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,
54 class YoutubeBaseInfoExtractor(InfoExtractor):
55 """Provide base functions for Youtube extractors"""
56 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
57 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
59 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
60 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
61 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
63 _NETRC_MACHINE = 'youtube'
64 # If True it will raise an error if no login info is provided
65 _LOGIN_REQUIRED = False
67 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
69 def _set_language(self):
71 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
72 # YouTube sets the expire time to about two months
73 expire_time=time.time() + 2 * 30 * 24 * 3600)
75 def _ids_to_results(self, ids):
77 self.url_result(vid_id, 'Youtube', video_id=vid_id)
82 Attempt to log in to YouTube.
83 True is returned if successful or skipped.
84 False is returned if login failed.
86 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
88 username, password = self._get_login_info()
89 # No authentication to be performed
91 if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
92 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
95 login_page = self._download_webpage(
96 self._LOGIN_URL, None,
97 note='Downloading login page',
98 errnote='unable to fetch login page', fatal=False)
99 if login_page is False:
102 login_form = self._hidden_inputs(login_page)
104 def req(url, f_req, note, errnote):
105 data = login_form.copy()
108 'checkConnection': 'youtube',
109 'checkedDomains': 'youtube',
111 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
112 'f.req': json.dumps(f_req),
113 'flowName': 'GlifWebSignIn',
114 'flowEntry': 'ServiceLogin',
116 return self._download_json(
117 url, None, note=note, errnote=errnote,
118 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
120 data=urlencode_postdata(data), headers={
121 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
122 'Google-Accounts-XSRF': 1,
126 self._downloader.report_warning(message)
130 None, [], None, 'US', None, None, 2, False, True,
134 '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',
136 1, [None, None, []], None, None, None, True
141 lookup_results = req(
142 self._LOOKUP_URL, lookup_req,
143 'Looking up account info', 'Unable to look up account info')
145 if lookup_results is False:
148 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
150 warn('Unable to extract user hash')
155 None, 1, None, [1, None, None, None, [password, None, True]],
157 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],
158 1, [None, None, []], None, None, None, True
161 challenge_results = req(
162 self._CHALLENGE_URL, challenge_req,
163 'Logging in', 'Unable to log in')
165 if challenge_results is False:
168 login_res = try_get(challenge_results, lambda x: x[0][5], list)
170 login_msg = try_get(login_res, lambda x: x[5], compat_str)
172 'Unable to login: %s' % 'Invalid password'
173 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
176 res = try_get(challenge_results, lambda x: x[0][-1], list)
178 warn('Unable to extract result entry')
181 tfa = try_get(res, lambda x: x[0][0], list)
183 tfa_str = try_get(tfa, lambda x: x[2], compat_str)
184 if tfa_str == 'TWO_STEP_VERIFICATION':
185 # SEND_SUCCESS - TFA code has been successfully sent to phone
186 # QUOTA_EXCEEDED - reached the limit of TFA codes
187 status = try_get(tfa, lambda x: x[5], compat_str)
188 if status == 'QUOTA_EXCEEDED':
189 warn('Exceeded the limit of TFA codes, try later')
192 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
194 warn('Unable to extract TL')
197 tfa_code = self._get_tfa_info('2-step verification code')
201 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
202 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
205 tfa_code = remove_start(tfa_code, 'G-')
208 user_hash, None, 2, None,
210 9, None, None, None, None, None, None, None,
211 [None, tfa_code, True, 2]
215 self._TFA_URL.format(tl), tfa_req,
216 'Submitting TFA code', 'Unable to submit TFA code')
218 if tfa_results is False:
221 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
223 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
225 'Unable to finish TFA: %s' % 'Invalid TFA code'
226 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
229 check_cookie_url = try_get(
230 tfa_results, lambda x: x[0][-1][2], compat_str)
232 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
234 if not check_cookie_url:
235 warn('Unable to extract CheckCookie URL')
238 check_cookie_results = self._download_webpage(
239 check_cookie_url, None, 'Checking cookie', fatal=False)
241 if check_cookie_results is False:
244 if 'https://myaccount.google.com/' not in check_cookie_results:
245 warn('Unable to log in')
250 def _download_webpage_handle(self, *args, **kwargs):
251 kwargs.setdefault('query', {})['disable_polymer'] = 'true'
252 return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
253 *args, **compat_kwargs(kwargs))
255 def _real_initialize(self):
256 if self._downloader is None:
259 if not self._login():
263 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
264 # Extract entries from page with "Load more" button
265 def _entries(self, page, playlist_id):
266 more_widget_html = content_html = page
267 for page_num in itertools.count(1):
268 for entry in self._process_page(content_html):
271 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
275 more = self._download_json(
276 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
277 'Downloading page #%s' % page_num,
278 transform_source=uppercase_escape)
279 content_html = more['content_html']
280 if not content_html.strip():
281 # Some webpages show a "Load more" button but they don't
284 more_widget_html = more['load_more_widget_html']
287 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
288 def _process_page(self, content):
289 for video_id, video_title in self.extract_videos_from_page(content):
290 yield self.url_result(video_id, 'Youtube', video_id, video_title)
292 def extract_videos_from_page(self, page):
295 for mobj in re.finditer(self._VIDEO_RE, page):
296 # The link with index 0 is not the first video of the playlist (not sure if still actual)
297 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
299 video_id = mobj.group('id')
300 video_title = unescapeHTML(mobj.group('title'))
302 video_title = video_title.strip()
304 idx = ids_in_page.index(video_id)
305 if video_title and not titles_in_page[idx]:
306 titles_in_page[idx] = video_title
308 ids_in_page.append(video_id)
309 titles_in_page.append(video_title)
310 return zip(ids_in_page, titles_in_page)
313 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
314 def _process_page(self, content):
315 for playlist_id in orderedSet(re.findall(
316 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
318 yield self.url_result(
319 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
321 def _real_extract(self, url):
322 playlist_id = self._match_id(url)
323 webpage = self._download_webpage(url, playlist_id)
324 title = self._og_search_title(webpage, fatal=False)
325 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
328 class YoutubeIE(YoutubeBaseInfoExtractor):
329 IE_DESC = 'YouTube.com'
330 _VALID_URL = r"""(?x)^
332 (?:https?://|//) # http(s):// or protocol-independent URL
333 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
334 (?:www\.)?deturl\.com/www\.youtube\.com/|
335 (?:www\.)?pwnyoutube\.com/|
336 (?:www\.)?hooktube\.com/|
337 (?:www\.)?yourepeat\.com/|
338 tube\.majestyc\.net/|
339 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
340 (?:.*?\#/)? # handle anchor (#/) redirect urls
341 (?: # the various things that can precede the ID:
342 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
343 |(?: # or the v= param in all its forms
344 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
345 (?:\?|\#!?) # the params delimiter ? or # or #!
346 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
351 youtu\.be| # just youtu.be/xxxx
352 vid\.plus| # or vid.plus/xxxx
353 zwearz\.com/watch| # or zwearz.com/watch/xxxx
355 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
357 )? # all until now is optional -> you can pass the naked ID
358 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
361 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
362 WL # WL are handled by the watch later IE
365 (?(1).+)? # if we found the ID, everything can follow
366 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
367 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
369 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
370 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
371 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
372 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
373 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
374 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
375 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
376 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
377 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
378 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
379 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
380 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
381 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
382 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
383 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
384 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
385 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
386 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
390 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
391 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
392 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
393 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
394 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
395 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
396 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
398 # Apple HTTP Live Streaming
399 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
400 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
401 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
402 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
403 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
404 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
405 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
406 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
409 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
410 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
411 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
412 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
413 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
414 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
415 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
416 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
417 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
418 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
419 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
420 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
423 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
424 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
425 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
426 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
427 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
428 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
429 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
432 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
433 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
434 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
435 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
436 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
437 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
438 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
439 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
440 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
441 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
442 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
443 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
444 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
445 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
446 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
447 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
448 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
449 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
450 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
451 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
452 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
453 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
456 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
457 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
459 # Dash webm audio with opus inside
460 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
461 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
462 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
465 '_rtmp': {'protocol': 'rtmp'},
467 _SUBTITLE_FORMATS = ('ttml', 'vtt')
474 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
478 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
479 'uploader': 'Philipp Hagemeister',
480 'uploader_id': 'phihag',
481 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
482 'upload_date': '20121002',
483 'license': 'Standard YouTube License',
484 '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 .',
485 'categories': ['Science & Technology'],
486 'tags': ['youtube-dl'],
489 'dislike_count': int,
495 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
496 'note': 'Test generic use_cipher_signature video (#897)',
500 'upload_date': '20120506',
501 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
502 'alt_title': 'I Love It (feat. Charli XCX)',
503 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
504 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
505 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
506 'iconic ep', 'iconic', 'love', 'it'],
508 'uploader': 'Icona Pop',
509 'uploader_id': 'IconaPop',
510 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
511 'license': 'Standard YouTube License',
512 'creator': 'Icona Pop',
513 'track': 'I Love It (feat. Charli XCX)',
514 'artist': 'Icona Pop',
518 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
519 'note': 'Test VEVO video with age protection (#956)',
523 'upload_date': '20130703',
524 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
525 'alt_title': 'Tunnel Vision',
526 'description': 'md5:64249768eec3bc4276236606ea996373',
528 'uploader': 'justintimberlakeVEVO',
529 'uploader_id': 'justintimberlakeVEVO',
530 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
531 'license': 'Standard YouTube License',
532 'creator': 'Justin Timberlake',
533 'track': 'Tunnel Vision',
534 'artist': 'Justin Timberlake',
539 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
540 'note': 'Embed-only video (#1746)',
544 'upload_date': '20120608',
545 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
546 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
547 'uploader': 'SET India',
548 'uploader_id': 'setindia',
549 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
550 'license': 'Standard YouTube License',
555 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
556 'note': 'Use the first video ID in the URL',
560 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
561 'uploader': 'Philipp Hagemeister',
562 'uploader_id': 'phihag',
563 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
564 'upload_date': '20121002',
565 'license': 'Standard YouTube License',
566 '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 .',
567 'categories': ['Science & Technology'],
568 'tags': ['youtube-dl'],
571 'dislike_count': int,
574 'skip_download': True,
578 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
579 'note': '256k DASH audio (format 141) via DASH manifest',
583 'upload_date': '20121002',
584 'uploader_id': '8KVIDEO',
585 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
587 'uploader': '8KVIDEO',
588 'license': 'Standard YouTube License',
589 'title': 'UHDTV TEST 8K VIDEO.mp4'
592 'youtube_include_dash_manifest': True,
595 'skip': 'format 141 not served anymore',
597 # DASH manifest with encrypted signature
599 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
603 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
604 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
606 'uploader': 'AfrojackVEVO',
607 'uploader_id': 'AfrojackVEVO',
608 'upload_date': '20131011',
609 'license': 'Standard YouTube License',
612 'youtube_include_dash_manifest': True,
613 'format': '141/bestaudio[ext=m4a]',
616 # JS player signature function name containing $
618 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
622 'title': 'Taylor Swift - Shake It Off',
623 'alt_title': 'Shake It Off',
624 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
626 'uploader': 'TaylorSwiftVEVO',
627 'uploader_id': 'TaylorSwiftVEVO',
628 'upload_date': '20140818',
629 'license': 'Standard YouTube License',
630 'creator': 'Taylor Swift',
633 'youtube_include_dash_manifest': True,
634 'format': '141/bestaudio[ext=m4a]',
639 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
644 'upload_date': '20100909',
645 'uploader': 'The Amazing Atheist',
646 'uploader_id': 'TheAmazingAtheist',
647 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
648 'license': 'Standard YouTube License',
649 'title': 'Burning Everyone\'s Koran',
650 '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',
653 # Normal age-gate video (No vevo, embed allowed)
655 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
659 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
660 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
662 'uploader': 'The Witcher',
663 'uploader_id': 'WitcherGame',
664 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
665 'upload_date': '20140605',
666 'license': 'Standard YouTube License',
670 # Age-gate video with encrypted signature
672 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
676 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
677 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
679 'uploader': 'LloydVEVO',
680 'uploader_id': 'LloydVEVO',
681 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
682 'upload_date': '20110629',
683 'license': 'Standard YouTube License',
687 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
688 # YouTube Red ad is not captured for creator
690 'url': '__2ABJjxzNo',
695 'upload_date': '20100430',
696 'uploader_id': 'deadmau5',
697 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
698 'creator': 'deadmau5',
699 'description': 'md5:12c56784b8032162bb936a5f76d55360',
700 'uploader': 'deadmau5',
701 'license': 'Standard YouTube License',
702 'title': 'Deadmau5 - Some Chords (HD)',
703 'alt_title': 'Some Chords',
705 'expected_warnings': [
706 'DASH manifest missing',
709 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
711 'url': 'lqQg6PlCWgI',
716 'upload_date': '20150827',
717 'uploader_id': 'olympic',
718 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
719 'license': 'Standard YouTube License',
720 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
721 'uploader': 'Olympic',
722 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
725 'skip_download': 'requires avconv',
730 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
734 'stretched_ratio': 16 / 9.,
736 'upload_date': '20110310',
737 'uploader_id': 'AllenMeow',
738 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
739 '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',
741 'license': 'Standard YouTube License',
742 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
745 # url_encoded_fmt_stream_map is empty string
747 'url': 'qEJwOuvDf7I',
751 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
753 'upload_date': '20150404',
754 'uploader_id': 'spbelect',
755 'uploader': 'Наблюдатели Петербурга',
758 'skip_download': 'requires avconv',
760 'skip': 'This live event has ended.',
762 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
764 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
768 'title': 'md5:7b81415841e02ecd4313668cde88737a',
769 'description': 'md5:116377fd2963b81ec4ce64b542173306',
771 'upload_date': '20150625',
772 'uploader_id': 'dorappi2000',
773 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
774 'uploader': 'dorappi2000',
775 'license': 'Standard YouTube License',
776 'formats': 'mincount:32',
779 # DASH manifest with segment_list
781 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
782 'md5': '8ce563a1d667b599d21064e982ab9e31',
786 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
787 'uploader': 'Airtek',
788 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
789 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
790 'license': 'Standard YouTube License',
791 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
794 'youtube_include_dash_manifest': True,
795 'format': '135', # bestvideo
797 'skip': 'This live event has ended.',
800 # Multifeed videos (multiple cameras), URL is for Main Camera
801 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
804 'title': 'teamPGP: Rocket League Noob Stream',
805 'description': 'md5:dc7872fb300e143831327f1bae3af010',
811 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
812 'description': 'md5:dc7872fb300e143831327f1bae3af010',
814 'upload_date': '20150721',
815 'uploader': 'Beer Games Beer',
816 'uploader_id': 'beergamesbeer',
817 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
818 'license': 'Standard YouTube License',
824 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
825 'description': 'md5:dc7872fb300e143831327f1bae3af010',
827 'upload_date': '20150721',
828 'uploader': 'Beer Games Beer',
829 'uploader_id': 'beergamesbeer',
830 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
831 'license': 'Standard YouTube License',
837 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
838 'description': 'md5:dc7872fb300e143831327f1bae3af010',
840 'upload_date': '20150721',
841 'uploader': 'Beer Games Beer',
842 'uploader_id': 'beergamesbeer',
843 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
844 'license': 'Standard YouTube License',
850 'title': 'teamPGP: Rocket League Noob Stream (zim)',
851 'description': 'md5:dc7872fb300e143831327f1bae3af010',
853 'upload_date': '20150721',
854 'uploader': 'Beer Games Beer',
855 'uploader_id': 'beergamesbeer',
856 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
857 'license': 'Standard YouTube License',
861 'skip_download': True,
865 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
866 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
869 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
872 'skip': 'Not multifeed anymore',
875 'url': 'https://vid.plus/FlRa-iH7PGw',
876 'only_matching': True,
879 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
880 'only_matching': True,
883 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
884 # Also tests cut-off URL expansion in video description (see
885 # https://github.com/rg3/youtube-dl/issues/1892,
886 # https://github.com/rg3/youtube-dl/issues/8164)
887 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
891 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
892 'alt_title': 'Dark Walk',
893 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
895 'upload_date': '20151119',
896 'uploader_id': 'IronSoulElf',
897 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
898 'uploader': 'IronSoulElf',
899 'license': 'Standard YouTube License',
900 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
903 'skip_download': True,
907 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
908 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
909 'only_matching': True,
912 # Video with yt:stretch=17:0
913 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
917 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
918 'description': 'md5:ee18a25c350637c8faff806845bddee9',
919 'upload_date': '20151107',
920 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
921 'uploader': 'CH GAMER DROID',
924 'skip_download': True,
926 'skip': 'This video does not exist.',
929 # Video licensed under Creative Commons
930 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
934 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
935 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
937 'upload_date': '20150127',
938 'uploader_id': 'BerkmanCenter',
939 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
940 'uploader': 'The Berkman Klein Center for Internet & Society',
941 'license': 'Creative Commons Attribution license (reuse allowed)',
944 'skip_download': True,
948 # Channel-like uploader_url
949 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
953 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
954 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
956 'upload_date': '20151119',
957 'uploader': 'Bernie 2016',
958 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
959 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
960 'license': 'Creative Commons Attribution license (reuse allowed)',
963 'skip_download': True,
967 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
968 'only_matching': True,
971 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
972 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
973 'only_matching': True,
976 # Rental video preview
977 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
981 'title': 'Piku - Trailer',
982 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
983 'upload_date': '20150811',
984 'uploader': 'FlixMatrix',
985 'uploader_id': 'FlixMatrixKaravan',
986 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
987 'license': 'Standard YouTube License',
990 'skip_download': True,
994 # YouTube Red video with episode data
995 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
999 'title': 'Isolation - Mind Field (Ep 1)',
1000 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
1002 'upload_date': '20170118',
1003 'uploader': 'Vsauce',
1004 'uploader_id': 'Vsauce',
1005 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1006 'license': 'Standard YouTube License',
1007 'series': 'Mind Field',
1009 'episode_number': 1,
1012 'skip_download': True,
1014 'expected_warnings': [
1015 'Skipping DASH manifest',
1019 # The following content has been identified by the YouTube community
1020 # as inappropriate or offensive to some audiences.
1021 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1023 'id': '6SJNVb0GnPI',
1025 'title': 'Race Differences in Intelligence',
1026 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1028 'upload_date': '20140124',
1029 'uploader': 'New Century Foundation',
1030 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1031 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1032 'license': 'Standard YouTube License',
1036 'skip_download': True,
1041 'url': '1t24XAntNCY',
1042 'only_matching': True,
1045 # geo restricted to JP
1046 'url': 'sJL6WA-aGkQ',
1047 'only_matching': True,
1050 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1051 'only_matching': True,
1055 def __init__(self, *args, **kwargs):
1056 super(YoutubeIE, self).__init__(*args, **kwargs)
1057 self._player_cache = {}
1059 def report_video_info_webpage_download(self, video_id):
1060 """Report attempt to download video info webpage."""
1061 self.to_screen('%s: Downloading video info webpage' % video_id)
1063 def report_information_extraction(self, video_id):
1064 """Report attempt to extract video information."""
1065 self.to_screen('%s: Extracting video information' % video_id)
1067 def report_unavailable_format(self, video_id, format):
1068 """Report extracted video URL."""
1069 self.to_screen('%s: Format %s not available' % (video_id, format))
1071 def report_rtmp_download(self):
1072 """Indicate the download will use the RTMP protocol."""
1073 self.to_screen('RTMP download detected')
1075 def _signature_cache_id(self, example_sig):
1076 """ Return a string representation of a signature """
1077 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1079 def _extract_signature_function(self, video_id, player_url, example_sig):
1081 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1084 raise ExtractorError('Cannot identify player %r' % player_url)
1085 player_type = id_m.group('ext')
1086 player_id = id_m.group('id')
1088 # Read from filesystem cache
1089 func_id = '%s_%s_%s' % (
1090 player_type, player_id, self._signature_cache_id(example_sig))
1091 assert os.path.basename(func_id) == func_id
1093 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1094 if cache_spec is not None:
1095 return lambda s: ''.join(s[i] for i in cache_spec)
1098 'Downloading player %s' % player_url
1099 if self._downloader.params.get('verbose') else
1100 'Downloading %s player %s' % (player_type, player_id)
1102 if player_type == 'js':
1103 code = self._download_webpage(
1104 player_url, video_id,
1106 errnote='Download of %s failed' % player_url)
1107 res = self._parse_sig_js(code)
1108 elif player_type == 'swf':
1109 urlh = self._request_webpage(
1110 player_url, video_id,
1112 errnote='Download of %s failed' % player_url)
1114 res = self._parse_sig_swf(code)
1116 assert False, 'Invalid player type %r' % player_type
1118 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1119 cache_res = res(test_string)
1120 cache_spec = [ord(c) for c in cache_res]
1122 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1125 def _print_sig_code(self, func, example_sig):
1126 def gen_sig_code(idxs):
1127 def _genslice(start, end, step):
1128 starts = '' if start == 0 else str(start)
1129 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1130 steps = '' if step == 1 else (':%d' % step)
1131 return 's[%s%s%s]' % (starts, ends, steps)
1134 # Quelch pyflakes warnings - start will be set when step is set
1135 start = '(Never used)'
1136 for i, prev in zip(idxs[1:], idxs[:-1]):
1137 if step is not None:
1138 if i - prev == step:
1140 yield _genslice(start, prev, step)
1143 if i - prev in [-1, 1]:
1148 yield 's[%d]' % prev
1152 yield _genslice(start, i, step)
1154 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1155 cache_res = func(test_string)
1156 cache_spec = [ord(c) for c in cache_res]
1157 expr_code = ' + '.join(gen_sig_code(cache_spec))
1158 signature_id_tuple = '(%s)' % (
1159 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1160 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1161 ' return %s\n') % (signature_id_tuple, expr_code)
1162 self.to_screen('Extracted signature function:\n' + code)
1164 def _parse_sig_js(self, jscode):
1165 funcname = self._search_regex(
1166 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1167 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1168 jscode, 'Initial JS player signature function name', group='sig')
1170 jsi = JSInterpreter(jscode)
1171 initial_function = jsi.extract_function(funcname)
1172 return lambda s: initial_function([s])
1174 def _parse_sig_swf(self, file_contents):
1175 swfi = SWFInterpreter(file_contents)
1176 TARGET_CLASSNAME = 'SignatureDecipher'
1177 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1178 initial_function = swfi.extract_function(searched_class, 'decipher')
1179 return lambda s: initial_function([s])
1181 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1182 """Turn the encrypted s field into a working signature"""
1184 if player_url is None:
1185 raise ExtractorError('Cannot decrypt signature without player_url')
1187 if player_url.startswith('//'):
1188 player_url = 'https:' + player_url
1189 elif not re.match(r'https?://', player_url):
1190 player_url = compat_urlparse.urljoin(
1191 'https://www.youtube.com', player_url)
1193 player_id = (player_url, self._signature_cache_id(s))
1194 if player_id not in self._player_cache:
1195 func = self._extract_signature_function(
1196 video_id, player_url, s
1198 self._player_cache[player_id] = func
1199 func = self._player_cache[player_id]
1200 if self._downloader.params.get('youtube_print_sig_code'):
1201 self._print_sig_code(func, s)
1203 except Exception as e:
1204 tb = traceback.format_exc()
1205 raise ExtractorError(
1206 'Signature extraction failed: ' + tb, cause=e)
1208 def _get_subtitles(self, video_id, webpage):
1210 subs_doc = self._download_xml(
1211 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1212 video_id, note=False)
1213 except ExtractorError as err:
1214 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1218 for track in subs_doc.findall('track'):
1219 lang = track.attrib['lang_code']
1220 if lang in sub_lang_list:
1223 for ext in self._SUBTITLE_FORMATS:
1224 params = compat_urllib_parse_urlencode({
1228 'name': track.attrib['name'].encode('utf-8'),
1230 sub_formats.append({
1231 'url': 'https://www.youtube.com/api/timedtext?' + params,
1234 sub_lang_list[lang] = sub_formats
1235 if not sub_lang_list:
1236 self._downloader.report_warning('video doesn\'t have subtitles')
1238 return sub_lang_list
1240 def _get_ytplayer_config(self, video_id, webpage):
1242 # User data may contain arbitrary character sequences that may affect
1243 # JSON extraction with regex, e.g. when '};' is contained the second
1244 # regex won't capture the whole JSON. Yet working around by trying more
1245 # concrete regex first keeping in mind proper quoted string handling
1246 # to be implemented in future that will replace this workaround (see
1247 # https://github.com/rg3/youtube-dl/issues/7468,
1248 # https://github.com/rg3/youtube-dl/pull/7599)
1249 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1250 r';ytplayer\.config\s*=\s*({.+?});',
1252 config = self._search_regex(
1253 patterns, webpage, 'ytplayer.config', default=None)
1255 return self._parse_json(
1256 uppercase_escape(config), video_id, fatal=False)
1258 def _get_automatic_captions(self, video_id, webpage):
1259 """We need the webpage for getting the captions url, pass it as an
1260 argument to speed up the process."""
1261 self.to_screen('%s: Looking for automatic captions' % video_id)
1262 player_config = self._get_ytplayer_config(video_id, webpage)
1263 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1264 if not player_config:
1265 self._downloader.report_warning(err_msg)
1268 args = player_config['args']
1269 caption_url = args.get('ttsurl')
1271 timestamp = args['timestamp']
1272 # We get the available subtitles
1273 list_params = compat_urllib_parse_urlencode({
1278 list_url = caption_url + '&' + list_params
1279 caption_list = self._download_xml(list_url, video_id)
1280 original_lang_node = caption_list.find('track')
1281 if original_lang_node is None:
1282 self._downloader.report_warning('Video doesn\'t have automatic captions')
1284 original_lang = original_lang_node.attrib['lang_code']
1285 caption_kind = original_lang_node.attrib.get('kind', '')
1288 for lang_node in caption_list.findall('target'):
1289 sub_lang = lang_node.attrib['lang_code']
1291 for ext in self._SUBTITLE_FORMATS:
1292 params = compat_urllib_parse_urlencode({
1293 'lang': original_lang,
1297 'kind': caption_kind,
1299 sub_formats.append({
1300 'url': caption_url + '&' + params,
1303 sub_lang_list[sub_lang] = sub_formats
1304 return sub_lang_list
1306 def make_captions(sub_url, sub_langs):
1307 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1308 caption_qs = compat_parse_qs(parsed_sub_url.query)
1310 for sub_lang in sub_langs:
1312 for ext in self._SUBTITLE_FORMATS:
1314 'tlang': [sub_lang],
1317 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1318 query=compat_urllib_parse_urlencode(caption_qs, True)))
1319 sub_formats.append({
1323 captions[sub_lang] = sub_formats
1326 # New captions format as of 22.06.2017
1327 player_response = args.get('player_response')
1328 if player_response and isinstance(player_response, compat_str):
1329 player_response = self._parse_json(
1330 player_response, video_id, fatal=False)
1332 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1333 base_url = renderer['captionTracks'][0]['baseUrl']
1335 for lang in renderer['translationLanguages']:
1336 lang_code = lang.get('languageCode')
1338 sub_lang_list.append(lang_code)
1339 return make_captions(base_url, sub_lang_list)
1341 # Some videos don't provide ttsurl but rather caption_tracks and
1342 # caption_translation_languages (e.g. 20LmZk1hakA)
1343 # Does not used anymore as of 22.06.2017
1344 caption_tracks = args['caption_tracks']
1345 caption_translation_languages = args['caption_translation_languages']
1346 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1348 for lang in caption_translation_languages.split(','):
1349 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1350 sub_lang = lang_qs.get('lc', [None])[0]
1352 sub_lang_list.append(sub_lang)
1353 return make_captions(caption_url, sub_lang_list)
1354 # An extractor error can be raise by the download process if there are
1355 # no automatic captions but there are subtitles
1356 except (KeyError, IndexError, ExtractorError):
1357 self._downloader.report_warning(err_msg)
1360 def _mark_watched(self, video_id, video_info):
1361 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1362 if not playback_url:
1364 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1365 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1367 # cpn generation algorithm is reverse engineered from base.js.
1368 # In fact it works even with dummy cpn.
1369 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1370 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1376 playback_url = compat_urlparse.urlunparse(
1377 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1379 self._download_webpage(
1380 playback_url, video_id, 'Marking watched',
1381 'Unable to mark watched', fatal=False)
1384 def _extract_urls(webpage):
1385 # Embedded YouTube player
1387 unescapeHTML(mobj.group('url'))
1388 for mobj in re.finditer(r'''(?x)
1398 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1399 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1402 # lazyYT YouTube embed
1403 entries.extend(list(map(
1405 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1407 # Wordpress "YouTube Video Importer" plugin
1408 matches = re.findall(r'''(?x)<div[^>]+
1409 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1410 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1411 entries.extend(m[-1] for m in matches)
1416 def _extract_url(webpage):
1417 urls = YoutubeIE._extract_urls(webpage)
1418 return urls[0] if urls else None
1421 def extract_id(cls, url):
1422 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1424 raise ExtractorError('Invalid URL: %s' % url)
1425 video_id = mobj.group(2)
1428 def _extract_annotations(self, video_id):
1429 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1430 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1433 def _extract_chapters(description, duration):
1436 chapter_lines = re.findall(
1437 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1439 if not chapter_lines:
1442 for next_num, (chapter_line, time_point) in enumerate(
1443 chapter_lines, start=1):
1444 start_time = parse_duration(time_point)
1445 if start_time is None:
1447 if start_time > duration:
1449 end_time = (duration if next_num == len(chapter_lines)
1450 else parse_duration(chapter_lines[next_num][1]))
1451 if end_time is None:
1453 if end_time > duration:
1455 if start_time > end_time:
1457 chapter_title = re.sub(
1458 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1459 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1461 'start_time': start_time,
1462 'end_time': end_time,
1463 'title': chapter_title,
1467 def _real_extract(self, url):
1468 url, smuggled_data = unsmuggle_url(url, {})
1471 'http' if self._downloader.params.get('prefer_insecure', False)
1476 parsed_url = compat_urllib_parse_urlparse(url)
1477 for component in [parsed_url.fragment, parsed_url.query]:
1478 query = compat_parse_qs(component)
1479 if start_time is None and 't' in query:
1480 start_time = parse_duration(query['t'][0])
1481 if start_time is None and 'start' in query:
1482 start_time = parse_duration(query['start'][0])
1483 if end_time is None and 'end' in query:
1484 end_time = parse_duration(query['end'][0])
1486 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1487 mobj = re.search(self._NEXT_URL_RE, url)
1489 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1490 video_id = self.extract_id(url)
1493 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1494 video_webpage = self._download_webpage(url, video_id)
1496 # Attempt to extract SWF player URL
1497 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1498 if mobj is not None:
1499 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1505 def add_dash_mpd(video_info):
1506 dash_mpd = video_info.get('dashmpd')
1507 if dash_mpd and dash_mpd[0] not in dash_mpds:
1508 dash_mpds.append(dash_mpd[0])
1513 def extract_view_count(v_info):
1514 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1517 embed_webpage = None
1518 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1520 # We simulate the access to the video from www.youtube.com/v/{video_id}
1521 # this can be viewed without login into Youtube
1522 url = proto + '://www.youtube.com/embed/%s' % video_id
1523 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1524 data = compat_urllib_parse_urlencode({
1525 'video_id': video_id,
1526 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1527 'sts': self._search_regex(
1528 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1530 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1531 video_info_webpage = self._download_webpage(
1532 video_info_url, video_id,
1533 note='Refetching age-gated info webpage',
1534 errnote='unable to download video info webpage')
1535 video_info = compat_parse_qs(video_info_webpage)
1536 add_dash_mpd(video_info)
1541 # Try looking directly into the video webpage
1542 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1544 args = ytplayer_config['args']
1545 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1546 # Convert to the same format returned by compat_parse_qs
1547 video_info = dict((k, [v]) for k, v in args.items())
1548 add_dash_mpd(video_info)
1549 # Rental video is not rented but preview is available (e.g.
1550 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1551 # https://github.com/rg3/youtube-dl/issues/10532)
1552 if not video_info and args.get('ypc_vid'):
1553 return self.url_result(
1554 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1555 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1557 sts = ytplayer_config.get('sts')
1558 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1559 # We also try looking in get_video_info since it may contain different dashmpd
1560 # URL that points to a DASH manifest with possibly different itag set (some itags
1561 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1562 # manifest pointed by get_video_info's dashmpd).
1563 # The general idea is to take a union of itags of both DASH manifests (for example
1564 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1565 self.report_video_info_webpage_download(video_id)
1566 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1568 'video_id': video_id,
1578 video_info_webpage = self._download_webpage(
1579 '%s://www.youtube.com/get_video_info' % proto,
1580 video_id, note=False,
1581 errnote='unable to download video info webpage',
1582 fatal=False, query=query)
1583 if not video_info_webpage:
1585 get_video_info = compat_parse_qs(video_info_webpage)
1586 add_dash_mpd(get_video_info)
1587 if view_count is None:
1588 view_count = extract_view_count(get_video_info)
1590 video_info = get_video_info
1591 if 'token' in get_video_info:
1592 # Different get_video_info requests may report different results, e.g.
1593 # some may report video unavailability, but some may serve it without
1594 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1595 # the original webpage as well as el=info and el=embedded get_video_info
1596 # requests report video unavailability due to geo restriction while
1597 # el=detailpage succeeds and returns valid data). This is probably
1598 # due to YouTube measures against IP ranges of hosting providers.
1599 # Working around by preferring the first succeeded video_info containing
1600 # the token if no such video_info yet was found.
1601 if 'token' not in video_info:
1602 video_info = get_video_info
1605 def extract_unavailable_message():
1606 return self._html_search_regex(
1607 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1608 video_webpage, 'unavailable message', default=None)
1610 if 'token' not in video_info:
1611 if 'reason' in video_info:
1612 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1613 regions_allowed = self._html_search_meta(
1614 'regionsAllowed', video_webpage, default=None)
1615 countries = regions_allowed.split(',') if regions_allowed else None
1616 self.raise_geo_restricted(
1617 msg=video_info['reason'][0], countries=countries)
1618 reason = video_info['reason'][0]
1619 if 'Invalid parameters' in reason:
1620 unavailable_message = extract_unavailable_message()
1621 if unavailable_message:
1622 reason = unavailable_message
1623 raise ExtractorError(
1624 'YouTube said: %s' % reason,
1625 expected=True, video_id=video_id)
1627 raise ExtractorError(
1628 '"token" parameter not in video info for unknown reason',
1632 if 'title' in video_info:
1633 video_title = video_info['title'][0]
1635 self._downloader.report_warning('Unable to extract video title')
1639 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1640 if video_description:
1643 redir_url = compat_urlparse.urljoin(url, m.group(1))
1644 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1645 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1646 qs = compat_parse_qs(parsed_redir_url.query)
1652 description_original = video_description = re.sub(r'''(?x)
1654 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1655 (?:title|href)="([^"]+)"\s+
1656 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1660 ''', replace_url, video_description)
1661 video_description = clean_html(video_description)
1663 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1665 video_description = unescapeHTML(fd_mobj.group(1))
1667 video_description = ''
1669 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1670 if not self._downloader.params.get('noplaylist'):
1673 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1674 for feed in multifeed_metadata_list.split(','):
1675 # Unquote should take place before split on comma (,) since textual
1676 # fields may contain comma as well (see
1677 # https://github.com/rg3/youtube-dl/issues/8536)
1678 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1680 '_type': 'url_transparent',
1681 'ie_key': 'Youtube',
1683 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1684 {'force_singlefeed': True}),
1685 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1687 feed_ids.append(feed_data['id'][0])
1689 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1690 % (', '.join(feed_ids), video_id))
1691 return self.playlist_result(entries, video_id, video_title, video_description)
1692 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1694 if view_count is None:
1695 view_count = extract_view_count(video_info)
1697 # Check for "rental" videos
1698 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1699 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1701 def _extract_filesize(media_url):
1702 return int_or_none(self._search_regex(
1703 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1705 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1706 self.report_rtmp_download()
1708 'format_id': '_rtmp',
1710 'url': video_info['conn'][0],
1711 'player_url': player_url,
1713 elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1714 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1715 if 'rtmpe%3Dyes' in encoded_url_map:
1716 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1718 fmt_list = video_info.get('fmt_list', [''])[0]
1720 for fmt in fmt_list.split(','):
1721 spec = fmt.split('/')
1723 width_height = spec[1].split('x')
1724 if len(width_height) == 2:
1725 formats_spec[spec[0]] = {
1726 'resolution': spec[1],
1727 'width': int_or_none(width_height[0]),
1728 'height': int_or_none(width_height[1]),
1730 q = qualities(['small', 'medium', 'hd720'])
1732 for url_data_str in encoded_url_map.split(','):
1733 url_data = compat_parse_qs(url_data_str)
1734 if 'itag' not in url_data or 'url' not in url_data:
1736 format_id = url_data['itag'][0]
1737 url = url_data['url'][0]
1739 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1740 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1741 jsplayer_url_json = self._search_regex(
1743 embed_webpage if age_gate else video_webpage,
1744 'JS player URL (1)', default=None)
1745 if not jsplayer_url_json and not age_gate:
1746 # We need the embed website after all
1747 if embed_webpage is None:
1748 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1749 embed_webpage = self._download_webpage(
1750 embed_url, video_id, 'Downloading embed webpage')
1751 jsplayer_url_json = self._search_regex(
1752 ASSETS_RE, embed_webpage, 'JS player URL')
1754 player_url = json.loads(jsplayer_url_json)
1755 if player_url is None:
1756 player_url_json = self._search_regex(
1757 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1758 video_webpage, 'age gate player URL')
1759 player_url = json.loads(player_url_json)
1761 if 'sig' in url_data:
1762 url += '&signature=' + url_data['sig'][0]
1763 elif 's' in url_data:
1764 encrypted_sig = url_data['s'][0]
1766 if self._downloader.params.get('verbose'):
1767 if player_url is None:
1768 player_version = 'unknown'
1769 player_desc = 'unknown'
1771 if player_url.endswith('swf'):
1772 player_version = self._search_regex(
1773 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1774 'flash player', fatal=False)
1775 player_desc = 'flash player %s' % player_version
1777 player_version = self._search_regex(
1778 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1779 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1781 'html5 player', fatal=False)
1782 player_desc = 'html5 player %s' % player_version
1784 parts_sizes = self._signature_cache_id(encrypted_sig)
1785 self.to_screen('{%s} signature length %s, %s' %
1786 (format_id, parts_sizes, player_desc))
1788 signature = self._decrypt_signature(
1789 encrypted_sig, video_id, player_url, age_gate)
1790 url += '&signature=' + signature
1791 if 'ratebypass' not in url:
1792 url += '&ratebypass=yes'
1795 'format_id': format_id,
1797 'player_url': player_url,
1799 if format_id in self._formats:
1800 dct.update(self._formats[format_id])
1801 if format_id in formats_spec:
1802 dct.update(formats_spec[format_id])
1804 # Some itags are not included in DASH manifest thus corresponding formats will
1805 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1806 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1807 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1808 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1810 filesize = int_or_none(url_data.get(
1811 'clen', [None])[0]) or _extract_filesize(url)
1813 quality = url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0]
1816 'filesize': filesize,
1817 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1820 'fps': int_or_none(url_data.get('fps', [None])[0]),
1821 'format_note': quality,
1822 'quality': q(quality),
1824 for key, value in more_fields.items():
1827 type_ = url_data.get('type', [None])[0]
1829 type_split = type_.split(';')
1830 kind_ext = type_split[0].split('/')
1831 if len(kind_ext) == 2:
1833 dct['ext'] = mimetype2ext(type_split[0])
1834 if kind in ('audio', 'video'):
1836 for mobj in re.finditer(
1837 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1838 if mobj.group('key') == 'codecs':
1839 codecs = mobj.group('val')
1842 dct.update(parse_codecs(codecs))
1843 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1844 dct['downloader_options'] = {
1845 # Youtube throttles chunks >~10M
1846 'http_chunk_size': 10485760,
1849 elif video_info.get('hlsvp'):
1850 manifest_url = video_info['hlsvp'][0]
1852 m3u8_formats = self._extract_m3u8_formats(
1853 manifest_url, video_id, 'mp4', fatal=False)
1854 for a_format in m3u8_formats:
1855 itag = self._search_regex(
1856 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1858 a_format['format_id'] = itag
1859 if itag in self._formats:
1860 dct = self._formats[itag].copy()
1861 dct.update(a_format)
1863 a_format['player_url'] = player_url
1864 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1865 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1866 formats.append(a_format)
1868 error_message = clean_html(video_info.get('reason', [None])[0])
1869 if not error_message:
1870 error_message = extract_unavailable_message()
1872 raise ExtractorError(error_message, expected=True)
1873 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1876 video_uploader = try_get(video_info, lambda x: x['author'][0], compat_str)
1878 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1880 self._downloader.report_warning('unable to extract uploader name')
1883 video_uploader_id = None
1884 video_uploader_url = None
1886 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1888 if mobj is not None:
1889 video_uploader_id = mobj.group('uploader_id')
1890 video_uploader_url = mobj.group('uploader_url')
1892 self._downloader.report_warning('unable to extract uploader nickname')
1895 # We try first to get a high quality image:
1896 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1897 video_webpage, re.DOTALL)
1898 if m_thumb is not None:
1899 video_thumbnail = m_thumb.group(1)
1900 elif 'thumbnail_url' not in video_info:
1901 self._downloader.report_warning('unable to extract video thumbnail')
1902 video_thumbnail = None
1903 else: # don't panic if we can't find it
1904 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1907 upload_date = self._html_search_meta(
1908 'datePublished', video_webpage, 'upload date', default=None)
1910 upload_date = self._search_regex(
1911 [r'(?s)id="eow-date.*?>(.*?)</span>',
1912 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1913 video_webpage, 'upload date', default=None)
1914 upload_date = unified_strdate(upload_date)
1916 video_license = self._html_search_regex(
1917 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1918 video_webpage, 'license', default=None)
1920 m_music = re.search(
1922 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1930 \bhref=["\']/red[^>]*>| # drop possible
1931 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1938 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1939 video_creator = clean_html(m_music.group('creator'))
1941 video_alt_title = video_creator = None
1943 def extract_meta(field):
1944 return self._html_search_regex(
1945 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
1946 video_webpage, field, default=None)
1948 track = extract_meta('Song')
1949 artist = extract_meta('Artist')
1951 m_episode = re.search(
1952 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1955 series = m_episode.group('series')
1956 season_number = int(m_episode.group('season'))
1957 episode_number = int(m_episode.group('episode'))
1959 series = season_number = episode_number = None
1961 m_cat_container = self._search_regex(
1962 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1963 video_webpage, 'categories', default=None)
1965 category = self._html_search_regex(
1966 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1968 video_categories = None if category is None else [category]
1970 video_categories = None
1973 unescapeHTML(m.group('content'))
1974 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1976 def _extract_count(count_name):
1977 return str_to_int(self._search_regex(
1978 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1979 % re.escape(count_name),
1980 video_webpage, count_name, default=None))
1982 like_count = _extract_count('like')
1983 dislike_count = _extract_count('dislike')
1986 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1987 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1989 video_duration = try_get(
1990 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1991 if not video_duration:
1992 video_duration = parse_duration(self._html_search_meta(
1993 'duration', video_webpage, 'video duration'))
1996 video_annotations = None
1997 if self._downloader.params.get('writeannotations', False):
1998 video_annotations = self._extract_annotations(video_id)
2000 chapters = self._extract_chapters(description_original, video_duration)
2002 # Look for the DASH manifest
2003 if self._downloader.params.get('youtube_include_dash_manifest', True):
2004 dash_mpd_fatal = True
2005 for mpd_url in dash_mpds:
2008 def decrypt_sig(mobj):
2010 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2011 return '/signature/%s' % dec_s
2013 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2015 for df in self._extract_mpd_formats(
2016 mpd_url, video_id, fatal=dash_mpd_fatal,
2017 formats_dict=self._formats):
2018 if not df.get('filesize'):
2019 df['filesize'] = _extract_filesize(df['url'])
2020 # Do not overwrite DASH format found in some previous DASH manifest
2021 if df['format_id'] not in dash_formats:
2022 dash_formats[df['format_id']] = df
2023 # Additional DASH manifests may end up in HTTP Error 403 therefore
2024 # allow them to fail without bug report message if we already have
2025 # some DASH manifest succeeded. This is temporary workaround to reduce
2026 # burst of bug reports until we figure out the reason and whether it
2027 # can be fixed at all.
2028 dash_mpd_fatal = False
2029 except (ExtractorError, KeyError) as e:
2030 self.report_warning(
2031 'Skipping DASH manifest: %r' % e, video_id)
2033 # Remove the formats we found through non-DASH, they
2034 # contain less info and it can be wrong, because we use
2035 # fixed values (for example the resolution). See
2036 # https://github.com/rg3/youtube-dl/issues/5774 for an
2038 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2039 formats.extend(dash_formats.values())
2041 # Check for malformed aspect ratio
2042 stretched_m = re.search(
2043 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2046 w = float(stretched_m.group('w'))
2047 h = float(stretched_m.group('h'))
2048 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2049 # We will only process correct ratios.
2053 if f.get('vcodec') != 'none':
2054 f['stretched_ratio'] = ratio
2056 self._sort_formats(formats)
2058 self.mark_watched(video_id, video_info)
2062 'uploader': video_uploader,
2063 'uploader_id': video_uploader_id,
2064 'uploader_url': video_uploader_url,
2065 'upload_date': upload_date,
2066 'license': video_license,
2067 'creator': video_creator or artist,
2068 'title': video_title,
2069 'alt_title': video_alt_title or track,
2070 'thumbnail': video_thumbnail,
2071 'description': video_description,
2072 'categories': video_categories,
2074 'subtitles': video_subtitles,
2075 'automatic_captions': automatic_captions,
2076 'duration': video_duration,
2077 'age_limit': 18 if age_gate else 0,
2078 'annotations': video_annotations,
2079 'chapters': chapters,
2080 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2081 'view_count': view_count,
2082 'like_count': like_count,
2083 'dislike_count': dislike_count,
2084 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2087 'start_time': start_time,
2088 'end_time': end_time,
2090 'season_number': season_number,
2091 'episode_number': episode_number,
2097 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2098 IE_DESC = 'YouTube.com playlists'
2099 _VALID_URL = r"""(?x)(?:
2105 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2106 \? (?:.*?[&;])*? (?:p|a|list)=
2109 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2112 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2113 # Top tracks, they can also include dots
2119 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2120 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2121 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2122 IE_NAME = 'youtube:playlist'
2124 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2126 'title': 'ytdl test PL',
2127 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2129 'playlist_count': 3,
2131 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2133 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2134 'title': 'YDL_Empty_List',
2136 'playlist_count': 0,
2137 'skip': 'This playlist is private',
2139 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2140 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2142 'title': '29C3: Not my department',
2143 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2145 'playlist_count': 95,
2147 'note': 'issue #673',
2148 'url': 'PLBB231211A4F62143',
2150 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2151 'id': 'PLBB231211A4F62143',
2153 'playlist_mincount': 26,
2155 'note': 'Large playlist',
2156 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2158 'title': 'Uploads from Cauchemar',
2159 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2161 'playlist_mincount': 799,
2163 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2165 'title': 'YDL_safe_search',
2166 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2168 'playlist_count': 2,
2169 'skip': 'This playlist is private',
2172 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2173 'playlist_count': 4,
2176 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2179 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2180 'playlist_mincount': 485,
2182 'title': '2017 華語最新單曲 (2/24更新)',
2183 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2186 'note': 'Embedded SWF player',
2187 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2188 'playlist_count': 4,
2191 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2194 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2195 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2197 'title': 'Uploads from Interstellar Movie',
2198 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2200 'playlist_mincount': 21,
2202 # Playlist URL that does not actually serve a playlist
2203 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2205 'id': 'FqZTN594JQw',
2207 'title': "Smiley's People 01 detective, Adventure Series, Action",
2208 'uploader': 'STREEM',
2209 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2210 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2211 'upload_date': '20150526',
2212 'license': 'Standard YouTube License',
2213 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2214 'categories': ['People & Blogs'],
2217 'dislike_count': int,
2220 'skip_download': True,
2222 'add_ie': [YoutubeIE.ie_key()],
2224 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2226 'id': 'yeWKywCrFtk',
2228 'title': 'Small Scale Baler and Braiding Rugs',
2229 'uploader': 'Backus-Page House Museum',
2230 'uploader_id': 'backuspagemuseum',
2231 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2232 'upload_date': '20161008',
2233 'license': 'Standard YouTube License',
2234 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2235 'categories': ['Nonprofits & Activism'],
2238 'dislike_count': int,
2242 'skip_download': True,
2245 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2246 'only_matching': True,
2248 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2249 'only_matching': True,
2252 def _real_initialize(self):
2255 def _extract_mix(self, playlist_id):
2256 # The mixes are generated from a single video
2257 # the id of the playlist is just 'RD' + video_id
2259 last_id = playlist_id[-11:]
2260 for n in itertools.count(1):
2261 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2262 webpage = self._download_webpage(
2263 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2264 new_ids = orderedSet(re.findall(
2265 r'''(?xs)data-video-username=".*?".*?
2266 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2268 # Fetch new pages until all the videos are repeated, it seems that
2269 # there are always 51 unique videos.
2270 new_ids = [_id for _id in new_ids if _id not in ids]
2276 url_results = self._ids_to_results(ids)
2278 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2280 search_title('playlist-title') or
2281 search_title('title long-title') or
2282 search_title('title'))
2283 title = clean_html(title_span)
2285 return self.playlist_result(url_results, playlist_id, title)
2287 def _extract_playlist(self, playlist_id):
2288 url = self._TEMPLATE_URL % playlist_id
2289 page = self._download_webpage(url, playlist_id)
2291 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2292 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2293 match = match.strip()
2294 # Check if the playlist exists or is private
2295 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2297 reason = mobj.group('reason')
2298 message = 'This playlist %s' % reason
2299 if 'private' in reason:
2300 message += ', use --username or --netrc to access it'
2302 raise ExtractorError(message, expected=True)
2303 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2304 raise ExtractorError(
2305 'Invalid parameters. Maybe URL is incorrect.',
2307 elif re.match(r'[^<]*Choose your language[^<]*', match):
2310 self.report_warning('Youtube gives an alert message: ' + match)
2312 playlist_title = self._html_search_regex(
2313 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2314 page, 'title', default=None)
2316 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2317 uploader = self._search_regex(
2318 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2319 page, 'uploader', default=None)
2321 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2324 uploader_id = mobj.group('uploader_id')
2325 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2327 uploader_id = uploader_url = None
2331 if not playlist_title:
2333 # Some playlist URLs don't actually serve a playlist (e.g.
2334 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2335 next(self._entries(page, playlist_id))
2336 except StopIteration:
2339 playlist = self.playlist_result(
2340 self._entries(page, playlist_id), playlist_id, playlist_title)
2342 'uploader': uploader,
2343 'uploader_id': uploader_id,
2344 'uploader_url': uploader_url,
2347 return has_videos, playlist
2349 def _check_download_just_video(self, url, playlist_id):
2350 # Check if it's a video-specific URL
2351 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2352 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2353 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2354 'video id', default=None)
2356 if self._downloader.params.get('noplaylist'):
2357 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2358 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2360 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2361 return video_id, None
2364 def _real_extract(self, url):
2365 # Extract playlist id
2366 mobj = re.match(self._VALID_URL, url)
2368 raise ExtractorError('Invalid URL: %s' % url)
2369 playlist_id = mobj.group(1) or mobj.group(2)
2371 video_id, video = self._check_download_just_video(url, playlist_id)
2375 if playlist_id.startswith(('RD', 'UL', 'PU')):
2376 # Mixes require a custom extraction process
2377 return self._extract_mix(playlist_id)
2379 has_videos, playlist = self._extract_playlist(playlist_id)
2380 if has_videos or not video_id:
2383 # Some playlist URLs don't actually serve a playlist (see
2384 # https://github.com/rg3/youtube-dl/issues/10537).
2385 # Fallback to plain video extraction if there is a video id
2386 # along with playlist id.
2387 return self.url_result(video_id, 'Youtube', video_id=video_id)
2390 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2391 IE_DESC = 'YouTube.com channels'
2392 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2393 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2394 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2395 IE_NAME = 'youtube:channel'
2397 'note': 'paginated channel',
2398 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2399 'playlist_mincount': 91,
2401 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2402 'title': 'Uploads from lex will',
2405 'note': 'Age restricted channel',
2406 # from https://www.youtube.com/user/DeusExOfficial
2407 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2408 'playlist_mincount': 64,
2410 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2411 'title': 'Uploads from Deus Ex',
2416 def suitable(cls, url):
2417 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2418 else super(YoutubeChannelIE, cls).suitable(url))
2420 def _build_template_url(self, url, channel_id):
2421 return self._TEMPLATE_URL % channel_id
2423 def _real_extract(self, url):
2424 channel_id = self._match_id(url)
2426 url = self._build_template_url(url, channel_id)
2428 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2429 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2430 # otherwise fallback on channel by page extraction
2431 channel_page = self._download_webpage(
2432 url + '?view=57', channel_id,
2433 'Downloading channel page', fatal=False)
2434 if channel_page is False:
2435 channel_playlist_id = False
2437 channel_playlist_id = self._html_search_meta(
2438 'channelId', channel_page, 'channel id', default=None)
2439 if not channel_playlist_id:
2440 channel_url = self._html_search_meta(
2441 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2442 channel_page, 'channel url', default=None)
2444 channel_playlist_id = self._search_regex(
2445 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2446 channel_url, 'channel id', default=None)
2447 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2448 playlist_id = 'UU' + channel_playlist_id[2:]
2449 return self.url_result(
2450 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2452 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2453 autogenerated = re.search(r'''(?x)
2455 channel-header-autogenerated-label|
2456 yt-channel-title-autogenerated
2457 )[^"]*"''', channel_page) is not None
2460 # The videos are contained in a single page
2461 # the ajax pages can't be used, they are empty
2464 video_id, 'Youtube', video_id=video_id,
2465 video_title=video_title)
2466 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2467 return self.playlist_result(entries, channel_id)
2470 next(self._entries(channel_page, channel_id))
2471 except StopIteration:
2472 alert_message = self._html_search_regex(
2473 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2474 channel_page, 'alert', default=None, group='alert')
2476 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2478 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2481 class YoutubeUserIE(YoutubeChannelIE):
2482 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2483 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2484 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2485 IE_NAME = 'youtube:user'
2488 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2489 'playlist_mincount': 320,
2491 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2492 'title': 'Uploads from The Linux Foundation',
2495 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2496 # but not https://www.youtube.com/user/12minuteathlete/videos
2497 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2498 'playlist_mincount': 249,
2500 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2501 'title': 'Uploads from 12 Minute Athlete',
2504 'url': 'ytuser:phihag',
2505 'only_matching': True,
2507 'url': 'https://www.youtube.com/c/gametrailers',
2508 'only_matching': True,
2510 'url': 'https://www.youtube.com/gametrailers',
2511 'only_matching': True,
2513 # This channel is not available, geo restricted to JP
2514 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2515 'only_matching': True,
2519 def suitable(cls, url):
2520 # Don't return True if the url can be extracted with other youtube
2521 # extractor, the regex would is too permissive and it would match.
2522 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2523 if any(ie.suitable(url) for ie in other_yt_ies):
2526 return super(YoutubeUserIE, cls).suitable(url)
2528 def _build_template_url(self, url, channel_id):
2529 mobj = re.match(self._VALID_URL, url)
2530 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2533 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2534 IE_DESC = 'YouTube.com live streams'
2535 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2536 IE_NAME = 'youtube:live'
2539 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2541 'id': 'a48o2S1cPoo',
2543 'title': 'The Young Turks - Live Main Show',
2544 'uploader': 'The Young Turks',
2545 'uploader_id': 'TheYoungTurks',
2546 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2547 'upload_date': '20150715',
2548 'license': 'Standard YouTube License',
2549 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2550 'categories': ['News & Politics'],
2551 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2553 'dislike_count': int,
2556 'skip_download': True,
2559 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2560 'only_matching': True,
2562 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2563 'only_matching': True,
2565 'url': 'https://www.youtube.com/TheYoungTurks/live',
2566 'only_matching': True,
2569 def _real_extract(self, url):
2570 mobj = re.match(self._VALID_URL, url)
2571 channel_id = mobj.group('id')
2572 base_url = mobj.group('base_url')
2573 webpage = self._download_webpage(url, channel_id, fatal=False)
2575 page_type = self._og_search_property(
2576 'type', webpage, 'page type', default='')
2577 video_id = self._html_search_meta(
2578 'videoId', webpage, 'video id', default=None)
2579 if page_type.startswith('video') and video_id and re.match(
2580 r'^[0-9A-Za-z_-]{11}$', video_id):
2581 return self.url_result(video_id, YoutubeIE.ie_key())
2582 return self.url_result(base_url)
2585 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2586 IE_DESC = 'YouTube.com user/channel playlists'
2587 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2588 IE_NAME = 'youtube:playlists'
2591 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2592 'playlist_mincount': 4,
2594 'id': 'ThirstForScience',
2595 'title': 'Thirst for Science',
2598 # with "Load more" button
2599 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2600 'playlist_mincount': 70,
2603 'title': 'Игорь Клейнер',
2606 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2607 'playlist_mincount': 17,
2609 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2610 'title': 'Chem Player',
2615 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2616 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2619 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2620 IE_DESC = 'YouTube.com searches'
2621 # there doesn't appear to be a real limit, for example if you search for
2622 # 'python' you get more than 8.000.000 results
2623 _MAX_RESULTS = float('inf')
2624 IE_NAME = 'youtube:search'
2625 _SEARCH_KEY = 'ytsearch'
2626 _EXTRA_QUERY_ARGS = {}
2629 def _get_n_results(self, query, n):
2630 """Get a specified number of results for a query"""
2636 'search_query': query.encode('utf-8'),
2638 url_query.update(self._EXTRA_QUERY_ARGS)
2639 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2641 for pagenum in itertools.count(1):
2642 data = self._download_json(
2643 result_url, video_id='query "%s"' % query,
2644 note='Downloading page %s' % pagenum,
2645 errnote='Unable to download API page',
2646 query={'spf': 'navigate'})
2647 html_content = data[1]['body']['content']
2649 if 'class="search-message' in html_content:
2650 raise ExtractorError(
2651 '[youtube] No video results', expected=True)
2653 new_videos = list(self._process_page(html_content))
2654 videos += new_videos
2655 if not new_videos or len(videos) > limit:
2657 next_link = self._html_search_regex(
2658 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2659 html_content, 'next link', default=None)
2660 if next_link is None:
2662 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2666 return self.playlist_result(videos, query)
2669 class YoutubeSearchDateIE(YoutubeSearchIE):
2670 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2671 _SEARCH_KEY = 'ytsearchdate'
2672 IE_DESC = 'YouTube.com searches, newest videos first'
2673 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2676 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2677 IE_DESC = 'YouTube.com search URLs'
2678 IE_NAME = 'youtube:search_url'
2679 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2681 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2682 'playlist_mincount': 5,
2684 'title': 'youtube-dl test video',
2687 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2688 'only_matching': True,
2691 def _real_extract(self, url):
2692 mobj = re.match(self._VALID_URL, url)
2693 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2694 webpage = self._download_webpage(url, query)
2695 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2698 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2699 IE_DESC = 'YouTube.com (multi-season) shows'
2700 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2701 IE_NAME = 'youtube:show'
2703 'url': 'https://www.youtube.com/show/airdisasters',
2704 'playlist_mincount': 5,
2706 'id': 'airdisasters',
2707 'title': 'Air Disasters',
2711 def _real_extract(self, url):
2712 playlist_id = self._match_id(url)
2713 return super(YoutubeShowIE, self)._real_extract(
2714 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2717 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2719 Base class for feed extractors
2720 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2722 _LOGIN_REQUIRED = True
2726 return 'youtube:%s' % self._FEED_NAME
2728 def _real_initialize(self):
2731 def _entries(self, page):
2732 # The extraction process is the same as for playlists, but the regex
2733 # for the video ids doesn't contain an index
2735 more_widget_html = content_html = page
2736 for page_num in itertools.count(1):
2737 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2739 # 'recommended' feed has infinite 'load more' and each new portion spins
2740 # the same videos in (sometimes) slightly different order, so we'll check
2741 # for unicity and break when portion has no new videos
2742 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2748 for entry in self._ids_to_results(new_ids):
2751 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2755 more = self._download_json(
2756 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2757 'Downloading page #%s' % page_num,
2758 transform_source=uppercase_escape)
2759 content_html = more['content_html']
2760 more_widget_html = more['load_more_widget_html']
2762 def _real_extract(self, url):
2763 page = self._download_webpage(
2764 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2765 self._PLAYLIST_TITLE)
2766 return self.playlist_result(
2767 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2770 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2771 IE_NAME = 'youtube:watchlater'
2772 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2773 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2776 'url': 'https://www.youtube.com/playlist?list=WL',
2777 'only_matching': True,
2779 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2780 'only_matching': True,
2783 def _real_extract(self, url):
2784 _, video = self._check_download_just_video(url, 'WL')
2787 _, playlist = self._extract_playlist('WL')
2791 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2792 IE_NAME = 'youtube:favorites'
2793 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2794 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2795 _LOGIN_REQUIRED = True
2797 def _real_extract(self, url):
2798 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2799 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2800 return self.url_result(playlist_id, 'YoutubePlaylist')
2803 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2804 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2805 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2806 _FEED_NAME = 'recommended'
2807 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2810 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2811 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2812 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2813 _FEED_NAME = 'subscriptions'
2814 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2817 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2818 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2819 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2820 _FEED_NAME = 'history'
2821 _PLAYLIST_TITLE = 'Youtube History'
2824 class YoutubeTruncatedURLIE(InfoExtractor):
2825 IE_NAME = 'youtube:truncated_url'
2826 IE_DESC = False # Do not list
2827 _VALID_URL = r'''(?x)
2829 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2832 annotation_id=annotation_[^&]+|
2838 attribution_link\?a=[^&]+
2844 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2845 'only_matching': True,
2847 'url': 'https://www.youtube.com/watch?',
2848 'only_matching': True,
2850 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2851 'only_matching': True,
2853 'url': 'https://www.youtube.com/watch?feature=foo',
2854 'only_matching': True,
2856 'url': 'https://www.youtube.com/watch?hl=en-GB',
2857 'only_matching': True,
2859 'url': 'https://www.youtube.com/watch?t=2372',
2860 'only_matching': True,
2863 def _real_extract(self, url):
2864 raise ExtractorError(
2865 'Did you forget to quote the URL? Remember that & is a meta '
2866 'character in most shells, so you want to put the URL in quotes, '
2868 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2869 ' or simply youtube-dl BaW_jenozKc .',
2873 class YoutubeTruncatedIDIE(InfoExtractor):
2874 IE_NAME = 'youtube:truncated_id'
2875 IE_DESC = False # Do not list
2876 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2879 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2880 'only_matching': True,
2883 def _real_extract(self, url):
2884 video_id = self._match_id(url)
2885 raise ExtractorError(
2886 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),