         % (video_id, el_type))
643                 video_info_webpage = self._download_webpage(video_info_url, video_id,
644                                         note=False,
645                                         errnote='unable to download video info webpage')
646                 video_info = compat_parse_qs(video_info_webpage)
647                 if 'token' in video_info:
648                     break
649         if 'token' not in video_info:
650             if 'reason' in video_info:
651                 raise ExtractorError(u'YouTube said: %s' % video_info['reason'][0], expected=True)
652             else:
653                 raise ExtractorError(u'"token" parameter not in video info for unknown reason')
654
655         # Check for "rental" videos
656         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
657             raise ExtractorError(u'"rental" videos not supported')
658
659         # Start extracting information
660         self.report_information_extraction(video_id)
661
662         # uploader
663         if 'author' not in video_info:
664             raise ExtractorError(u'Unable to extract uploader name')
665         video_uploader = compat_urllib_parse.unquote_plus(video_info['author'][0])
666
667         # uploader_id
668         video_uploader_id = None
669         mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
670         if mobj is not None:
671             video_uploader_id = mobj.group(1)
672         else:
673             self._downloader.report_warning(u'unable to extract uploader nickname')
674
675         # title
676         if 'title' not in video_info:
677             raise ExtractorError(u'Unable to extract video title')
678         video_title = compat_urllib_parse.unquote_plus(video_info['title'][0])
679
680         # thumbnail image
681         # We try first to get a high quality image:
682         m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
683                             video_webpage, re.DOTALL)
684         if m_thumb is not None:
685             video_thumbnail = m_thumb.group(1)
686         elif 'thumbnail_url' not in video_info:
687             self._downloader.report_warning(u'unable to extract video thumbnail')
688             video_thumbnail = ''
689         else:   # don't panic if we can't find it
690             video_thumbnail = compat_urllib_parse.unquote_plus(video_info['thumbnail_url'][0])
691
692         # upload date
693         upload_date = None
694         mobj = re.search(r'id="eow-date.*?>(.*?)</span>', video_webpage, re.DOTALL)
695         if mobj is not None:
696             upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
697             upload_date = unified_strdate(upload_date)
698
699         # description
700         video_description = get_element_by_id("eow-description", video_webpage)
701         if video_description:
702             video_description = clean_html(video_description)
703         else:
704             fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
705             if fd_mobj:
706                 video_description = unescapeHTML(fd_mobj.group(1))
707             else:
708                 video_description = u''
709
710         # subtitles
711         video_subtitles = None
712
713         if self._downloader.params.get('writesubtitles', False) or self._downloader.params.get('allsubtitles', False):
714             video_subtitles = self._extract_subtitles(video_id)
715         elif self._downloader.params.get('writeautomaticsub', False):
716             video_subtitles = self._request_automatic_caption(video_id, video_webpage)
717
718         if self._downloader.params.get('listsubtitles', False):
719             self._list_available_subtitles(video_id)
720             return
721
722         if 'length_seconds' not in video_info:
723             self._downloader.report_warning(u'unable to extract video duration')
724             video_duration = ''
725         else:
726             video_duration = compat_urllib_parse.unquote_plus(video_info['length_seconds'][0])
727
728         # Decide which formats to download
729
730         try:
731             mobj = re.search(r';ytplayer.config = ({.*?});', video_webpage)
732             if not mobj:
733                 raise ValueError('Could not find vevo ID')
734             info = json.loads(mobj.group(1))
735             args = info['args']
736             # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
737             # this signatures are encrypted
738             m_s = re.search(r'[&,]s=', args['url_encoded_fmt_stream_map'])
739             if m_s is not None:
740                 self.to_screen(u'%s: Encrypted signatures detected.' % video_id)
741                 video_info['url_encoded_fmt_stream_map'] = [args['url_encoded_fmt_stream_map']]
742             m_s = re.search(r'[&,]s=', args.get('adaptive_fmts', u''))
743             if m_s is not None:
744                 if 'url_encoded_fmt_stream_map' in video_info:
745                     video_info['url_encoded_fmt_stream_map'][0] += ',' + args['adaptive_fmts']
746                 else:
747                     video_info['url_encoded_fmt_stream_map'] = [args['adaptive_fmts']]
748             elif 'adaptive_fmts' in video_info:
749                 if 'url_encoded_fmt_stream_map' in video_info:
750                     video_info['url_encoded_fmt_stream_map'][0] += ',' + video_info['adaptive_fmts'][0]
751                 else:
752                     video_info['url_encoded_fmt_stream_map'] = video_info['adaptive_fmts']
753         except ValueError:
754             pass
755
756         if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
757             self.report_rtmp_download()
758             video_url_list = [(None, video_info['conn'][0])]
759         elif 'url_encoded_fmt_stream_map' in video_info and len(video_info['url_encoded_fmt_stream_map']) >= 1:
760             if 'rtmpe%3Dyes' in video_info['url_encoded_fmt_stream_map'][0]:
761                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
762             url_map = {}
763             for url_data_str in video_info['url_encoded_fmt_stream_map'][0].split(','):
764                 url_data = compat_parse_qs(url_data_str)
765                 if 'itag' in url_data and 'url' in url_data:
766                     url = url_data['url'][0]
767                     if 'sig' in url_data:
768                         url += '&signature=' + url_data['sig'][0]
769                     elif 's' in url_data:
770                         if self._downloader.params.get('verbose'):
771                             s = url_data['s'][0]
772                             if age_gate:
773                                 player_version = self._search_regex(r'ad3-(.+?)\.swf',
774                                     video_info['ad3_module'][0] if 'ad3_module' in video_info else 'NOT FOUND',
775                                     'flash player', fatal=False)
776                                 player = 'flash player %s' % player_version
777                             else:
778                                 player = u'html5 player %s' % self._search_regex(r'html5player-(.+?)\.js', video_webpage,
779                                     'html5 player', fatal=False)
780                             parts_sizes = u'.'.join(compat_str(len(part)) for part in s.split('.'))
781                             self.to_screen(u'encrypted signature length %d (%s), itag %s, %s' %
782                                 (len(s), parts_sizes, url_data['itag'][0], player))
783                         encrypted_sig = url_data['s'][0]
784                         if age_gate:
785                             signature = self._decrypt_signature_age_gate(encrypted_sig)
786                         else:
787                             signature = self._decrypt_signature(encrypted_sig)
788                         url += '&signature=' + signature
789                     if 'ratebypass' not in url:
790                         url += '&ratebypass=yes'
791                     url_map[url_data['itag'][0]] = url
792             video_url_list = self._get_video_url_list(url_map)
793             if not video_url_list:
794                 return
795         elif video_info.get('hlsvp'):
796             manifest_url = video_info['hlsvp'][0]
797             url_map = self._extract_from_m3u8(manifest_url, video_id)
798             video_url_list = self._get_video_url_list(url_map)
799             if not video_url_list:
800                 return
801
802         else:
803             raise ExtractorError(u'no conn or url_encoded_fmt_stream_map information found in video info')
804
805         results = []
806         for format_param, video_real_url in video_url_list:
807             # Extension
808             video_extension = self._video_extensions.get(format_param, 'flv')
809
810             video_format = '{0} - {1}{2}'.format(format_param if format_param else video_extension,
811                                               self._video_dimensions.get(format_param, '???'),
812                                               ' ('+self._special_itags[format_param]+')' if format_param in self._special_itags else '')
813
814             results.append({
815                 'id':       video_id,
816                 'url':      video_real_url,
817                 'uploader': video_uploader,
818                 'uploader_id': video_uploader_id,
819                 'upload_date':  upload_date,
820                 'title':    video_title,
821                 'ext':      video_extension,
822                 'format':   video_format,
823                 'thumbnail':    video_thumbnail,
824                 'description':  video_description,
825                 'player_url':   player_url,
826                 'subtitles':    video_subtitles,
827                 'duration':     video_duration
828             })
829         return results
830
831 class YoutubePlaylistIE(InfoExtractor):
832     IE_DESC = u'YouTube.com playlists'
833     _VALID_URL = r"""(?:
834                         (?:https?://)?
835                         (?:\w+\.)?
836                         youtube\.com/
837                         (?:
838                            (?:course|view_play_list|my_playlists|artist|playlist|watch)
839                            \? (?:.*?&)*? (?:p|a|list)=
840                         |  p/
841                         )
842                         ((?:PL|EC|UU|FL)?[0-9A-Za-z-_]{10,})
843                         .*
844                      |
845                         ((?:PL|EC|UU|FL)[0-9A-Za-z-_]{10,})
846                      )"""
847     _TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/playlists/%s?max-results=%i&start-index=%i&v=2&alt=json&safeSearch=none'
848     _MAX_RESULTS = 50
849     IE_NAME = u'youtube:playlist'
850
851     @classmethod
852     def suitable(cls, url):
853         """Receives a URL and returns True if suitable for this IE."""
854         return re.match(cls._VALID_URL, url, re.VERBOSE) is not None
855
856     def _real_extract(self, url):
857         # Extract playlist id
858         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
859         if mobj is None:
860             raise ExtractorError(u'Invalid URL: %s' % url)
861
862         # Download playlist videos from API
863         playlist_id = mobj.group(1) or mobj.group(2)
864         videos = []
865
866         for page_num in itertools.count(1):
867             start_index = self._MAX_RESULTS * (page_num - 1) + 1
868             if start_index >= 1000:
869                 self._downloader.report_warning(u'Max number of results reached')
870                 break
871             url = self._TEMPLATE_URL % (playlist_id, self._MAX_RESULTS, start_index)
872             page = self._download_webpage(url, playlist_id, u'Downloading page #%s' % page_num)
873
874             try:
875                 response = json.loads(page)
876             except ValueError as err:
877                 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
878
879             if 'feed' not in response:
880                 raise ExtractorError(u'Got a malformed response from YouTube API')
881             playlist_title = response['feed']['title']['$t']
882             if 'entry' not in response['feed']:
883                 # Number of videos is a multiple of self._MAX_RESULTS
884                 break
885
886             for entry in response['feed']['entry']:
887                 index = entry['yt$position']['$t']
888                 if 'media$group' in entry and 'yt$videoid' in entry['media$group']:
889                     videos.append((
890                         index,
891                         'https://www.youtube.com/watch?v=' + entry['media$group']['yt$videoid']['$t']
892                     ))
893
894         videos = [v[1] for v in sorted(videos)]
895
896         url_results = [self.url_result(vurl, 'Youtube') for vurl in videos]
897         return [self.playlist_result(url_results, playlist_id, playlist_title)]
898
899
900 class YoutubeChannelIE(InfoExtractor):
901     IE_DESC = u'YouTube.com channels'
902     _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
903     _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
904     _MORE_PAGES_INDICATOR = 'yt-uix-load-more'
905     _MORE_PAGES_URL = 'http://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
906     IE_NAME = u'youtube:channel'
907
908     def extract_videos_from_page(self, page):
909         ids_in_page = []
910         for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&?', page):
911             if mobj.group(1) not in ids_in_page:
912                 ids_in_page.append(mobj.group(1))
913         return ids_in_page
914
915     def _real_extract(self, url):
916         # Extract channel id
917         mobj = re.match(self._VALID_URL, url)
918         if mobj is None:
919             raise ExtractorError(u'Invalid URL: %s' % url)
920
921         # Download channel page
922         channel_id = mobj.group(1)
923         video_ids = []
924         pagenum = 1
925
926         url = self._TEMPLATE_URL % (channel_id, pagenum)
927         page = self._download_webpage(url, channel_id,
928                                       u'Downloading page #%s' % pagenum)
929
930         # Extract video identifiers
931         ids_in_page = self.extract_videos_from_page(page)
932         video_ids.extend(ids_in_page)
933
934         # Download any subsequent channel pages using the json-based channel_ajax query
935         if self._MORE_PAGES_INDICATOR in page:
936             for pagenum in itertools.count(1):
937                 url = self._MORE_PAGES_URL % (pagenum, channel_id)
938                 page = self._download_webpage(url, channel_id,
939                                               u'Downloading page #%s' % pagenum)
940
941                 page = json.loads(page)
942
943                 ids_in_page = self.extract_videos_from_page(page['content_html'])
944                 video_ids.extend(ids_in_page)
945
946                 if self._MORE_PAGES_INDICATOR  not in page['load_more_widget_html']:
947                     break
948
949         self._downloader.to_screen(u'[youtube] Channel %s: Found %i videos' % (channel_id, len(video_ids)))
950
951         urls = ['http://www.youtube.com/watch?v=%s' % id for id in video_ids]
952         url_entries = [self.url_result(eurl, 'Youtube') for eurl in urls]
953         return [self.playlist_result(url_entries, channel_id)]
954
955
956 class YoutubeUserIE(InfoExtractor):
957     IE_DESC = u'YouTube.com user videos (URL or "ytuser" keyword)'
958     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?)|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
959     _TEMPLATE_URL = 'http://gdata.youtube.com/feeds/api/users/%s'
960     _GDATA_PAGE_SIZE = 50
961     _GDATA_URL = 'http://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
962     IE_NAME = u'youtube:user'
963
964     @classmethod
965     def suitable(cls, url):
966         # Don't return True if the url can be extracted with other youtube
967         # extractor, the regex would is too permissive and it would match.
968         other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
969         if any(ie.suitable(url) for ie in other_ies): return False
970         else: return super(YoutubeUserIE, cls).suitable(url)
971
972     def _real_extract(self, url):
973         # Extract username
974         mobj = re.match(self._VALID_URL, url)
975         if mobj is None:
976             raise ExtractorError(u'Invalid URL: %s' % url)
977
978         username = mobj.group(1)
979
980         # Download video ids using YouTube Data API. Result size per
981         # query is limited (currently to 50 videos) so we need to query
982         # page by page until there are no video ids - it means we got
983         # all of them.
984
985         video_ids = []
986
987         for pagenum in itertools.count(0):
988             start_index = pagenum * self._GDATA_PAGE_SIZE + 1
989
990             gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
991             page = self._download_webpage(gdata_url, username,
992                                           u'Downloading video ids from %d to %d' % (start_index, start_index + self._GDATA_PAGE_SIZE))
993
994             try:
995                 response = json.loads(page)
996             except ValueError as err:
997                 raise ExtractorError(u'Invalid JSON in API response: ' + compat_str(err))
998
999             # Extract video identifiers
1000             ids_in_page = []
1001             for entry in response['feed']['entry']:
1002                 ids_in_page.append(entry['id']['$t'].split('/')[-1])
1003             video_ids.extend(ids_in_page)
1004
1005             # A little optimization - if current page is not
1006             # "full", ie. does not contain PAGE_SIZE video ids then
1007             # we can assume that this page is the last one - there
1008             # are no more ids on further pages - no need to query
1009             # again.
1010
1011             if len(ids_in_page) < self._GDATA_PAGE_SIZE:
1012                 break
1013
1014         urls = ['http://www.youtube.com/watch?v=%s' % video_id for video_id in video_ids]
1015         url_results = [self.url_result(rurl, 'Youtube') for rurl in urls]
1016         return [self.playlist_result(url_results, playlist_title = username)]
1017
1018 class YoutubeSearchIE(SearchInfoExtractor):
1019     IE_DESC = u'YouTube.com searches'
1020     _API_URL = 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1021     _MAX_RESULTS = 1000
1022     IE_NAME = u'youtube:search'
1023     _SEARCH_KEY = 'ytsearch'
1024
1025     def report_download_page(self, query, pagenum):
1026         """Report attempt to download search page with given number."""
1027         self._downloader.to_screen(u'[youtube] query "%s": Downloading page %s' % (query, pagenum))
1028
1029     def _get_n_results(self, query, n):
1030         """Get a specified number of results for a query"""
1031
1032         video_ids = []
1033         pagenum = 0
1034         limit = n
1035
1036         while (50 * pagenum) < limit:
1037             self.report_download_page(query, pagenum+1)
1038             result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
1039             request = compat_urllib_request.Request(result_url)
1040             try:
1041                 data = compat_urllib_request.urlopen(request).read().decode('utf-8')
1042             except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1043                 raise ExtractorError(u'Unable to download API page: %s' % compat_str(err))
1044             api_response = json.loads(data)['data']
1045
1046             if not 'items' in api_response:
1047                 raise ExtractorError(u'[youtube] No video results')
1048
1049             new_ids = list(video['id'] for video in api_response['items'])
1050             video_ids += new_ids
1051
1052             limit = min(n, api_response['totalItems'])
1053             pagenum += 1
1054
1055         if len(video_ids) > n:
1056             video_ids = video_ids[:n]
1057         videos = [self.url_result('http://www.youtube.com/watch?v=%s' % id, 'Youtube') for id in video_ids]
1058         return self.playlist_result(videos, query)
1059
1060
1061 class YoutubeShowIE(InfoExtractor):
1062     IE_DESC = u'YouTube.com (multi-season) shows'
1063     _VALID_URL = r'https?://www\.youtube\.com/show/(.*)'
1064     IE_NAME = u'youtube:show'
1065
1066     def _real_extract(self, url):
1067         mobj = re.match(self._VALID_URL, url)
1068         show_name = mobj.group(1)
1069         webpage = self._download_webpage(url, show_name, u'Downloading show webpage')
1070         # There's one playlist for each season of the show
1071         m_seasons = list(re.finditer(r'href="(/playlist\?list=.*?)"', webpage))
1072         self.to_screen(u'%s: Found %s seasons' % (show_name, len(m_seasons)))
1073         return [self.url_result('https://www.youtube.com' + season.group(1), 'YoutubePlaylist') for season in m_seasons]
1074
1075
1076 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1077     """
1078     Base class for extractors that fetch info from
1079     http://www.youtube.com/feed_ajax
1080     Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1081     """
1082     _LOGIN_REQUIRED = True
1083     _PAGING_STEP = 30
1084     # use action_load_personal_feed instead of action_load_system_feed
1085     _PERSONAL_FEED = False
1086
1087     @property
1088     def _FEED_TEMPLATE(self):
1089         action = 'action_load_system_feed'
1090         if self._PERSONAL_FEED:
1091             action = 'action_load_personal_feed'
1092         return 'http://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action, self._FEED_NAME)
1093
1094     @property
1095     def IE_NAME(self):
1096         return u'youtube:%s' % self._FEED_NAME
1097
1098     def _real_initialize(self):
1099         self._login()
1100
1101     def _real_extract(self, url):
1102         feed_entries = []
1103         # The step argument is available only in 2.7 or higher
1104         for i in itertools.count(0):
1105             paging = i*self._PAGING_STEP
1106             info = self._download_webpage(self._FEED_TEMPLATE % paging,
1107                                           u'%s feed' % self._FEED_NAME,
1108                                           u'Downloading page %s' % i)
1109             info = json.loads(info)
1110             feed_html = info['feed_html']
1111             m_ids = re.finditer(r'"/watch\?v=(.*?)["&]', feed_html)
1112             ids = orderedSet(m.group(1) for m in m_ids)
1113             feed_entries.extend(self.url_result(id, 'Youtube') for id in ids)
1114             if info['paging'] is None:
1115                 break
1116         return self.playlist_result(feed_entries, playlist_title=self._PLAYLIST_TITLE)
1117
1118 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1119     IE_DESC = u'YouTube.com subscriptions feed, "ytsubs" keyword(requires authentication)'
1120     _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1121     _FEED_NAME = 'subscriptions'
1122     _PLAYLIST_TITLE = u'Youtube Subscriptions'
1123
1124 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1125     IE_DESC = u'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1126     _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1127     _FEED_NAME = 'recommended'
1128     _PLAYLIST_TITLE = u'Youtube Recommended videos'
1129
1130 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor):
1131     IE_DESC = u'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1132     _VALID_URL = r'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1133     _FEED_NAME = 'watch_later'
1134     _PLAYLIST_TITLE = u'Youtube Watch Later'
1135     _PAGING_STEP = 100
1136     _PERSONAL_FEED = True
1137
1138 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1139     IE_NAME = u'youtube:favorites'
1140     IE_DESC = u'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1141     _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1142     _LOGIN_REQUIRED = True
1143
1144     def _real_extract(self, url):
1145         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1146         playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, u'favourites playlist id')
1147         return self.url_result(playlist_id, 'YoutubePlaylist')