3 from __future__ import unicode_literals
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlencode,
24 compat_urllib_parse_urlparse,
33 get_element_by_attribute,
53 class YoutubeBaseInfoExtractor(InfoExtractor):
54 """Provide base functions for Youtube extractors"""
55 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
56 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
58 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
59 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
60 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
62 _NETRC_MACHINE = 'youtube'
63 # If True it will raise an error if no login info is provided
64 _LOGIN_REQUIRED = False
66 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
68 def _set_language(self):
70 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
71 # YouTube sets the expire time to about two months
72 expire_time=time.time() + 2 * 30 * 24 * 3600)
74 def _ids_to_results(self, ids):
76 self.url_result(vid_id, 'Youtube', video_id=vid_id)
81 Attempt to log in to YouTube.
82 True is returned if successful or skipped.
83 False is returned if login failed.
85 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
87 (username, password) = self._get_login_info()
88 # No authentication to be performed
90 if self._LOGIN_REQUIRED:
91 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
94 login_page = self._download_webpage(
95 self._LOGIN_URL, None,
96 note='Downloading login page',
97 errnote='unable to fetch login page', fatal=False)
98 if login_page is False:
101 login_form = self._hidden_inputs(login_page)
103 def req(url, f_req, note, errnote):
104 data = login_form.copy()
107 'checkConnection': 'youtube',
108 'checkedDomains': 'youtube',
110 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
111 'f.req': json.dumps(f_req),
112 'flowName': 'GlifWebSignIn',
113 'flowEntry': 'ServiceLogin',
115 return self._download_json(
116 url, None, note=note, errnote=errnote,
117 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
119 data=urlencode_postdata(data), headers={
120 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
121 'Google-Accounts-XSRF': 1,
125 self._downloader.report_warning(message)
129 None, [], None, 'US', None, None, 2, False, True,
133 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
135 1, [None, None, []], None, None, None, True
140 lookup_results = req(
141 self._LOOKUP_URL, lookup_req,
142 'Looking up account info', 'Unable to look up account info')
144 if lookup_results is False:
147 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
149 warn('Unable to extract user hash')
154 None, 1, None, [1, None, None, None, [password, None, True]],
156 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
157 1, [None, None, []], None, None, None, True
160 challenge_results = req(
161 self._CHALLENGE_URL, challenge_req,
162 'Logging in', 'Unable to log in')
164 if challenge_results is False:
167 login_res = try_get(challenge_results, lambda x: x[0][5], list)
169 login_msg = try_get(login_res, lambda x: x[5], compat_str)
171 'Unable to login: %s' % 'Invalid password'
172 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
175 res = try_get(challenge_results, lambda x: x[0][-1], list)
177 warn('Unable to extract result entry')
180 tfa = try_get(res, lambda x: x[0][0], list)
182 tfa_str = try_get(tfa, lambda x: x[2], compat_str)
183 if tfa_str == 'TWO_STEP_VERIFICATION':
184 # SEND_SUCCESS - TFA code has been successfully sent to phone
185 # QUOTA_EXCEEDED - reached the limit of TFA codes
186 status = try_get(tfa, lambda x: x[5], compat_str)
187 if status == 'QUOTA_EXCEEDED':
188 warn('Exceeded the limit of TFA codes, try later')
191 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
193 warn('Unable to extract TL')
196 tfa_code = self._get_tfa_info('2-step verification code')
200 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
201 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
204 tfa_code = remove_start(tfa_code, 'G-')
207 user_hash, None, 2, None,
209 9, None, None, None, None, None, None, None,
210 [None, tfa_code, True, 2]
214 self._TFA_URL.format(tl), tfa_req,
215 'Submitting TFA code', 'Unable to submit TFA code')
217 if tfa_results is False:
220 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
222 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
224 'Unable to finish TFA: %s' % 'Invalid TFA code'
225 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
228 check_cookie_url = try_get(
229 tfa_results, lambda x: x[0][-1][2], compat_str)
231 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
233 if not check_cookie_url:
234 warn('Unable to extract CheckCookie URL')
237 check_cookie_results = self._download_webpage(
238 check_cookie_url, None, 'Checking cookie', fatal=False)
240 if check_cookie_results is False:
243 if 'https://myaccount.google.com/' not in check_cookie_results:
244 warn('Unable to log in')
249 def _download_webpage(self, *args, **kwargs):
250 kwargs.setdefault('query', {})['disable_polymer'] = 'true'
251 return super(YoutubeBaseInfoExtractor, self)._download_webpage(
252 *args, **compat_kwargs(kwargs))
254 def _real_initialize(self):
255 if self._downloader is None:
258 if not self._login():
262 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
263 # Extract entries from page with "Load more" button
264 def _entries(self, page, playlist_id):
265 more_widget_html = content_html = page
266 for page_num in itertools.count(1):
267 for entry in self._process_page(content_html):
270 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
274 more = self._download_json(
275 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
276 'Downloading page #%s' % page_num,
277 transform_source=uppercase_escape)
278 content_html = more['content_html']
279 if not content_html.strip():
280 # Some webpages show a "Load more" button but they don't
283 more_widget_html = more['load_more_widget_html']
286 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
287 def _process_page(self, content):
288 for video_id, video_title in self.extract_videos_from_page(content):
289 yield self.url_result(video_id, 'Youtube', video_id, video_title)
291 def extract_videos_from_page(self, page):
294 for mobj in re.finditer(self._VIDEO_RE, page):
295 # The link with index 0 is not the first video of the playlist (not sure if still actual)
296 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
298 video_id = mobj.group('id')
299 video_title = unescapeHTML(mobj.group('title'))
301 video_title = video_title.strip()
303 idx = ids_in_page.index(video_id)
304 if video_title and not titles_in_page[idx]:
305 titles_in_page[idx] = video_title
307 ids_in_page.append(video_id)
308 titles_in_page.append(video_title)
309 return zip(ids_in_page, titles_in_page)
312 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
313 def _process_page(self, content):
314 for playlist_id in orderedSet(re.findall(
315 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
317 yield self.url_result(
318 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
320 def _real_extract(self, url):
321 playlist_id = self._match_id(url)
322 webpage = self._download_webpage(url, playlist_id)
323 title = self._og_search_title(webpage, fatal=False)
324 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
327 class YoutubeIE(YoutubeBaseInfoExtractor):
328 IE_DESC = 'YouTube.com'
329 _VALID_URL = r"""(?x)^
331 (?:https?://|//) # http(s):// or protocol-independent URL
332 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
333 (?:www\.)?deturl\.com/www\.youtube\.com/|
334 (?:www\.)?pwnyoutube\.com/|
335 (?:www\.)?yourepeat\.com/|
336 tube\.majestyc\.net/|
337 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
338 (?:.*?\#/)? # handle anchor (#/) redirect urls
339 (?: # the various things that can precede the ID:
340 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
341 |(?: # or the v= param in all its forms
342 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
343 (?:\?|\#!?) # the params delimiter ? or # or #!
344 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
349 youtu\.be| # just youtu.be/xxxx
350 vid\.plus| # or vid.plus/xxxx
351 zwearz\.com/watch| # or zwearz.com/watch/xxxx
353 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
355 )? # all until now is optional -> you can pass the naked ID
356 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
359 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
360 WL # WL are handled by the watch later IE
363 (?(1).+)? # if we found the ID, everything can follow
364 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
365 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
367 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
368 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
369 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
370 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
371 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
372 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
373 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
374 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
375 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
376 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
377 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
378 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
379 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
380 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
381 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
382 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
383 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
384 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
388 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
389 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
390 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
391 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
392 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
393 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
394 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
396 # Apple HTTP Live Streaming
397 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
398 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
399 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
400 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
401 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
402 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
403 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
404 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
407 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
408 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
409 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
410 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
411 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
412 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
413 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
414 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
415 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
416 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
417 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
418 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
421 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
422 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
423 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
424 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
425 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
426 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
427 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
430 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
431 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
432 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
433 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
434 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
435 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
436 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
437 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
438 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
439 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
440 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
441 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
442 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
443 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
444 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
445 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
446 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
447 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
448 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
449 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
450 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
451 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
454 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
455 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
457 # Dash webm audio with opus inside
458 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
459 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
460 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
463 '_rtmp': {'protocol': 'rtmp'},
465 _SUBTITLE_FORMATS = ('ttml', 'vtt')
472 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
476 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
477 'uploader': 'Philipp Hagemeister',
478 'uploader_id': 'phihag',
479 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
480 'upload_date': '20121002',
481 'license': 'Standard YouTube License',
482 '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 .',
483 'categories': ['Science & Technology'],
484 'tags': ['youtube-dl'],
487 'dislike_count': int,
493 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
494 'note': 'Test generic use_cipher_signature video (#897)',
498 'upload_date': '20120506',
499 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
500 'alt_title': 'I Love It (feat. Charli XCX)',
501 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
502 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
503 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
504 'iconic ep', 'iconic', 'love', 'it'],
506 'uploader': 'Icona Pop',
507 'uploader_id': 'IconaPop',
508 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
509 'license': 'Standard YouTube License',
510 'creator': 'Icona Pop',
514 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
515 'note': 'Test VEVO video with age protection (#956)',
519 'upload_date': '20130703',
520 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
521 'alt_title': 'Tunnel Vision',
522 'description': 'md5:64249768eec3bc4276236606ea996373',
524 'uploader': 'justintimberlakeVEVO',
525 'uploader_id': 'justintimberlakeVEVO',
526 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
527 'license': 'Standard YouTube License',
528 'creator': 'Justin Timberlake',
533 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
534 'note': 'Embed-only video (#1746)',
538 'upload_date': '20120608',
539 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
540 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
541 'uploader': 'SET India',
542 'uploader_id': 'setindia',
543 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
544 'license': 'Standard YouTube License',
549 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
550 'note': 'Use the first video ID in the URL',
554 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
555 'uploader': 'Philipp Hagemeister',
556 'uploader_id': 'phihag',
557 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
558 'upload_date': '20121002',
559 'license': 'Standard YouTube License',
560 '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 .',
561 'categories': ['Science & Technology'],
562 'tags': ['youtube-dl'],
565 'dislike_count': int,
568 'skip_download': True,
572 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
573 'note': '256k DASH audio (format 141) via DASH manifest',
577 'upload_date': '20121002',
578 'uploader_id': '8KVIDEO',
579 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
581 'uploader': '8KVIDEO',
582 'license': 'Standard YouTube License',
583 'title': 'UHDTV TEST 8K VIDEO.mp4'
586 'youtube_include_dash_manifest': True,
589 'skip': 'format 141 not served anymore',
591 # DASH manifest with encrypted signature
593 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
597 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
598 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
600 'uploader': 'AfrojackVEVO',
601 'uploader_id': 'AfrojackVEVO',
602 'upload_date': '20131011',
603 'license': 'Standard YouTube License',
606 'youtube_include_dash_manifest': True,
607 'format': '141/bestaudio[ext=m4a]',
610 # JS player signature function name containing $
612 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
616 'title': 'Taylor Swift - Shake It Off',
617 'alt_title': 'Shake It Off',
618 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
620 'uploader': 'TaylorSwiftVEVO',
621 'uploader_id': 'TaylorSwiftVEVO',
622 'upload_date': '20140818',
623 'license': 'Standard YouTube License',
624 'creator': 'Taylor Swift',
627 'youtube_include_dash_manifest': True,
628 'format': '141/bestaudio[ext=m4a]',
633 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
638 'upload_date': '20100909',
639 'uploader': 'The Amazing Atheist',
640 'uploader_id': 'TheAmazingAtheist',
641 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
642 'license': 'Standard YouTube License',
643 'title': 'Burning Everyone\'s Koran',
644 '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',
647 # Normal age-gate video (No vevo, embed allowed)
649 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
653 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
654 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
656 'uploader': 'The Witcher',
657 'uploader_id': 'WitcherGame',
658 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
659 'upload_date': '20140605',
660 'license': 'Standard YouTube License',
664 # Age-gate video with encrypted signature
666 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
670 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
671 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
673 'uploader': 'LloydVEVO',
674 'uploader_id': 'LloydVEVO',
675 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
676 'upload_date': '20110629',
677 'license': 'Standard YouTube License',
681 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
682 # YouTube Red ad is not captured for creator
684 'url': '__2ABJjxzNo',
689 'upload_date': '20100430',
690 'uploader_id': 'deadmau5',
691 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
692 'creator': 'deadmau5',
693 'description': 'md5:12c56784b8032162bb936a5f76d55360',
694 'uploader': 'deadmau5',
695 'license': 'Standard YouTube License',
696 'title': 'Deadmau5 - Some Chords (HD)',
697 'alt_title': 'Some Chords',
699 'expected_warnings': [
700 'DASH manifest missing',
703 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
705 'url': 'lqQg6PlCWgI',
710 'upload_date': '20150827',
711 'uploader_id': 'olympic',
712 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
713 'license': 'Standard YouTube License',
714 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
715 'uploader': 'Olympic',
716 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
719 'skip_download': 'requires avconv',
724 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
728 'stretched_ratio': 16 / 9.,
730 'upload_date': '20110310',
731 'uploader_id': 'AllenMeow',
732 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
733 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯',
735 'license': 'Standard YouTube License',
736 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
739 # url_encoded_fmt_stream_map is empty string
741 'url': 'qEJwOuvDf7I',
745 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
747 'upload_date': '20150404',
748 'uploader_id': 'spbelect',
749 'uploader': 'Наблюдатели Петербурга',
752 'skip_download': 'requires avconv',
754 'skip': 'This live event has ended.',
756 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
758 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
762 'title': 'md5:7b81415841e02ecd4313668cde88737a',
763 'description': 'md5:116377fd2963b81ec4ce64b542173306',
765 'upload_date': '20150625',
766 'uploader_id': 'dorappi2000',
767 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
768 'uploader': 'dorappi2000',
769 'license': 'Standard YouTube License',
770 'formats': 'mincount:32',
773 # DASH manifest with segment_list
775 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
776 'md5': '8ce563a1d667b599d21064e982ab9e31',
780 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
781 'uploader': 'Airtek',
782 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
783 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
784 'license': 'Standard YouTube License',
785 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
788 'youtube_include_dash_manifest': True,
789 'format': '135', # bestvideo
791 'skip': 'This live event has ended.',
794 # Multifeed videos (multiple cameras), URL is for Main Camera
795 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
798 'title': 'teamPGP: Rocket League Noob Stream',
799 'description': 'md5:dc7872fb300e143831327f1bae3af010',
805 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
806 'description': 'md5:dc7872fb300e143831327f1bae3af010',
808 'upload_date': '20150721',
809 'uploader': 'Beer Games Beer',
810 'uploader_id': 'beergamesbeer',
811 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
812 'license': 'Standard YouTube License',
818 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
819 'description': 'md5:dc7872fb300e143831327f1bae3af010',
821 'upload_date': '20150721',
822 'uploader': 'Beer Games Beer',
823 'uploader_id': 'beergamesbeer',
824 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
825 'license': 'Standard YouTube License',
831 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
832 'description': 'md5:dc7872fb300e143831327f1bae3af010',
834 'upload_date': '20150721',
835 'uploader': 'Beer Games Beer',
836 'uploader_id': 'beergamesbeer',
837 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
838 'license': 'Standard YouTube License',
844 'title': 'teamPGP: Rocket League Noob Stream (zim)',
845 'description': 'md5:dc7872fb300e143831327f1bae3af010',
847 'upload_date': '20150721',
848 'uploader': 'Beer Games Beer',
849 'uploader_id': 'beergamesbeer',
850 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
851 'license': 'Standard YouTube License',
855 'skip_download': True,
859 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
860 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
863 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
866 'skip': 'Not multifeed anymore',
869 'url': 'https://vid.plus/FlRa-iH7PGw',
870 'only_matching': True,
873 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
874 'only_matching': True,
877 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
878 # Also tests cut-off URL expansion in video description (see
879 # https://github.com/rg3/youtube-dl/issues/1892,
880 # https://github.com/rg3/youtube-dl/issues/8164)
881 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
885 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
886 'alt_title': 'Dark Walk',
887 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
889 'upload_date': '20151119',
890 'uploader_id': 'IronSoulElf',
891 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
892 'uploader': 'IronSoulElf',
893 'license': 'Standard YouTube License',
894 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
897 'skip_download': True,
901 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
902 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
903 'only_matching': True,
906 # Video with yt:stretch=17:0
907 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
911 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
912 'description': 'md5:ee18a25c350637c8faff806845bddee9',
913 'upload_date': '20151107',
914 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
915 'uploader': 'CH GAMER DROID',
918 'skip_download': True,
920 'skip': 'This video does not exist.',
923 # Video licensed under Creative Commons
924 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
928 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
929 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
931 'upload_date': '20150127',
932 'uploader_id': 'BerkmanCenter',
933 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
934 'uploader': 'The Berkman Klein Center for Internet & Society',
935 'license': 'Creative Commons Attribution license (reuse allowed)',
938 'skip_download': True,
942 # Channel-like uploader_url
943 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
947 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
948 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
950 'upload_date': '20151119',
951 'uploader': 'Bernie 2016',
952 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
953 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
954 'license': 'Creative Commons Attribution license (reuse allowed)',
957 'skip_download': True,
961 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
962 'only_matching': True,
965 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
966 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
967 'only_matching': True,
970 # Rental video preview
971 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
975 'title': 'Piku - Trailer',
976 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
977 'upload_date': '20150811',
978 'uploader': 'FlixMatrix',
979 'uploader_id': 'FlixMatrixKaravan',
980 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
981 'license': 'Standard YouTube License',
984 'skip_download': True,
988 # YouTube Red video with episode data
989 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
993 'title': 'Isolation - Mind Field (Ep 1)',
994 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
996 'upload_date': '20170118',
997 'uploader': 'Vsauce',
998 'uploader_id': 'Vsauce',
999 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1000 'license': 'Standard YouTube License',
1001 'series': 'Mind Field',
1003 'episode_number': 1,
1006 'skip_download': True,
1008 'expected_warnings': [
1009 'Skipping DASH manifest',
1013 # The following content has been identified by the YouTube community
1014 # as inappropriate or offensive to some audiences.
1015 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1017 'id': '6SJNVb0GnPI',
1019 'title': 'Race Differences in Intelligence',
1020 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1022 'upload_date': '20140124',
1023 'uploader': 'New Century Foundation',
1024 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1025 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1026 'license': 'Standard YouTube License',
1030 'skip_download': True,
1035 'url': '1t24XAntNCY',
1036 'only_matching': True,
1039 # geo restricted to JP
1040 'url': 'sJL6WA-aGkQ',
1041 'only_matching': True,
1044 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1045 'only_matching': True,
1049 def __init__(self, *args, **kwargs):
1050 super(YoutubeIE, self).__init__(*args, **kwargs)
1051 self._player_cache = {}
1053 def report_video_info_webpage_download(self, video_id):
1054 """Report attempt to download video info webpage."""
1055 self.to_screen('%s: Downloading video info webpage' % video_id)
1057 def report_information_extraction(self, video_id):
1058 """Report attempt to extract video information."""
1059 self.to_screen('%s: Extracting video information' % video_id)
1061 def report_unavailable_format(self, video_id, format):
1062 """Report extracted video URL."""
1063 self.to_screen('%s: Format %s not available' % (video_id, format))
1065 def report_rtmp_download(self):
1066 """Indicate the download will use the RTMP protocol."""
1067 self.to_screen('RTMP download detected')
1069 def _signature_cache_id(self, example_sig):
1070 """ Return a string representation of a signature """
1071 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1073 def _extract_signature_function(self, video_id, player_url, example_sig):
1075 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1078 raise ExtractorError('Cannot identify player %r' % player_url)
1079 player_type = id_m.group('ext')
1080 player_id = id_m.group('id')
1082 # Read from filesystem cache
1083 func_id = '%s_%s_%s' % (
1084 player_type, player_id, self._signature_cache_id(example_sig))
1085 assert os.path.basename(func_id) == func_id
1087 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1088 if cache_spec is not None:
1089 return lambda s: ''.join(s[i] for i in cache_spec)
1092 'Downloading player %s' % player_url
1093 if self._downloader.params.get('verbose') else
1094 'Downloading %s player %s' % (player_type, player_id)
1096 if player_type == 'js':
1097 code = self._download_webpage(
1098 player_url, video_id,
1100 errnote='Download of %s failed' % player_url)
1101 res = self._parse_sig_js(code)
1102 elif player_type == 'swf':
1103 urlh = self._request_webpage(
1104 player_url, video_id,
1106 errnote='Download of %s failed' % player_url)
1108 res = self._parse_sig_swf(code)
1110 assert False, 'Invalid player type %r' % player_type
1112 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1113 cache_res = res(test_string)
1114 cache_spec = [ord(c) for c in cache_res]
1116 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1119 def _print_sig_code(self, func, example_sig):
1120 def gen_sig_code(idxs):
1121 def _genslice(start, end, step):
1122 starts = '' if start == 0 else str(start)
1123 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1124 steps = '' if step == 1 else (':%d' % step)
1125 return 's[%s%s%s]' % (starts, ends, steps)
1128 # Quelch pyflakes warnings - start will be set when step is set
1129 start = '(Never used)'
1130 for i, prev in zip(idxs[1:], idxs[:-1]):
1131 if step is not None:
1132 if i - prev == step:
1134 yield _genslice(start, prev, step)
1137 if i - prev in [-1, 1]:
1142 yield 's[%d]' % prev
1146 yield _genslice(start, i, step)
1148 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1149 cache_res = func(test_string)
1150 cache_spec = [ord(c) for c in cache_res]
1151 expr_code = ' + '.join(gen_sig_code(cache_spec))
1152 signature_id_tuple = '(%s)' % (
1153 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1154 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1155 ' return %s\n') % (signature_id_tuple, expr_code)
1156 self.to_screen('Extracted signature function:\n' + code)
1158 def _parse_sig_js(self, jscode):
1159 funcname = self._search_regex(
1160 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1161 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1162 jscode, 'Initial JS player signature function name', group='sig')
1164 jsi = JSInterpreter(jscode)
1165 initial_function = jsi.extract_function(funcname)
1166 return lambda s: initial_function([s])
1168 def _parse_sig_swf(self, file_contents):
1169 swfi = SWFInterpreter(file_contents)
1170 TARGET_CLASSNAME = 'SignatureDecipher'
1171 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1172 initial_function = swfi.extract_function(searched_class, 'decipher')
1173 return lambda s: initial_function([s])
1175 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1176 """Turn the encrypted s field into a working signature"""
1178 if player_url is None:
1179 raise ExtractorError('Cannot decrypt signature without player_url')
1181 if player_url.startswith('//'):
1182 player_url = 'https:' + player_url
1183 elif not re.match(r'https?://', player_url):
1184 player_url = compat_urlparse.urljoin(
1185 'https://www.youtube.com', player_url)
1187 player_id = (player_url, self._signature_cache_id(s))
1188 if player_id not in self._player_cache:
1189 func = self._extract_signature_function(
1190 video_id, player_url, s
1192 self._player_cache[player_id] = func
1193 func = self._player_cache[player_id]
1194 if self._downloader.params.get('youtube_print_sig_code'):
1195 self._print_sig_code(func, s)
1197 except Exception as e:
1198 tb = traceback.format_exc()
1199 raise ExtractorError(
1200 'Signature extraction failed: ' + tb, cause=e)
1202 def _get_subtitles(self, video_id, webpage):
1204 subs_doc = self._download_xml(
1205 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1206 video_id, note=False)
1207 except ExtractorError as err:
1208 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1212 for track in subs_doc.findall('track'):
1213 lang = track.attrib['lang_code']
1214 if lang in sub_lang_list:
1217 for ext in self._SUBTITLE_FORMATS:
1218 params = compat_urllib_parse_urlencode({
1222 'name': track.attrib['name'].encode('utf-8'),
1224 sub_formats.append({
1225 'url': 'https://www.youtube.com/api/timedtext?' + params,
1228 sub_lang_list[lang] = sub_formats
1229 if not sub_lang_list:
1230 self._downloader.report_warning('video doesn\'t have subtitles')
1232 return sub_lang_list
1234 def _get_ytplayer_config(self, video_id, webpage):
1236 # User data may contain arbitrary character sequences that may affect
1237 # JSON extraction with regex, e.g. when '};' is contained the second
1238 # regex won't capture the whole JSON. Yet working around by trying more
1239 # concrete regex first keeping in mind proper quoted string handling
1240 # to be implemented in future that will replace this workaround (see
1241 # https://github.com/rg3/youtube-dl/issues/7468,
1242 # https://github.com/rg3/youtube-dl/pull/7599)
1243 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1244 r';ytplayer\.config\s*=\s*({.+?});',
1246 config = self._search_regex(
1247 patterns, webpage, 'ytplayer.config', default=None)
1249 return self._parse_json(
1250 uppercase_escape(config), video_id, fatal=False)
1252 def _get_automatic_captions(self, video_id, webpage):
1253 """We need the webpage for getting the captions url, pass it as an
1254 argument to speed up the process."""
1255 self.to_screen('%s: Looking for automatic captions' % video_id)
1256 player_config = self._get_ytplayer_config(video_id, webpage)
1257 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1258 if not player_config:
1259 self._downloader.report_warning(err_msg)
1262 args = player_config['args']
1263 caption_url = args.get('ttsurl')
1265 timestamp = args['timestamp']
1266 # We get the available subtitles
1267 list_params = compat_urllib_parse_urlencode({
1272 list_url = caption_url + '&' + list_params
1273 caption_list = self._download_xml(list_url, video_id)
1274 original_lang_node = caption_list.find('track')
1275 if original_lang_node is None:
1276 self._downloader.report_warning('Video doesn\'t have automatic captions')
1278 original_lang = original_lang_node.attrib['lang_code']
1279 caption_kind = original_lang_node.attrib.get('kind', '')
1282 for lang_node in caption_list.findall('target'):
1283 sub_lang = lang_node.attrib['lang_code']
1285 for ext in self._SUBTITLE_FORMATS:
1286 params = compat_urllib_parse_urlencode({
1287 'lang': original_lang,
1291 'kind': caption_kind,
1293 sub_formats.append({
1294 'url': caption_url + '&' + params,
1297 sub_lang_list[sub_lang] = sub_formats
1298 return sub_lang_list
1300 def make_captions(sub_url, sub_langs):
1301 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1302 caption_qs = compat_parse_qs(parsed_sub_url.query)
1304 for sub_lang in sub_langs:
1306 for ext in self._SUBTITLE_FORMATS:
1308 'tlang': [sub_lang],
1311 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1312 query=compat_urllib_parse_urlencode(caption_qs, True)))
1313 sub_formats.append({
1317 captions[sub_lang] = sub_formats
1320 # New captions format as of 22.06.2017
1321 player_response = args.get('player_response')
1322 if player_response and isinstance(player_response, compat_str):
1323 player_response = self._parse_json(
1324 player_response, video_id, fatal=False)
1326 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1327 base_url = renderer['captionTracks'][0]['baseUrl']
1329 for lang in renderer['translationLanguages']:
1330 lang_code = lang.get('languageCode')
1332 sub_lang_list.append(lang_code)
1333 return make_captions(base_url, sub_lang_list)
1335 # Some videos don't provide ttsurl but rather caption_tracks and
1336 # caption_translation_languages (e.g. 20LmZk1hakA)
1337 # Does not used anymore as of 22.06.2017
1338 caption_tracks = args['caption_tracks']
1339 caption_translation_languages = args['caption_translation_languages']
1340 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1342 for lang in caption_translation_languages.split(','):
1343 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1344 sub_lang = lang_qs.get('lc', [None])[0]
1346 sub_lang_list.append(sub_lang)
1347 return make_captions(caption_url, sub_lang_list)
1348 # An extractor error can be raise by the download process if there are
1349 # no automatic captions but there are subtitles
1350 except (KeyError, IndexError, ExtractorError):
1351 self._downloader.report_warning(err_msg)
1354 def _mark_watched(self, video_id, video_info):
1355 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1356 if not playback_url:
1358 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1359 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1361 # cpn generation algorithm is reverse engineered from base.js.
1362 # In fact it works even with dummy cpn.
1363 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1364 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1370 playback_url = compat_urlparse.urlunparse(
1371 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1373 self._download_webpage(
1374 playback_url, video_id, 'Marking watched',
1375 'Unable to mark watched', fatal=False)
1378 def extract_id(cls, url):
1379 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1381 raise ExtractorError('Invalid URL: %s' % url)
1382 video_id = mobj.group(2)
1385 def _extract_annotations(self, video_id):
1386 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1387 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1390 def _extract_chapters(description, duration):
1393 chapter_lines = re.findall(
1394 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1396 if not chapter_lines:
1399 for next_num, (chapter_line, time_point) in enumerate(
1400 chapter_lines, start=1):
1401 start_time = parse_duration(time_point)
1402 if start_time is None:
1404 if start_time > duration:
1406 end_time = (duration if next_num == len(chapter_lines)
1407 else parse_duration(chapter_lines[next_num][1]))
1408 if end_time is None:
1410 if end_time > duration:
1412 if start_time > end_time:
1414 chapter_title = re.sub(
1415 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1416 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1418 'start_time': start_time,
1419 'end_time': end_time,
1420 'title': chapter_title,
1424 def _real_extract(self, url):
1425 url, smuggled_data = unsmuggle_url(url, {})
1428 'http' if self._downloader.params.get('prefer_insecure', False)
1433 parsed_url = compat_urllib_parse_urlparse(url)
1434 for component in [parsed_url.fragment, parsed_url.query]:
1435 query = compat_parse_qs(component)
1436 if start_time is None and 't' in query:
1437 start_time = parse_duration(query['t'][0])
1438 if start_time is None and 'start' in query:
1439 start_time = parse_duration(query['start'][0])
1440 if end_time is None and 'end' in query:
1441 end_time = parse_duration(query['end'][0])
1443 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1444 mobj = re.search(self._NEXT_URL_RE, url)
1446 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1447 video_id = self.extract_id(url)
1450 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1451 video_webpage = self._download_webpage(url, video_id)
1453 # Attempt to extract SWF player URL
1454 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1455 if mobj is not None:
1456 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1462 def add_dash_mpd(video_info):
1463 dash_mpd = video_info.get('dashmpd')
1464 if dash_mpd and dash_mpd[0] not in dash_mpds:
1465 dash_mpds.append(dash_mpd[0])
1470 def extract_view_count(v_info):
1471 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1474 embed_webpage = None
1475 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1477 # We simulate the access to the video from www.youtube.com/v/{video_id}
1478 # this can be viewed without login into Youtube
1479 url = proto + '://www.youtube.com/embed/%s' % video_id
1480 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1481 data = compat_urllib_parse_urlencode({
1482 'video_id': video_id,
1483 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1484 'sts': self._search_regex(
1485 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1487 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1488 video_info_webpage = self._download_webpage(
1489 video_info_url, video_id,
1490 note='Refetching age-gated info webpage',
1491 errnote='unable to download video info webpage')
1492 video_info = compat_parse_qs(video_info_webpage)
1493 add_dash_mpd(video_info)
1498 # Try looking directly into the video webpage
1499 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1501 args = ytplayer_config['args']
1502 if args.get('url_encoded_fmt_stream_map'):
1503 # Convert to the same format returned by compat_parse_qs
1504 video_info = dict((k, [v]) for k, v in args.items())
1505 add_dash_mpd(video_info)
1506 # Rental video is not rented but preview is available (e.g.
1507 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1508 # https://github.com/rg3/youtube-dl/issues/10532)
1509 if not video_info and args.get('ypc_vid'):
1510 return self.url_result(
1511 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1512 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1514 sts = ytplayer_config.get('sts')
1515 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1516 # We also try looking in get_video_info since it may contain different dashmpd
1517 # URL that points to a DASH manifest with possibly different itag set (some itags
1518 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1519 # manifest pointed by get_video_info's dashmpd).
1520 # The general idea is to take a union of itags of both DASH manifests (for example
1521 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1522 self.report_video_info_webpage_download(video_id)
1523 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1525 'video_id': video_id,
1535 video_info_webpage = self._download_webpage(
1536 '%s://www.youtube.com/get_video_info' % proto,
1537 video_id, note=False,
1538 errnote='unable to download video info webpage',
1539 fatal=False, query=query)
1540 if not video_info_webpage:
1542 get_video_info = compat_parse_qs(video_info_webpage)
1543 add_dash_mpd(get_video_info)
1544 if view_count is None:
1545 view_count = extract_view_count(get_video_info)
1547 video_info = get_video_info
1548 if 'token' in get_video_info:
1549 # Different get_video_info requests may report different results, e.g.
1550 # some may report video unavailability, but some may serve it without
1551 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1552 # the original webpage as well as el=info and el=embedded get_video_info
1553 # requests report video unavailability due to geo restriction while
1554 # el=detailpage succeeds and returns valid data). This is probably
1555 # due to YouTube measures against IP ranges of hosting providers.
1556 # Working around by preferring the first succeeded video_info containing
1557 # the token if no such video_info yet was found.
1558 if 'token' not in video_info:
1559 video_info = get_video_info
1561 if 'token' not in video_info:
1562 if 'reason' in video_info:
1563 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1564 regions_allowed = self._html_search_meta(
1565 'regionsAllowed', video_webpage, default=None)
1566 countries = regions_allowed.split(',') if regions_allowed else None
1567 self.raise_geo_restricted(
1568 msg=video_info['reason'][0], countries=countries)
1569 raise ExtractorError(
1570 'YouTube said: %s' % video_info['reason'][0],
1571 expected=True, video_id=video_id)
1573 raise ExtractorError(
1574 '"token" parameter not in video info for unknown reason',
1578 if 'title' in video_info:
1579 video_title = video_info['title'][0]
1581 self._downloader.report_warning('Unable to extract video title')
1585 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1586 if video_description:
1587 description_original = video_description = re.sub(r'''(?x)
1589 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1590 (?:title|href)="([^"]+)"\s+
1591 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1595 ''', r'\1', video_description)
1596 video_description = clean_html(video_description)
1598 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1600 video_description = unescapeHTML(fd_mobj.group(1))
1602 video_description = ''
1604 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1605 if not self._downloader.params.get('noplaylist'):
1608 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1609 for feed in multifeed_metadata_list.split(','):
1610 # Unquote should take place before split on comma (,) since textual
1611 # fields may contain comma as well (see
1612 # https://github.com/rg3/youtube-dl/issues/8536)
1613 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1615 '_type': 'url_transparent',
1616 'ie_key': 'Youtube',
1618 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1619 {'force_singlefeed': True}),
1620 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1622 feed_ids.append(feed_data['id'][0])
1624 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1625 % (', '.join(feed_ids), video_id))
1626 return self.playlist_result(entries, video_id, video_title, video_description)
1627 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1629 if view_count is None:
1630 view_count = extract_view_count(video_info)
1632 # Check for "rental" videos
1633 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1634 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1636 # Start extracting information
1637 self.report_information_extraction(video_id)
1640 if 'author' not in video_info:
1641 raise ExtractorError('Unable to extract uploader name')
1642 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1645 video_uploader_id = None
1646 video_uploader_url = None
1648 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1650 if mobj is not None:
1651 video_uploader_id = mobj.group('uploader_id')
1652 video_uploader_url = mobj.group('uploader_url')
1654 self._downloader.report_warning('unable to extract uploader nickname')
1657 # We try first to get a high quality image:
1658 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1659 video_webpage, re.DOTALL)
1660 if m_thumb is not None:
1661 video_thumbnail = m_thumb.group(1)
1662 elif 'thumbnail_url' not in video_info:
1663 self._downloader.report_warning('unable to extract video thumbnail')
1664 video_thumbnail = None
1665 else: # don't panic if we can't find it
1666 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1669 upload_date = self._html_search_meta(
1670 'datePublished', video_webpage, 'upload date', default=None)
1672 upload_date = self._search_regex(
1673 [r'(?s)id="eow-date.*?>(.*?)</span>',
1674 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1675 video_webpage, 'upload date', default=None)
1676 upload_date = unified_strdate(upload_date)
1678 video_license = self._html_search_regex(
1679 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1680 video_webpage, 'license', default=None)
1682 m_music = re.search(
1684 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1692 \bhref=["\']/red[^>]*>| # drop possible
1693 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1700 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1701 video_creator = clean_html(m_music.group('creator'))
1703 video_alt_title = video_creator = None
1705 m_episode = re.search(
1706 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1709 series = m_episode.group('series')
1710 season_number = int(m_episode.group('season'))
1711 episode_number = int(m_episode.group('episode'))
1713 series = season_number = episode_number = None
1715 m_cat_container = self._search_regex(
1716 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1717 video_webpage, 'categories', default=None)
1719 category = self._html_search_regex(
1720 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1722 video_categories = None if category is None else [category]
1724 video_categories = None
1727 unescapeHTML(m.group('content'))
1728 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1730 def _extract_count(count_name):
1731 return str_to_int(self._search_regex(
1732 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1733 % re.escape(count_name),
1734 video_webpage, count_name, default=None))
1736 like_count = _extract_count('like')
1737 dislike_count = _extract_count('dislike')
1740 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1741 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1743 video_duration = try_get(
1744 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1745 if not video_duration:
1746 video_duration = parse_duration(self._html_search_meta(
1747 'duration', video_webpage, 'video duration'))
1750 video_annotations = None
1751 if self._downloader.params.get('writeannotations', False):
1752 video_annotations = self._extract_annotations(video_id)
1754 chapters = self._extract_chapters(description_original, video_duration)
1756 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1757 self.report_rtmp_download()
1759 'format_id': '_rtmp',
1761 'url': video_info['conn'][0],
1762 'player_url': player_url,
1764 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1765 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1766 if 'rtmpe%3Dyes' in encoded_url_map:
1767 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1769 fmt_list = video_info.get('fmt_list', [''])[0]
1771 for fmt in fmt_list.split(','):
1772 spec = fmt.split('/')
1774 width_height = spec[1].split('x')
1775 if len(width_height) == 2:
1776 formats_spec[spec[0]] = {
1777 'resolution': spec[1],
1778 'width': int_or_none(width_height[0]),
1779 'height': int_or_none(width_height[1]),
1782 for url_data_str in encoded_url_map.split(','):
1783 url_data = compat_parse_qs(url_data_str)
1784 if 'itag' not in url_data or 'url' not in url_data:
1786 format_id = url_data['itag'][0]
1787 url = url_data['url'][0]
1789 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1790 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1791 jsplayer_url_json = self._search_regex(
1793 embed_webpage if age_gate else video_webpage,
1794 'JS player URL (1)', default=None)
1795 if not jsplayer_url_json and not age_gate:
1796 # We need the embed website after all
1797 if embed_webpage is None:
1798 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1799 embed_webpage = self._download_webpage(
1800 embed_url, video_id, 'Downloading embed webpage')
1801 jsplayer_url_json = self._search_regex(
1802 ASSETS_RE, embed_webpage, 'JS player URL')
1804 player_url = json.loads(jsplayer_url_json)
1805 if player_url is None:
1806 player_url_json = self._search_regex(
1807 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1808 video_webpage, 'age gate player URL')
1809 player_url = json.loads(player_url_json)
1811 if 'sig' in url_data:
1812 url += '&signature=' + url_data['sig'][0]
1813 elif 's' in url_data:
1814 encrypted_sig = url_data['s'][0]
1816 if self._downloader.params.get('verbose'):
1817 if player_url is None:
1818 player_version = 'unknown'
1819 player_desc = 'unknown'
1821 if player_url.endswith('swf'):
1822 player_version = self._search_regex(
1823 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1824 'flash player', fatal=False)
1825 player_desc = 'flash player %s' % player_version
1827 player_version = self._search_regex(
1828 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1829 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1831 'html5 player', fatal=False)
1832 player_desc = 'html5 player %s' % player_version
1834 parts_sizes = self._signature_cache_id(encrypted_sig)
1835 self.to_screen('{%s} signature length %s, %s' %
1836 (format_id, parts_sizes, player_desc))
1838 signature = self._decrypt_signature(
1839 encrypted_sig, video_id, player_url, age_gate)
1840 url += '&signature=' + signature
1841 if 'ratebypass' not in url:
1842 url += '&ratebypass=yes'
1845 'format_id': format_id,
1847 'player_url': player_url,
1849 if format_id in self._formats:
1850 dct.update(self._formats[format_id])
1851 if format_id in formats_spec:
1852 dct.update(formats_spec[format_id])
1854 # Some itags are not included in DASH manifest thus corresponding formats will
1855 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1856 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1857 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1858 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1861 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1862 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1865 'fps': int_or_none(url_data.get('fps', [None])[0]),
1866 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1868 for key, value in more_fields.items():
1871 type_ = url_data.get('type', [None])[0]
1873 type_split = type_.split(';')
1874 kind_ext = type_split[0].split('/')
1875 if len(kind_ext) == 2:
1877 dct['ext'] = mimetype2ext(type_split[0])
1878 if kind in ('audio', 'video'):
1880 for mobj in re.finditer(
1881 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1882 if mobj.group('key') == 'codecs':
1883 codecs = mobj.group('val')
1886 dct.update(parse_codecs(codecs))
1888 elif video_info.get('hlsvp'):
1889 manifest_url = video_info['hlsvp'][0]
1891 m3u8_formats = self._extract_m3u8_formats(
1892 manifest_url, video_id, 'mp4', fatal=False)
1893 for a_format in m3u8_formats:
1894 itag = self._search_regex(
1895 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1897 a_format['format_id'] = itag
1898 if itag in self._formats:
1899 dct = self._formats[itag].copy()
1900 dct.update(a_format)
1902 a_format['player_url'] = player_url
1903 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1904 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1905 formats.append(a_format)
1907 unavailable_message = self._html_search_regex(
1908 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1909 video_webpage, 'unavailable message', default=None)
1910 if unavailable_message:
1911 raise ExtractorError(unavailable_message, expected=True)
1912 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1914 # Look for the DASH manifest
1915 if self._downloader.params.get('youtube_include_dash_manifest', True):
1916 dash_mpd_fatal = True
1917 for mpd_url in dash_mpds:
1920 def decrypt_sig(mobj):
1922 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1923 return '/signature/%s' % dec_s
1925 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1927 for df in self._extract_mpd_formats(
1928 mpd_url, video_id, fatal=dash_mpd_fatal,
1929 formats_dict=self._formats):
1930 # Do not overwrite DASH format found in some previous DASH manifest
1931 if df['format_id'] not in dash_formats:
1932 dash_formats[df['format_id']] = df
1933 # Additional DASH manifests may end up in HTTP Error 403 therefore
1934 # allow them to fail without bug report message if we already have
1935 # some DASH manifest succeeded. This is temporary workaround to reduce
1936 # burst of bug reports until we figure out the reason and whether it
1937 # can be fixed at all.
1938 dash_mpd_fatal = False
1939 except (ExtractorError, KeyError) as e:
1940 self.report_warning(
1941 'Skipping DASH manifest: %r' % e, video_id)
1943 # Remove the formats we found through non-DASH, they
1944 # contain less info and it can be wrong, because we use
1945 # fixed values (for example the resolution). See
1946 # https://github.com/rg3/youtube-dl/issues/5774 for an
1948 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1949 formats.extend(dash_formats.values())
1951 # Check for malformed aspect ratio
1952 stretched_m = re.search(
1953 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1956 w = float(stretched_m.group('w'))
1957 h = float(stretched_m.group('h'))
1958 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1959 # We will only process correct ratios.
1963 if f.get('vcodec') != 'none':
1964 f['stretched_ratio'] = ratio
1966 self._sort_formats(formats)
1968 self.mark_watched(video_id, video_info)
1972 'uploader': video_uploader,
1973 'uploader_id': video_uploader_id,
1974 'uploader_url': video_uploader_url,
1975 'upload_date': upload_date,
1976 'license': video_license,
1977 'creator': video_creator,
1978 'title': video_title,
1979 'alt_title': video_alt_title,
1980 'thumbnail': video_thumbnail,
1981 'description': video_description,
1982 'categories': video_categories,
1984 'subtitles': video_subtitles,
1985 'automatic_captions': automatic_captions,
1986 'duration': video_duration,
1987 'age_limit': 18 if age_gate else 0,
1988 'annotations': video_annotations,
1989 'chapters': chapters,
1990 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1991 'view_count': view_count,
1992 'like_count': like_count,
1993 'dislike_count': dislike_count,
1994 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1997 'start_time': start_time,
1998 'end_time': end_time,
2000 'season_number': season_number,
2001 'episode_number': episode_number,
2005 class YoutubeSharedVideoIE(InfoExtractor):
2006 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
2007 IE_NAME = 'youtube:shared'
2010 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
2012 'id': 'uPDB5I9wfp8',
2014 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
2015 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
2016 'upload_date': '20160219',
2017 'uploader': 'Pocoyo - Português (BR)',
2018 'uploader_id': 'PocoyoBrazil',
2020 'add_ie': ['Youtube'],
2022 # There are already too many Youtube downloads
2023 'skip_download': True,
2027 def _real_extract(self, url):
2028 video_id = self._match_id(url)
2030 webpage = self._download_webpage(url, video_id)
2032 real_video_id = self._html_search_meta(
2033 'videoId', webpage, 'YouTube video id', fatal=True)
2035 return self.url_result(real_video_id, YoutubeIE.ie_key())
2038 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2039 IE_DESC = 'YouTube.com playlists'
2040 _VALID_URL = r"""(?x)(?:
2046 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2047 \? (?:.*?[&;])*? (?:p|a|list)=
2050 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2053 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2054 # Top tracks, they can also include dots
2060 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2061 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2062 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2063 IE_NAME = 'youtube:playlist'
2065 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2067 'title': 'ytdl test PL',
2068 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2070 'playlist_count': 3,
2072 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2074 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2075 'title': 'YDL_Empty_List',
2077 'playlist_count': 0,
2078 'skip': 'This playlist is private',
2080 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2081 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2083 'title': '29C3: Not my department',
2084 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2086 'playlist_count': 95,
2088 'note': 'issue #673',
2089 'url': 'PLBB231211A4F62143',
2091 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2092 'id': 'PLBB231211A4F62143',
2094 'playlist_mincount': 26,
2096 'note': 'Large playlist',
2097 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2099 'title': 'Uploads from Cauchemar',
2100 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2102 'playlist_mincount': 799,
2104 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2106 'title': 'YDL_safe_search',
2107 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2109 'playlist_count': 2,
2110 'skip': 'This playlist is private',
2113 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2114 'playlist_count': 4,
2117 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2120 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2121 'playlist_mincount': 485,
2123 'title': '2017 華語最新單曲 (2/24更新)',
2124 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2127 'note': 'Embedded SWF player',
2128 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2129 'playlist_count': 4,
2132 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2135 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2136 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2138 'title': 'Uploads from Interstellar Movie',
2139 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2141 'playlist_mincount': 21,
2143 # Playlist URL that does not actually serve a playlist
2144 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2146 'id': 'FqZTN594JQw',
2148 'title': "Smiley's People 01 detective, Adventure Series, Action",
2149 'uploader': 'STREEM',
2150 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2151 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2152 'upload_date': '20150526',
2153 'license': 'Standard YouTube License',
2154 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2155 'categories': ['People & Blogs'],
2158 'dislike_count': int,
2161 'skip_download': True,
2163 'add_ie': [YoutubeIE.ie_key()],
2165 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2167 'id': 'yeWKywCrFtk',
2169 'title': 'Small Scale Baler and Braiding Rugs',
2170 'uploader': 'Backus-Page House Museum',
2171 'uploader_id': 'backuspagemuseum',
2172 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2173 'upload_date': '20161008',
2174 'license': 'Standard YouTube License',
2175 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2176 'categories': ['Nonprofits & Activism'],
2179 'dislike_count': int,
2183 'skip_download': True,
2186 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2187 'only_matching': True,
2189 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2190 'only_matching': True,
2193 def _real_initialize(self):
2196 def _extract_mix(self, playlist_id):
2197 # The mixes are generated from a single video
2198 # the id of the playlist is just 'RD' + video_id
2200 last_id = playlist_id[-11:]
2201 for n in itertools.count(1):
2202 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2203 webpage = self._download_webpage(
2204 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2205 new_ids = orderedSet(re.findall(
2206 r'''(?xs)data-video-username=".*?".*?
2207 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2209 # Fetch new pages until all the videos are repeated, it seems that
2210 # there are always 51 unique videos.
2211 new_ids = [_id for _id in new_ids if _id not in ids]
2217 url_results = self._ids_to_results(ids)
2219 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2221 search_title('playlist-title') or
2222 search_title('title long-title') or
2223 search_title('title'))
2224 title = clean_html(title_span)
2226 return self.playlist_result(url_results, playlist_id, title)
2228 def _extract_playlist(self, playlist_id):
2229 url = self._TEMPLATE_URL % playlist_id
2230 page = self._download_webpage(url, playlist_id)
2232 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2233 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2234 match = match.strip()
2235 # Check if the playlist exists or is private
2236 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2238 reason = mobj.group('reason')
2239 message = 'This playlist %s' % reason
2240 if 'private' in reason:
2241 message += ', use --username or --netrc to access it'
2243 raise ExtractorError(message, expected=True)
2244 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2245 raise ExtractorError(
2246 'Invalid parameters. Maybe URL is incorrect.',
2248 elif re.match(r'[^<]*Choose your language[^<]*', match):
2251 self.report_warning('Youtube gives an alert message: ' + match)
2253 playlist_title = self._html_search_regex(
2254 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2255 page, 'title', default=None)
2259 if not playlist_title:
2261 # Some playlist URLs don't actually serve a playlist (e.g.
2262 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2263 next(self._entries(page, playlist_id))
2264 except StopIteration:
2267 return has_videos, self.playlist_result(
2268 self._entries(page, playlist_id), playlist_id, playlist_title)
2270 def _check_download_just_video(self, url, playlist_id):
2271 # Check if it's a video-specific URL
2272 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2273 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2274 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2275 'video id', default=None)
2277 if self._downloader.params.get('noplaylist'):
2278 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2279 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2281 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2282 return video_id, None
2285 def _real_extract(self, url):
2286 # Extract playlist id
2287 mobj = re.match(self._VALID_URL, url)
2289 raise ExtractorError('Invalid URL: %s' % url)
2290 playlist_id = mobj.group(1) or mobj.group(2)
2292 video_id, video = self._check_download_just_video(url, playlist_id)
2296 if playlist_id.startswith(('RD', 'UL', 'PU')):
2297 # Mixes require a custom extraction process
2298 return self._extract_mix(playlist_id)
2300 has_videos, playlist = self._extract_playlist(playlist_id)
2301 if has_videos or not video_id:
2304 # Some playlist URLs don't actually serve a playlist (see
2305 # https://github.com/rg3/youtube-dl/issues/10537).
2306 # Fallback to plain video extraction if there is a video id
2307 # along with playlist id.
2308 return self.url_result(video_id, 'Youtube', video_id=video_id)
2311 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2312 IE_DESC = 'YouTube.com channels'
2313 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2314 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2315 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2316 IE_NAME = 'youtube:channel'
2318 'note': 'paginated channel',
2319 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2320 'playlist_mincount': 91,
2322 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2323 'title': 'Uploads from lex will',
2326 'note': 'Age restricted channel',
2327 # from https://www.youtube.com/user/DeusExOfficial
2328 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2329 'playlist_mincount': 64,
2331 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2332 'title': 'Uploads from Deus Ex',
2337 def suitable(cls, url):
2338 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2339 else super(YoutubeChannelIE, cls).suitable(url))
2341 def _build_template_url(self, url, channel_id):
2342 return self._TEMPLATE_URL % channel_id
2344 def _real_extract(self, url):
2345 channel_id = self._match_id(url)
2347 url = self._build_template_url(url, channel_id)
2349 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2350 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2351 # otherwise fallback on channel by page extraction
2352 channel_page = self._download_webpage(
2353 url + '?view=57', channel_id,
2354 'Downloading channel page', fatal=False)
2355 if channel_page is False:
2356 channel_playlist_id = False
2358 channel_playlist_id = self._html_search_meta(
2359 'channelId', channel_page, 'channel id', default=None)
2360 if not channel_playlist_id:
2361 channel_url = self._html_search_meta(
2362 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2363 channel_page, 'channel url', default=None)
2365 channel_playlist_id = self._search_regex(
2366 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2367 channel_url, 'channel id', default=None)
2368 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2369 playlist_id = 'UU' + channel_playlist_id[2:]
2370 return self.url_result(
2371 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2373 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2374 autogenerated = re.search(r'''(?x)
2376 channel-header-autogenerated-label|
2377 yt-channel-title-autogenerated
2378 )[^"]*"''', channel_page) is not None
2381 # The videos are contained in a single page
2382 # the ajax pages can't be used, they are empty
2385 video_id, 'Youtube', video_id=video_id,
2386 video_title=video_title)
2387 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2388 return self.playlist_result(entries, channel_id)
2391 next(self._entries(channel_page, channel_id))
2392 except StopIteration:
2393 alert_message = self._html_search_regex(
2394 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2395 channel_page, 'alert', default=None, group='alert')
2397 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2399 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2402 class YoutubeUserIE(YoutubeChannelIE):
2403 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2404 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2405 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2406 IE_NAME = 'youtube:user'
2409 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2410 'playlist_mincount': 320,
2412 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2413 'title': 'Uploads from The Linux Foundation',
2416 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2417 # but not https://www.youtube.com/user/12minuteathlete/videos
2418 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2419 'playlist_mincount': 249,
2421 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2422 'title': 'Uploads from 12 Minute Athlete',
2425 'url': 'ytuser:phihag',
2426 'only_matching': True,
2428 'url': 'https://www.youtube.com/c/gametrailers',
2429 'only_matching': True,
2431 'url': 'https://www.youtube.com/gametrailers',
2432 'only_matching': True,
2434 # This channel is not available, geo restricted to JP
2435 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2436 'only_matching': True,
2440 def suitable(cls, url):
2441 # Don't return True if the url can be extracted with other youtube
2442 # extractor, the regex would is too permissive and it would match.
2443 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2444 if any(ie.suitable(url) for ie in other_yt_ies):
2447 return super(YoutubeUserIE, cls).suitable(url)
2449 def _build_template_url(self, url, channel_id):
2450 mobj = re.match(self._VALID_URL, url)
2451 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2454 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2455 IE_DESC = 'YouTube.com live streams'
2456 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2457 IE_NAME = 'youtube:live'
2460 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2462 'id': 'a48o2S1cPoo',
2464 'title': 'The Young Turks - Live Main Show',
2465 'uploader': 'The Young Turks',
2466 'uploader_id': 'TheYoungTurks',
2467 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2468 'upload_date': '20150715',
2469 'license': 'Standard YouTube License',
2470 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2471 'categories': ['News & Politics'],
2472 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2474 'dislike_count': int,
2477 'skip_download': True,
2480 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2481 'only_matching': True,
2483 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2484 'only_matching': True,
2486 'url': 'https://www.youtube.com/TheYoungTurks/live',
2487 'only_matching': True,
2490 def _real_extract(self, url):
2491 mobj = re.match(self._VALID_URL, url)
2492 channel_id = mobj.group('id')
2493 base_url = mobj.group('base_url')
2494 webpage = self._download_webpage(url, channel_id, fatal=False)
2496 page_type = self._og_search_property(
2497 'type', webpage, 'page type', default=None)
2498 video_id = self._html_search_meta(
2499 'videoId', webpage, 'video id', default=None)
2500 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2501 return self.url_result(video_id, YoutubeIE.ie_key())
2502 return self.url_result(base_url)
2505 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2506 IE_DESC = 'YouTube.com user/channel playlists'
2507 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2508 IE_NAME = 'youtube:playlists'
2511 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2512 'playlist_mincount': 4,
2514 'id': 'ThirstForScience',
2515 'title': 'Thirst for Science',
2518 # with "Load more" button
2519 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2520 'playlist_mincount': 70,
2523 'title': 'Игорь Клейнер',
2526 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2527 'playlist_mincount': 17,
2529 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2530 'title': 'Chem Player',
2535 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2536 IE_DESC = 'YouTube.com searches'
2537 # there doesn't appear to be a real limit, for example if you search for
2538 # 'python' you get more than 8.000.000 results
2539 _MAX_RESULTS = float('inf')
2540 IE_NAME = 'youtube:search'
2541 _SEARCH_KEY = 'ytsearch'
2542 _EXTRA_QUERY_ARGS = {}
2545 def _get_n_results(self, query, n):
2546 """Get a specified number of results for a query"""
2552 'search_query': query.encode('utf-8'),
2554 url_query.update(self._EXTRA_QUERY_ARGS)
2555 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2557 for pagenum in itertools.count(1):
2558 data = self._download_json(
2559 result_url, video_id='query "%s"' % query,
2560 note='Downloading page %s' % pagenum,
2561 errnote='Unable to download API page',
2562 query={'spf': 'navigate'})
2563 html_content = data[1]['body']['content']
2565 if 'class="search-message' in html_content:
2566 raise ExtractorError(
2567 '[youtube] No video results', expected=True)
2569 new_videos = self._ids_to_results(orderedSet(re.findall(
2570 r'href="/watch\?v=(.{11})', html_content)))
2571 videos += new_videos
2572 if not new_videos or len(videos) > limit:
2574 next_link = self._html_search_regex(
2575 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2576 html_content, 'next link', default=None)
2577 if next_link is None:
2579 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2583 return self.playlist_result(videos, query)
2586 class YoutubeSearchDateIE(YoutubeSearchIE):
2587 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2588 _SEARCH_KEY = 'ytsearchdate'
2589 IE_DESC = 'YouTube.com searches, newest videos first'
2590 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2593 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2594 IE_DESC = 'YouTube.com search URLs'
2595 IE_NAME = 'youtube:search_url'
2596 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2597 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2599 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2600 'playlist_mincount': 5,
2602 'title': 'youtube-dl test video',
2605 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2606 'only_matching': True,
2609 def _real_extract(self, url):
2610 mobj = re.match(self._VALID_URL, url)
2611 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2612 webpage = self._download_webpage(url, query)
2613 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2616 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2617 IE_DESC = 'YouTube.com (multi-season) shows'
2618 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2619 IE_NAME = 'youtube:show'
2621 'url': 'https://www.youtube.com/show/airdisasters',
2622 'playlist_mincount': 5,
2624 'id': 'airdisasters',
2625 'title': 'Air Disasters',
2629 def _real_extract(self, url):
2630 playlist_id = self._match_id(url)
2631 return super(YoutubeShowIE, self)._real_extract(
2632 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2635 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2637 Base class for feed extractors
2638 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2640 _LOGIN_REQUIRED = True
2644 return 'youtube:%s' % self._FEED_NAME
2646 def _real_initialize(self):
2649 def _real_extract(self, url):
2650 page = self._download_webpage(
2651 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2653 # The extraction process is the same as for playlists, but the regex
2654 # for the video ids doesn't contain an index
2656 more_widget_html = content_html = page
2657 for page_num in itertools.count(1):
2658 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2660 # 'recommended' feed has infinite 'load more' and each new portion spins
2661 # the same videos in (sometimes) slightly different order, so we'll check
2662 # for unicity and break when portion has no new videos
2663 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2669 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2673 more = self._download_json(
2674 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2675 'Downloading page #%s' % page_num,
2676 transform_source=uppercase_escape)
2677 content_html = more['content_html']
2678 more_widget_html = more['load_more_widget_html']
2680 return self.playlist_result(
2681 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2684 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2685 IE_NAME = 'youtube:watchlater'
2686 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2687 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2690 'url': 'https://www.youtube.com/playlist?list=WL',
2691 'only_matching': True,
2693 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2694 'only_matching': True,
2697 def _real_extract(self, url):
2698 _, video = self._check_download_just_video(url, 'WL')
2701 _, playlist = self._extract_playlist('WL')
2705 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2706 IE_NAME = 'youtube:favorites'
2707 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2708 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2709 _LOGIN_REQUIRED = True
2711 def _real_extract(self, url):
2712 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2713 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2714 return self.url_result(playlist_id, 'YoutubePlaylist')
2717 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2718 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2719 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2720 _FEED_NAME = 'recommended'
2721 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2724 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2725 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2726 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2727 _FEED_NAME = 'subscriptions'
2728 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2731 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2732 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2733 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2734 _FEED_NAME = 'history'
2735 _PLAYLIST_TITLE = 'Youtube History'
2738 class YoutubeTruncatedURLIE(InfoExtractor):
2739 IE_NAME = 'youtube:truncated_url'
2740 IE_DESC = False # Do not list
2741 _VALID_URL = r'''(?x)
2743 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2746 annotation_id=annotation_[^&]+|
2752 attribution_link\?a=[^&]+
2758 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2759 'only_matching': True,
2761 'url': 'https://www.youtube.com/watch?',
2762 'only_matching': True,
2764 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2765 'only_matching': True,
2767 'url': 'https://www.youtube.com/watch?feature=foo',
2768 'only_matching': True,
2770 'url': 'https://www.youtube.com/watch?hl=en-GB',
2771 'only_matching': True,
2773 'url': 'https://www.youtube.com/watch?t=2372',
2774 'only_matching': True,
2777 def _real_extract(self, url):
2778 raise ExtractorError(
2779 'Did you forget to quote the URL? Remember that & is a meta '
2780 'character in most shells, so you want to put the URL in quotes, '
2782 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2783 ' or simply youtube-dl BaW_jenozKc .',
2787 class YoutubeTruncatedIDIE(InfoExtractor):
2788 IE_NAME = 'youtube:truncated_id'
2789 IE_DESC = False # Do not list
2790 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2793 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2794 'only_matching': True,
2797 def _real_extract(self, url):
2798 video_id = self._match_id(url)
2799 raise ExtractorError(
2800 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),