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',
516 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
517 'note': 'Test VEVO video with age protection (#956)',
521 'upload_date': '20130703',
522 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
523 'alt_title': 'Tunnel Vision',
524 'description': 'md5:64249768eec3bc4276236606ea996373',
526 'uploader': 'justintimberlakeVEVO',
527 'uploader_id': 'justintimberlakeVEVO',
528 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
529 'license': 'Standard YouTube License',
530 'creator': 'Justin Timberlake',
535 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
536 'note': 'Embed-only video (#1746)',
540 'upload_date': '20120608',
541 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
542 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
543 'uploader': 'SET India',
544 'uploader_id': 'setindia',
545 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
546 'license': 'Standard YouTube License',
551 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
552 'note': 'Use the first video ID in the URL',
556 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
557 'uploader': 'Philipp Hagemeister',
558 'uploader_id': 'phihag',
559 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
560 'upload_date': '20121002',
561 'license': 'Standard YouTube License',
562 '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 .',
563 'categories': ['Science & Technology'],
564 'tags': ['youtube-dl'],
567 'dislike_count': int,
570 'skip_download': True,
574 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
575 'note': '256k DASH audio (format 141) via DASH manifest',
579 'upload_date': '20121002',
580 'uploader_id': '8KVIDEO',
581 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
583 'uploader': '8KVIDEO',
584 'license': 'Standard YouTube License',
585 'title': 'UHDTV TEST 8K VIDEO.mp4'
588 'youtube_include_dash_manifest': True,
591 'skip': 'format 141 not served anymore',
593 # DASH manifest with encrypted signature
595 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
599 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
600 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
602 'uploader': 'AfrojackVEVO',
603 'uploader_id': 'AfrojackVEVO',
604 'upload_date': '20131011',
605 'license': 'Standard YouTube License',
608 'youtube_include_dash_manifest': True,
609 'format': '141/bestaudio[ext=m4a]',
612 # JS player signature function name containing $
614 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
618 'title': 'Taylor Swift - Shake It Off',
619 'alt_title': 'Shake It Off',
620 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
622 'uploader': 'TaylorSwiftVEVO',
623 'uploader_id': 'TaylorSwiftVEVO',
624 'upload_date': '20140818',
625 'license': 'Standard YouTube License',
626 'creator': 'Taylor Swift',
629 'youtube_include_dash_manifest': True,
630 'format': '141/bestaudio[ext=m4a]',
635 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
640 'upload_date': '20100909',
641 'uploader': 'The Amazing Atheist',
642 'uploader_id': 'TheAmazingAtheist',
643 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
644 'license': 'Standard YouTube License',
645 'title': 'Burning Everyone\'s Koran',
646 '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',
649 # Normal age-gate video (No vevo, embed allowed)
651 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
655 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
656 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
658 'uploader': 'The Witcher',
659 'uploader_id': 'WitcherGame',
660 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
661 'upload_date': '20140605',
662 'license': 'Standard YouTube License',
666 # Age-gate video with encrypted signature
668 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
672 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
673 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
675 'uploader': 'LloydVEVO',
676 'uploader_id': 'LloydVEVO',
677 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
678 'upload_date': '20110629',
679 'license': 'Standard YouTube License',
683 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
684 # YouTube Red ad is not captured for creator
686 'url': '__2ABJjxzNo',
691 'upload_date': '20100430',
692 'uploader_id': 'deadmau5',
693 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
694 'creator': 'deadmau5',
695 'description': 'md5:12c56784b8032162bb936a5f76d55360',
696 'uploader': 'deadmau5',
697 'license': 'Standard YouTube License',
698 'title': 'Deadmau5 - Some Chords (HD)',
699 'alt_title': 'Some Chords',
701 'expected_warnings': [
702 'DASH manifest missing',
705 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
707 'url': 'lqQg6PlCWgI',
712 'upload_date': '20150827',
713 'uploader_id': 'olympic',
714 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
715 'license': 'Standard YouTube License',
716 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
717 'uploader': 'Olympic',
718 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
721 'skip_download': 'requires avconv',
726 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
730 'stretched_ratio': 16 / 9.,
732 'upload_date': '20110310',
733 'uploader_id': 'AllenMeow',
734 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
735 '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',
737 'license': 'Standard YouTube License',
738 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
741 # url_encoded_fmt_stream_map is empty string
743 'url': 'qEJwOuvDf7I',
747 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
749 'upload_date': '20150404',
750 'uploader_id': 'spbelect',
751 'uploader': 'Наблюдатели Петербурга',
754 'skip_download': 'requires avconv',
756 'skip': 'This live event has ended.',
758 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
760 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
764 'title': 'md5:7b81415841e02ecd4313668cde88737a',
765 'description': 'md5:116377fd2963b81ec4ce64b542173306',
767 'upload_date': '20150625',
768 'uploader_id': 'dorappi2000',
769 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
770 'uploader': 'dorappi2000',
771 'license': 'Standard YouTube License',
772 'formats': 'mincount:32',
775 # DASH manifest with segment_list
777 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
778 'md5': '8ce563a1d667b599d21064e982ab9e31',
782 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
783 'uploader': 'Airtek',
784 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
785 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
786 'license': 'Standard YouTube License',
787 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
790 'youtube_include_dash_manifest': True,
791 'format': '135', # bestvideo
793 'skip': 'This live event has ended.',
796 # Multifeed videos (multiple cameras), URL is for Main Camera
797 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
800 'title': 'teamPGP: Rocket League Noob Stream',
801 'description': 'md5:dc7872fb300e143831327f1bae3af010',
807 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
808 'description': 'md5:dc7872fb300e143831327f1bae3af010',
810 'upload_date': '20150721',
811 'uploader': 'Beer Games Beer',
812 'uploader_id': 'beergamesbeer',
813 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
814 'license': 'Standard YouTube License',
820 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
821 'description': 'md5:dc7872fb300e143831327f1bae3af010',
823 'upload_date': '20150721',
824 'uploader': 'Beer Games Beer',
825 'uploader_id': 'beergamesbeer',
826 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
827 'license': 'Standard YouTube License',
833 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
834 'description': 'md5:dc7872fb300e143831327f1bae3af010',
836 'upload_date': '20150721',
837 'uploader': 'Beer Games Beer',
838 'uploader_id': 'beergamesbeer',
839 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
840 'license': 'Standard YouTube License',
846 'title': 'teamPGP: Rocket League Noob Stream (zim)',
847 'description': 'md5:dc7872fb300e143831327f1bae3af010',
849 'upload_date': '20150721',
850 'uploader': 'Beer Games Beer',
851 'uploader_id': 'beergamesbeer',
852 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
853 'license': 'Standard YouTube License',
857 'skip_download': True,
861 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
862 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
865 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
868 'skip': 'Not multifeed anymore',
871 'url': 'https://vid.plus/FlRa-iH7PGw',
872 'only_matching': True,
875 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
876 'only_matching': True,
879 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
880 # Also tests cut-off URL expansion in video description (see
881 # https://github.com/rg3/youtube-dl/issues/1892,
882 # https://github.com/rg3/youtube-dl/issues/8164)
883 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
887 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
888 'alt_title': 'Dark Walk',
889 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
891 'upload_date': '20151119',
892 'uploader_id': 'IronSoulElf',
893 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
894 'uploader': 'IronSoulElf',
895 'license': 'Standard YouTube License',
896 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
899 'skip_download': True,
903 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
904 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
905 'only_matching': True,
908 # Video with yt:stretch=17:0
909 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
913 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
914 'description': 'md5:ee18a25c350637c8faff806845bddee9',
915 'upload_date': '20151107',
916 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
917 'uploader': 'CH GAMER DROID',
920 'skip_download': True,
922 'skip': 'This video does not exist.',
925 # Video licensed under Creative Commons
926 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
930 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
931 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
933 'upload_date': '20150127',
934 'uploader_id': 'BerkmanCenter',
935 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
936 'uploader': 'The Berkman Klein Center for Internet & Society',
937 'license': 'Creative Commons Attribution license (reuse allowed)',
940 'skip_download': True,
944 # Channel-like uploader_url
945 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
949 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
950 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
952 'upload_date': '20151119',
953 'uploader': 'Bernie 2016',
954 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
955 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
956 'license': 'Creative Commons Attribution license (reuse allowed)',
959 'skip_download': True,
963 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
964 'only_matching': True,
967 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
968 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
969 'only_matching': True,
972 # Rental video preview
973 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
977 'title': 'Piku - Trailer',
978 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
979 'upload_date': '20150811',
980 'uploader': 'FlixMatrix',
981 'uploader_id': 'FlixMatrixKaravan',
982 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
983 'license': 'Standard YouTube License',
986 'skip_download': True,
990 # YouTube Red video with episode data
991 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
995 'title': 'Isolation - Mind Field (Ep 1)',
996 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
998 'upload_date': '20170118',
999 'uploader': 'Vsauce',
1000 'uploader_id': 'Vsauce',
1001 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1002 'license': 'Standard YouTube License',
1003 'series': 'Mind Field',
1005 'episode_number': 1,
1008 'skip_download': True,
1010 'expected_warnings': [
1011 'Skipping DASH manifest',
1015 # The following content has been identified by the YouTube community
1016 # as inappropriate or offensive to some audiences.
1017 'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1019 'id': '6SJNVb0GnPI',
1021 'title': 'Race Differences in Intelligence',
1022 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1024 'upload_date': '20140124',
1025 'uploader': 'New Century Foundation',
1026 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1027 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1028 'license': 'Standard YouTube License',
1032 'skip_download': True,
1037 'url': '1t24XAntNCY',
1038 'only_matching': True,
1041 # geo restricted to JP
1042 'url': 'sJL6WA-aGkQ',
1043 'only_matching': True,
1046 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1047 'only_matching': True,
1051 def __init__(self, *args, **kwargs):
1052 super(YoutubeIE, self).__init__(*args, **kwargs)
1053 self._player_cache = {}
1055 def report_video_info_webpage_download(self, video_id):
1056 """Report attempt to download video info webpage."""
1057 self.to_screen('%s: Downloading video info webpage' % video_id)
1059 def report_information_extraction(self, video_id):
1060 """Report attempt to extract video information."""
1061 self.to_screen('%s: Extracting video information' % video_id)
1063 def report_unavailable_format(self, video_id, format):
1064 """Report extracted video URL."""
1065 self.to_screen('%s: Format %s not available' % (video_id, format))
1067 def report_rtmp_download(self):
1068 """Indicate the download will use the RTMP protocol."""
1069 self.to_screen('RTMP download detected')
1071 def _signature_cache_id(self, example_sig):
1072 """ Return a string representation of a signature """
1073 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1075 def _extract_signature_function(self, video_id, player_url, example_sig):
1077 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1080 raise ExtractorError('Cannot identify player %r' % player_url)
1081 player_type = id_m.group('ext')
1082 player_id = id_m.group('id')
1084 # Read from filesystem cache
1085 func_id = '%s_%s_%s' % (
1086 player_type, player_id, self._signature_cache_id(example_sig))
1087 assert os.path.basename(func_id) == func_id
1089 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1090 if cache_spec is not None:
1091 return lambda s: ''.join(s[i] for i in cache_spec)
1094 'Downloading player %s' % player_url
1095 if self._downloader.params.get('verbose') else
1096 'Downloading %s player %s' % (player_type, player_id)
1098 if player_type == 'js':
1099 code = self._download_webpage(
1100 player_url, video_id,
1102 errnote='Download of %s failed' % player_url)
1103 res = self._parse_sig_js(code)
1104 elif player_type == 'swf':
1105 urlh = self._request_webpage(
1106 player_url, video_id,
1108 errnote='Download of %s failed' % player_url)
1110 res = self._parse_sig_swf(code)
1112 assert False, 'Invalid player type %r' % player_type
1114 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1115 cache_res = res(test_string)
1116 cache_spec = [ord(c) for c in cache_res]
1118 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1121 def _print_sig_code(self, func, example_sig):
1122 def gen_sig_code(idxs):
1123 def _genslice(start, end, step):
1124 starts = '' if start == 0 else str(start)
1125 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1126 steps = '' if step == 1 else (':%d' % step)
1127 return 's[%s%s%s]' % (starts, ends, steps)
1130 # Quelch pyflakes warnings - start will be set when step is set
1131 start = '(Never used)'
1132 for i, prev in zip(idxs[1:], idxs[:-1]):
1133 if step is not None:
1134 if i - prev == step:
1136 yield _genslice(start, prev, step)
1139 if i - prev in [-1, 1]:
1144 yield 's[%d]' % prev
1148 yield _genslice(start, i, step)
1150 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1151 cache_res = func(test_string)
1152 cache_spec = [ord(c) for c in cache_res]
1153 expr_code = ' + '.join(gen_sig_code(cache_spec))
1154 signature_id_tuple = '(%s)' % (
1155 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1156 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1157 ' return %s\n') % (signature_id_tuple, expr_code)
1158 self.to_screen('Extracted signature function:\n' + code)
1160 def _parse_sig_js(self, jscode):
1161 funcname = self._search_regex(
1162 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1163 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1164 jscode, 'Initial JS player signature function name', group='sig')
1166 jsi = JSInterpreter(jscode)
1167 initial_function = jsi.extract_function(funcname)
1168 return lambda s: initial_function([s])
1170 def _parse_sig_swf(self, file_contents):
1171 swfi = SWFInterpreter(file_contents)
1172 TARGET_CLASSNAME = 'SignatureDecipher'
1173 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1174 initial_function = swfi.extract_function(searched_class, 'decipher')
1175 return lambda s: initial_function([s])
1177 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1178 """Turn the encrypted s field into a working signature"""
1180 if player_url is None:
1181 raise ExtractorError('Cannot decrypt signature without player_url')
1183 if player_url.startswith('//'):
1184 player_url = 'https:' + player_url
1185 elif not re.match(r'https?://', player_url):
1186 player_url = compat_urlparse.urljoin(
1187 'https://www.youtube.com', player_url)
1189 player_id = (player_url, self._signature_cache_id(s))
1190 if player_id not in self._player_cache:
1191 func = self._extract_signature_function(
1192 video_id, player_url, s
1194 self._player_cache[player_id] = func
1195 func = self._player_cache[player_id]
1196 if self._downloader.params.get('youtube_print_sig_code'):
1197 self._print_sig_code(func, s)
1199 except Exception as e:
1200 tb = traceback.format_exc()
1201 raise ExtractorError(
1202 'Signature extraction failed: ' + tb, cause=e)
1204 def _get_subtitles(self, video_id, webpage):
1206 subs_doc = self._download_xml(
1207 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1208 video_id, note=False)
1209 except ExtractorError as err:
1210 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1214 for track in subs_doc.findall('track'):
1215 lang = track.attrib['lang_code']
1216 if lang in sub_lang_list:
1219 for ext in self._SUBTITLE_FORMATS:
1220 params = compat_urllib_parse_urlencode({
1224 'name': track.attrib['name'].encode('utf-8'),
1226 sub_formats.append({
1227 'url': 'https://www.youtube.com/api/timedtext?' + params,
1230 sub_lang_list[lang] = sub_formats
1231 if not sub_lang_list:
1232 self._downloader.report_warning('video doesn\'t have subtitles')
1234 return sub_lang_list
1236 def _get_ytplayer_config(self, video_id, webpage):
1238 # User data may contain arbitrary character sequences that may affect
1239 # JSON extraction with regex, e.g. when '};' is contained the second
1240 # regex won't capture the whole JSON. Yet working around by trying more
1241 # concrete regex first keeping in mind proper quoted string handling
1242 # to be implemented in future that will replace this workaround (see
1243 # https://github.com/rg3/youtube-dl/issues/7468,
1244 # https://github.com/rg3/youtube-dl/pull/7599)
1245 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1246 r';ytplayer\.config\s*=\s*({.+?});',
1248 config = self._search_regex(
1249 patterns, webpage, 'ytplayer.config', default=None)
1251 return self._parse_json(
1252 uppercase_escape(config), video_id, fatal=False)
1254 def _get_automatic_captions(self, video_id, webpage):
1255 """We need the webpage for getting the captions url, pass it as an
1256 argument to speed up the process."""
1257 self.to_screen('%s: Looking for automatic captions' % video_id)
1258 player_config = self._get_ytplayer_config(video_id, webpage)
1259 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1260 if not player_config:
1261 self._downloader.report_warning(err_msg)
1264 args = player_config['args']
1265 caption_url = args.get('ttsurl')
1267 timestamp = args['timestamp']
1268 # We get the available subtitles
1269 list_params = compat_urllib_parse_urlencode({
1274 list_url = caption_url + '&' + list_params
1275 caption_list = self._download_xml(list_url, video_id)
1276 original_lang_node = caption_list.find('track')
1277 if original_lang_node is None:
1278 self._downloader.report_warning('Video doesn\'t have automatic captions')
1280 original_lang = original_lang_node.attrib['lang_code']
1281 caption_kind = original_lang_node.attrib.get('kind', '')
1284 for lang_node in caption_list.findall('target'):
1285 sub_lang = lang_node.attrib['lang_code']
1287 for ext in self._SUBTITLE_FORMATS:
1288 params = compat_urllib_parse_urlencode({
1289 'lang': original_lang,
1293 'kind': caption_kind,
1295 sub_formats.append({
1296 'url': caption_url + '&' + params,
1299 sub_lang_list[sub_lang] = sub_formats
1300 return sub_lang_list
1302 def make_captions(sub_url, sub_langs):
1303 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1304 caption_qs = compat_parse_qs(parsed_sub_url.query)
1306 for sub_lang in sub_langs:
1308 for ext in self._SUBTITLE_FORMATS:
1310 'tlang': [sub_lang],
1313 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1314 query=compat_urllib_parse_urlencode(caption_qs, True)))
1315 sub_formats.append({
1319 captions[sub_lang] = sub_formats
1322 # New captions format as of 22.06.2017
1323 player_response = args.get('player_response')
1324 if player_response and isinstance(player_response, compat_str):
1325 player_response = self._parse_json(
1326 player_response, video_id, fatal=False)
1328 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1329 base_url = renderer['captionTracks'][0]['baseUrl']
1331 for lang in renderer['translationLanguages']:
1332 lang_code = lang.get('languageCode')
1334 sub_lang_list.append(lang_code)
1335 return make_captions(base_url, sub_lang_list)
1337 # Some videos don't provide ttsurl but rather caption_tracks and
1338 # caption_translation_languages (e.g. 20LmZk1hakA)
1339 # Does not used anymore as of 22.06.2017
1340 caption_tracks = args['caption_tracks']
1341 caption_translation_languages = args['caption_translation_languages']
1342 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1344 for lang in caption_translation_languages.split(','):
1345 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1346 sub_lang = lang_qs.get('lc', [None])[0]
1348 sub_lang_list.append(sub_lang)
1349 return make_captions(caption_url, sub_lang_list)
1350 # An extractor error can be raise by the download process if there are
1351 # no automatic captions but there are subtitles
1352 except (KeyError, IndexError, ExtractorError):
1353 self._downloader.report_warning(err_msg)
1356 def _mark_watched(self, video_id, video_info):
1357 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1358 if not playback_url:
1360 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1361 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1363 # cpn generation algorithm is reverse engineered from base.js.
1364 # In fact it works even with dummy cpn.
1365 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1366 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1372 playback_url = compat_urlparse.urlunparse(
1373 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1375 self._download_webpage(
1376 playback_url, video_id, 'Marking watched',
1377 'Unable to mark watched', fatal=False)
1380 def _extract_urls(webpage):
1381 # Embedded YouTube player
1383 unescapeHTML(mobj.group('url'))
1384 for mobj in re.finditer(r'''(?x)
1394 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1395 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1398 # lazyYT YouTube embed
1399 entries.extend(list(map(
1401 re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1403 # Wordpress "YouTube Video Importer" plugin
1404 matches = re.findall(r'''(?x)<div[^>]+
1405 class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1406 data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1407 entries.extend(m[-1] for m in matches)
1412 def _extract_url(webpage):
1413 urls = YoutubeIE._extract_urls(webpage)
1414 return urls[0] if urls else None
1417 def extract_id(cls, url):
1418 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1420 raise ExtractorError('Invalid URL: %s' % url)
1421 video_id = mobj.group(2)
1424 def _extract_annotations(self, video_id):
1425 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1426 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1429 def _extract_chapters(description, duration):
1432 chapter_lines = re.findall(
1433 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1435 if not chapter_lines:
1438 for next_num, (chapter_line, time_point) in enumerate(
1439 chapter_lines, start=1):
1440 start_time = parse_duration(time_point)
1441 if start_time is None:
1443 if start_time > duration:
1445 end_time = (duration if next_num == len(chapter_lines)
1446 else parse_duration(chapter_lines[next_num][1]))
1447 if end_time is None:
1449 if end_time > duration:
1451 if start_time > end_time:
1453 chapter_title = re.sub(
1454 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1455 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1457 'start_time': start_time,
1458 'end_time': end_time,
1459 'title': chapter_title,
1463 def _real_extract(self, url):
1464 url, smuggled_data = unsmuggle_url(url, {})
1467 'http' if self._downloader.params.get('prefer_insecure', False)
1472 parsed_url = compat_urllib_parse_urlparse(url)
1473 for component in [parsed_url.fragment, parsed_url.query]:
1474 query = compat_parse_qs(component)
1475 if start_time is None and 't' in query:
1476 start_time = parse_duration(query['t'][0])
1477 if start_time is None and 'start' in query:
1478 start_time = parse_duration(query['start'][0])
1479 if end_time is None and 'end' in query:
1480 end_time = parse_duration(query['end'][0])
1482 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1483 mobj = re.search(self._NEXT_URL_RE, url)
1485 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1486 video_id = self.extract_id(url)
1489 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1490 video_webpage = self._download_webpage(url, video_id)
1492 # Attempt to extract SWF player URL
1493 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1494 if mobj is not None:
1495 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1501 def add_dash_mpd(video_info):
1502 dash_mpd = video_info.get('dashmpd')
1503 if dash_mpd and dash_mpd[0] not in dash_mpds:
1504 dash_mpds.append(dash_mpd[0])
1509 def extract_view_count(v_info):
1510 return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1513 embed_webpage = None
1514 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1516 # We simulate the access to the video from www.youtube.com/v/{video_id}
1517 # this can be viewed without login into Youtube
1518 url = proto + '://www.youtube.com/embed/%s' % video_id
1519 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1520 data = compat_urllib_parse_urlencode({
1521 'video_id': video_id,
1522 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1523 'sts': self._search_regex(
1524 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1526 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1527 video_info_webpage = self._download_webpage(
1528 video_info_url, video_id,
1529 note='Refetching age-gated info webpage',
1530 errnote='unable to download video info webpage')
1531 video_info = compat_parse_qs(video_info_webpage)
1532 add_dash_mpd(video_info)
1537 # Try looking directly into the video webpage
1538 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1540 args = ytplayer_config['args']
1541 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1542 # Convert to the same format returned by compat_parse_qs
1543 video_info = dict((k, [v]) for k, v in args.items())
1544 add_dash_mpd(video_info)
1545 # Rental video is not rented but preview is available (e.g.
1546 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1547 # https://github.com/rg3/youtube-dl/issues/10532)
1548 if not video_info and args.get('ypc_vid'):
1549 return self.url_result(
1550 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1551 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1553 sts = ytplayer_config.get('sts')
1554 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1555 # We also try looking in get_video_info since it may contain different dashmpd
1556 # URL that points to a DASH manifest with possibly different itag set (some itags
1557 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1558 # manifest pointed by get_video_info's dashmpd).
1559 # The general idea is to take a union of itags of both DASH manifests (for example
1560 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1561 self.report_video_info_webpage_download(video_id)
1562 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1564 'video_id': video_id,
1574 video_info_webpage = self._download_webpage(
1575 '%s://www.youtube.com/get_video_info' % proto,
1576 video_id, note=False,
1577 errnote='unable to download video info webpage',
1578 fatal=False, query=query)
1579 if not video_info_webpage:
1581 get_video_info = compat_parse_qs(video_info_webpage)
1582 add_dash_mpd(get_video_info)
1583 if view_count is None:
1584 view_count = extract_view_count(get_video_info)
1586 video_info = get_video_info
1587 if 'token' in get_video_info:
1588 # Different get_video_info requests may report different results, e.g.
1589 # some may report video unavailability, but some may serve it without
1590 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1591 # the original webpage as well as el=info and el=embedded get_video_info
1592 # requests report video unavailability due to geo restriction while
1593 # el=detailpage succeeds and returns valid data). This is probably
1594 # due to YouTube measures against IP ranges of hosting providers.
1595 # Working around by preferring the first succeeded video_info containing
1596 # the token if no such video_info yet was found.
1597 if 'token' not in video_info:
1598 video_info = get_video_info
1601 def extract_unavailable_message():
1602 return self._html_search_regex(
1603 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1604 video_webpage, 'unavailable message', default=None)
1606 if 'token' not in video_info:
1607 if 'reason' in video_info:
1608 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1609 regions_allowed = self._html_search_meta(
1610 'regionsAllowed', video_webpage, default=None)
1611 countries = regions_allowed.split(',') if regions_allowed else None
1612 self.raise_geo_restricted(
1613 msg=video_info['reason'][0], countries=countries)
1614 reason = video_info['reason'][0]
1615 if 'Invalid parameters' in reason:
1616 unavailable_message = extract_unavailable_message()
1617 if unavailable_message:
1618 reason = unavailable_message
1619 raise ExtractorError(
1620 'YouTube said: %s' % reason,
1621 expected=True, video_id=video_id)
1623 raise ExtractorError(
1624 '"token" parameter not in video info for unknown reason',
1628 if 'title' in video_info:
1629 video_title = video_info['title'][0]
1631 self._downloader.report_warning('Unable to extract video title')
1635 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1636 if video_description:
1639 redir_url = compat_urlparse.urljoin(url, m.group(1))
1640 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1641 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1642 qs = compat_parse_qs(parsed_redir_url.query)
1648 description_original = video_description = re.sub(r'''(?x)
1650 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1651 (?:title|href)="([^"]+)"\s+
1652 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1656 ''', replace_url, video_description)
1657 video_description = clean_html(video_description)
1659 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1661 video_description = unescapeHTML(fd_mobj.group(1))
1663 video_description = ''
1665 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1666 if not self._downloader.params.get('noplaylist'):
1669 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1670 for feed in multifeed_metadata_list.split(','):
1671 # Unquote should take place before split on comma (,) since textual
1672 # fields may contain comma as well (see
1673 # https://github.com/rg3/youtube-dl/issues/8536)
1674 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1676 '_type': 'url_transparent',
1677 'ie_key': 'Youtube',
1679 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1680 {'force_singlefeed': True}),
1681 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1683 feed_ids.append(feed_data['id'][0])
1685 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1686 % (', '.join(feed_ids), video_id))
1687 return self.playlist_result(entries, video_id, video_title, video_description)
1688 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1690 if view_count is None:
1691 view_count = extract_view_count(video_info)
1693 # Check for "rental" videos
1694 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1695 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1697 # Start extracting information
1698 self.report_information_extraction(video_id)
1701 video_uploader = try_get(video_info, lambda x: x['author'][0], compat_str)
1703 video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
1705 self._downloader.report_warning('unable to extract uploader name')
1708 video_uploader_id = None
1709 video_uploader_url = None
1711 r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1713 if mobj is not None:
1714 video_uploader_id = mobj.group('uploader_id')
1715 video_uploader_url = mobj.group('uploader_url')
1717 self._downloader.report_warning('unable to extract uploader nickname')
1720 # We try first to get a high quality image:
1721 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1722 video_webpage, re.DOTALL)
1723 if m_thumb is not None:
1724 video_thumbnail = m_thumb.group(1)
1725 elif 'thumbnail_url' not in video_info:
1726 self._downloader.report_warning('unable to extract video thumbnail')
1727 video_thumbnail = None
1728 else: # don't panic if we can't find it
1729 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1732 upload_date = self._html_search_meta(
1733 'datePublished', video_webpage, 'upload date', default=None)
1735 upload_date = self._search_regex(
1736 [r'(?s)id="eow-date.*?>(.*?)</span>',
1737 r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
1738 video_webpage, 'upload date', default=None)
1739 upload_date = unified_strdate(upload_date)
1741 video_license = self._html_search_regex(
1742 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1743 video_webpage, 'license', default=None)
1745 m_music = re.search(
1747 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1755 \bhref=["\']/red[^>]*>| # drop possible
1756 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1763 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1764 video_creator = clean_html(m_music.group('creator'))
1766 video_alt_title = video_creator = None
1768 m_episode = re.search(
1769 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1772 series = m_episode.group('series')
1773 season_number = int(m_episode.group('season'))
1774 episode_number = int(m_episode.group('episode'))
1776 series = season_number = episode_number = None
1778 m_cat_container = self._search_regex(
1779 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1780 video_webpage, 'categories', default=None)
1782 category = self._html_search_regex(
1783 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1785 video_categories = None if category is None else [category]
1787 video_categories = None
1790 unescapeHTML(m.group('content'))
1791 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1793 def _extract_count(count_name):
1794 return str_to_int(self._search_regex(
1795 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1796 % re.escape(count_name),
1797 video_webpage, count_name, default=None))
1799 like_count = _extract_count('like')
1800 dislike_count = _extract_count('dislike')
1803 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1804 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1806 video_duration = try_get(
1807 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1808 if not video_duration:
1809 video_duration = parse_duration(self._html_search_meta(
1810 'duration', video_webpage, 'video duration'))
1813 video_annotations = None
1814 if self._downloader.params.get('writeannotations', False):
1815 video_annotations = self._extract_annotations(video_id)
1817 chapters = self._extract_chapters(description_original, video_duration)
1819 def _extract_filesize(media_url):
1820 return int_or_none(self._search_regex(
1821 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1823 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1824 self.report_rtmp_download()
1826 'format_id': '_rtmp',
1828 'url': video_info['conn'][0],
1829 'player_url': player_url,
1831 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):
1832 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1833 if 'rtmpe%3Dyes' in encoded_url_map:
1834 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1836 fmt_list = video_info.get('fmt_list', [''])[0]
1838 for fmt in fmt_list.split(','):
1839 spec = fmt.split('/')
1841 width_height = spec[1].split('x')
1842 if len(width_height) == 2:
1843 formats_spec[spec[0]] = {
1844 'resolution': spec[1],
1845 'width': int_or_none(width_height[0]),
1846 'height': int_or_none(width_height[1]),
1848 q = qualities(['small', 'medium', 'hd720'])
1850 for url_data_str in encoded_url_map.split(','):
1851 url_data = compat_parse_qs(url_data_str)
1852 if 'itag' not in url_data or 'url' not in url_data:
1854 format_id = url_data['itag'][0]
1855 url = url_data['url'][0]
1857 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1858 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1859 jsplayer_url_json = self._search_regex(
1861 embed_webpage if age_gate else video_webpage,
1862 'JS player URL (1)', default=None)
1863 if not jsplayer_url_json and not age_gate:
1864 # We need the embed website after all
1865 if embed_webpage is None:
1866 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1867 embed_webpage = self._download_webpage(
1868 embed_url, video_id, 'Downloading embed webpage')
1869 jsplayer_url_json = self._search_regex(
1870 ASSETS_RE, embed_webpage, 'JS player URL')
1872 player_url = json.loads(jsplayer_url_json)
1873 if player_url is None:
1874 player_url_json = self._search_regex(
1875 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1876 video_webpage, 'age gate player URL')
1877 player_url = json.loads(player_url_json)
1879 if 'sig' in url_data:
1880 url += '&signature=' + url_data['sig'][0]
1881 elif 's' in url_data:
1882 encrypted_sig = url_data['s'][0]
1884 if self._downloader.params.get('verbose'):
1885 if player_url is None:
1886 player_version = 'unknown'
1887 player_desc = 'unknown'
1889 if player_url.endswith('swf'):
1890 player_version = self._search_regex(
1891 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1892 'flash player', fatal=False)
1893 player_desc = 'flash player %s' % player_version
1895 player_version = self._search_regex(
1896 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1897 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1899 'html5 player', fatal=False)
1900 player_desc = 'html5 player %s' % player_version
1902 parts_sizes = self._signature_cache_id(encrypted_sig)
1903 self.to_screen('{%s} signature length %s, %s' %
1904 (format_id, parts_sizes, player_desc))
1906 signature = self._decrypt_signature(
1907 encrypted_sig, video_id, player_url, age_gate)
1908 url += '&signature=' + signature
1909 if 'ratebypass' not in url:
1910 url += '&ratebypass=yes'
1913 'format_id': format_id,
1915 'player_url': player_url,
1917 if format_id in self._formats:
1918 dct.update(self._formats[format_id])
1919 if format_id in formats_spec:
1920 dct.update(formats_spec[format_id])
1922 # Some itags are not included in DASH manifest thus corresponding formats will
1923 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1924 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1925 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1926 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1928 filesize = int_or_none(url_data.get(
1929 'clen', [None])[0]) or _extract_filesize(url)
1931 quality = url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0]
1934 'filesize': filesize,
1935 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1938 'fps': int_or_none(url_data.get('fps', [None])[0]),
1939 'format_note': quality,
1940 'quality': q(quality),
1942 for key, value in more_fields.items():
1945 type_ = url_data.get('type', [None])[0]
1947 type_split = type_.split(';')
1948 kind_ext = type_split[0].split('/')
1949 if len(kind_ext) == 2:
1951 dct['ext'] = mimetype2ext(type_split[0])
1952 if kind in ('audio', 'video'):
1954 for mobj in re.finditer(
1955 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1956 if mobj.group('key') == 'codecs':
1957 codecs = mobj.group('val')
1960 dct.update(parse_codecs(codecs))
1961 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
1962 dct['downloader_options'] = {
1963 # Youtube throttles chunks >~10M
1964 'http_chunk_size': 10485760,
1967 elif video_info.get('hlsvp'):
1968 manifest_url = video_info['hlsvp'][0]
1970 m3u8_formats = self._extract_m3u8_formats(
1971 manifest_url, video_id, 'mp4', fatal=False)
1972 for a_format in m3u8_formats:
1973 itag = self._search_regex(
1974 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1976 a_format['format_id'] = itag
1977 if itag in self._formats:
1978 dct = self._formats[itag].copy()
1979 dct.update(a_format)
1981 a_format['player_url'] = player_url
1982 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1983 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1984 formats.append(a_format)
1986 error_message = clean_html(video_info.get('reason', [None])[0])
1987 if not error_message:
1988 error_message = extract_unavailable_message()
1990 raise ExtractorError(error_message, expected=True)
1991 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1993 # Look for the DASH manifest
1994 if self._downloader.params.get('youtube_include_dash_manifest', True):
1995 dash_mpd_fatal = True
1996 for mpd_url in dash_mpds:
1999 def decrypt_sig(mobj):
2001 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2002 return '/signature/%s' % dec_s
2004 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2006 for df in self._extract_mpd_formats(
2007 mpd_url, video_id, fatal=dash_mpd_fatal,
2008 formats_dict=self._formats):
2009 if not df.get('filesize'):
2010 df['filesize'] = _extract_filesize(df['url'])
2011 # Do not overwrite DASH format found in some previous DASH manifest
2012 if df['format_id'] not in dash_formats:
2013 dash_formats[df['format_id']] = df
2014 # Additional DASH manifests may end up in HTTP Error 403 therefore
2015 # allow them to fail without bug report message if we already have
2016 # some DASH manifest succeeded. This is temporary workaround to reduce
2017 # burst of bug reports until we figure out the reason and whether it
2018 # can be fixed at all.
2019 dash_mpd_fatal = False
2020 except (ExtractorError, KeyError) as e:
2021 self.report_warning(
2022 'Skipping DASH manifest: %r' % e, video_id)
2024 # Remove the formats we found through non-DASH, they
2025 # contain less info and it can be wrong, because we use
2026 # fixed values (for example the resolution). See
2027 # https://github.com/rg3/youtube-dl/issues/5774 for an
2029 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2030 formats.extend(dash_formats.values())
2032 # Check for malformed aspect ratio
2033 stretched_m = re.search(
2034 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2037 w = float(stretched_m.group('w'))
2038 h = float(stretched_m.group('h'))
2039 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2040 # We will only process correct ratios.
2044 if f.get('vcodec') != 'none':
2045 f['stretched_ratio'] = ratio
2047 self._sort_formats(formats)
2049 self.mark_watched(video_id, video_info)
2053 'uploader': video_uploader,
2054 'uploader_id': video_uploader_id,
2055 'uploader_url': video_uploader_url,
2056 'upload_date': upload_date,
2057 'license': video_license,
2058 'creator': video_creator,
2059 'title': video_title,
2060 'alt_title': video_alt_title,
2061 'thumbnail': video_thumbnail,
2062 'description': video_description,
2063 'categories': video_categories,
2065 'subtitles': video_subtitles,
2066 'automatic_captions': automatic_captions,
2067 'duration': video_duration,
2068 'age_limit': 18 if age_gate else 0,
2069 'annotations': video_annotations,
2070 'chapters': chapters,
2071 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2072 'view_count': view_count,
2073 'like_count': like_count,
2074 'dislike_count': dislike_count,
2075 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
2078 'start_time': start_time,
2079 'end_time': end_time,
2081 'season_number': season_number,
2082 'episode_number': episode_number,
2086 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2087 IE_DESC = 'YouTube.com playlists'
2088 _VALID_URL = r"""(?x)(?:
2094 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2095 \? (?:.*?[&;])*? (?:p|a|list)=
2098 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2101 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2102 # Top tracks, they can also include dots
2108 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2109 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2110 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2111 IE_NAME = 'youtube:playlist'
2113 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2115 'title': 'ytdl test PL',
2116 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2118 'playlist_count': 3,
2120 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2122 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2123 'title': 'YDL_Empty_List',
2125 'playlist_count': 0,
2126 'skip': 'This playlist is private',
2128 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2129 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2131 'title': '29C3: Not my department',
2132 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2134 'playlist_count': 95,
2136 'note': 'issue #673',
2137 'url': 'PLBB231211A4F62143',
2139 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2140 'id': 'PLBB231211A4F62143',
2142 'playlist_mincount': 26,
2144 'note': 'Large playlist',
2145 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2147 'title': 'Uploads from Cauchemar',
2148 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2150 'playlist_mincount': 799,
2152 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2154 'title': 'YDL_safe_search',
2155 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2157 'playlist_count': 2,
2158 'skip': 'This playlist is private',
2161 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2162 'playlist_count': 4,
2165 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2168 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2169 'playlist_mincount': 485,
2171 'title': '2017 華語最新單曲 (2/24更新)',
2172 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2175 'note': 'Embedded SWF player',
2176 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2177 'playlist_count': 4,
2180 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2183 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2184 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2186 'title': 'Uploads from Interstellar Movie',
2187 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2189 'playlist_mincount': 21,
2191 # Playlist URL that does not actually serve a playlist
2192 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2194 'id': 'FqZTN594JQw',
2196 'title': "Smiley's People 01 detective, Adventure Series, Action",
2197 'uploader': 'STREEM',
2198 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2199 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2200 'upload_date': '20150526',
2201 'license': 'Standard YouTube License',
2202 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2203 'categories': ['People & Blogs'],
2206 'dislike_count': int,
2209 'skip_download': True,
2211 'add_ie': [YoutubeIE.ie_key()],
2213 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2215 'id': 'yeWKywCrFtk',
2217 'title': 'Small Scale Baler and Braiding Rugs',
2218 'uploader': 'Backus-Page House Museum',
2219 'uploader_id': 'backuspagemuseum',
2220 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2221 'upload_date': '20161008',
2222 'license': 'Standard YouTube License',
2223 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2224 'categories': ['Nonprofits & Activism'],
2227 'dislike_count': int,
2231 'skip_download': True,
2234 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2235 'only_matching': True,
2237 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2238 'only_matching': True,
2241 def _real_initialize(self):
2244 def _extract_mix(self, playlist_id):
2245 # The mixes are generated from a single video
2246 # the id of the playlist is just 'RD' + video_id
2248 last_id = playlist_id[-11:]
2249 for n in itertools.count(1):
2250 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2251 webpage = self._download_webpage(
2252 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2253 new_ids = orderedSet(re.findall(
2254 r'''(?xs)data-video-username=".*?".*?
2255 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2257 # Fetch new pages until all the videos are repeated, it seems that
2258 # there are always 51 unique videos.
2259 new_ids = [_id for _id in new_ids if _id not in ids]
2265 url_results = self._ids_to_results(ids)
2267 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2269 search_title('playlist-title') or
2270 search_title('title long-title') or
2271 search_title('title'))
2272 title = clean_html(title_span)
2274 return self.playlist_result(url_results, playlist_id, title)
2276 def _extract_playlist(self, playlist_id):
2277 url = self._TEMPLATE_URL % playlist_id
2278 page = self._download_webpage(url, playlist_id)
2280 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2281 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2282 match = match.strip()
2283 # Check if the playlist exists or is private
2284 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2286 reason = mobj.group('reason')
2287 message = 'This playlist %s' % reason
2288 if 'private' in reason:
2289 message += ', use --username or --netrc to access it'
2291 raise ExtractorError(message, expected=True)
2292 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2293 raise ExtractorError(
2294 'Invalid parameters. Maybe URL is incorrect.',
2296 elif re.match(r'[^<]*Choose your language[^<]*', match):
2299 self.report_warning('Youtube gives an alert message: ' + match)
2301 playlist_title = self._html_search_regex(
2302 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2303 page, 'title', default=None)
2305 _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2306 uploader = self._search_regex(
2307 r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2308 page, 'uploader', default=None)
2310 r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2313 uploader_id = mobj.group('uploader_id')
2314 uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2316 uploader_id = uploader_url = None
2320 if not playlist_title:
2322 # Some playlist URLs don't actually serve a playlist (e.g.
2323 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2324 next(self._entries(page, playlist_id))
2325 except StopIteration:
2328 playlist = self.playlist_result(
2329 self._entries(page, playlist_id), playlist_id, playlist_title)
2331 'uploader': uploader,
2332 'uploader_id': uploader_id,
2333 'uploader_url': uploader_url,
2336 return has_videos, playlist
2338 def _check_download_just_video(self, url, playlist_id):
2339 # Check if it's a video-specific URL
2340 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2341 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2342 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2343 'video id', default=None)
2345 if self._downloader.params.get('noplaylist'):
2346 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2347 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2349 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2350 return video_id, None
2353 def _real_extract(self, url):
2354 # Extract playlist id
2355 mobj = re.match(self._VALID_URL, url)
2357 raise ExtractorError('Invalid URL: %s' % url)
2358 playlist_id = mobj.group(1) or mobj.group(2)
2360 video_id, video = self._check_download_just_video(url, playlist_id)
2364 if playlist_id.startswith(('RD', 'UL', 'PU')):
2365 # Mixes require a custom extraction process
2366 return self._extract_mix(playlist_id)
2368 has_videos, playlist = self._extract_playlist(playlist_id)
2369 if has_videos or not video_id:
2372 # Some playlist URLs don't actually serve a playlist (see
2373 # https://github.com/rg3/youtube-dl/issues/10537).
2374 # Fallback to plain video extraction if there is a video id
2375 # along with playlist id.
2376 return self.url_result(video_id, 'Youtube', video_id=video_id)
2379 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2380 IE_DESC = 'YouTube.com channels'
2381 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2382 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2383 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2384 IE_NAME = 'youtube:channel'
2386 'note': 'paginated channel',
2387 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2388 'playlist_mincount': 91,
2390 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2391 'title': 'Uploads from lex will',
2394 'note': 'Age restricted channel',
2395 # from https://www.youtube.com/user/DeusExOfficial
2396 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2397 'playlist_mincount': 64,
2399 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2400 'title': 'Uploads from Deus Ex',
2405 def suitable(cls, url):
2406 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2407 else super(YoutubeChannelIE, cls).suitable(url))
2409 def _build_template_url(self, url, channel_id):
2410 return self._TEMPLATE_URL % channel_id
2412 def _real_extract(self, url):
2413 channel_id = self._match_id(url)
2415 url = self._build_template_url(url, channel_id)
2417 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2418 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2419 # otherwise fallback on channel by page extraction
2420 channel_page = self._download_webpage(
2421 url + '?view=57', channel_id,
2422 'Downloading channel page', fatal=False)
2423 if channel_page is False:
2424 channel_playlist_id = False
2426 channel_playlist_id = self._html_search_meta(
2427 'channelId', channel_page, 'channel id', default=None)
2428 if not channel_playlist_id:
2429 channel_url = self._html_search_meta(
2430 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2431 channel_page, 'channel url', default=None)
2433 channel_playlist_id = self._search_regex(
2434 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2435 channel_url, 'channel id', default=None)
2436 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2437 playlist_id = 'UU' + channel_playlist_id[2:]
2438 return self.url_result(
2439 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2441 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2442 autogenerated = re.search(r'''(?x)
2444 channel-header-autogenerated-label|
2445 yt-channel-title-autogenerated
2446 )[^"]*"''', channel_page) is not None
2449 # The videos are contained in a single page
2450 # the ajax pages can't be used, they are empty
2453 video_id, 'Youtube', video_id=video_id,
2454 video_title=video_title)
2455 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2456 return self.playlist_result(entries, channel_id)
2459 next(self._entries(channel_page, channel_id))
2460 except StopIteration:
2461 alert_message = self._html_search_regex(
2462 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2463 channel_page, 'alert', default=None, group='alert')
2465 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2467 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2470 class YoutubeUserIE(YoutubeChannelIE):
2471 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2472 _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_-]+)'
2473 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2474 IE_NAME = 'youtube:user'
2477 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2478 'playlist_mincount': 320,
2480 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2481 'title': 'Uploads from The Linux Foundation',
2484 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2485 # but not https://www.youtube.com/user/12minuteathlete/videos
2486 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2487 'playlist_mincount': 249,
2489 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2490 'title': 'Uploads from 12 Minute Athlete',
2493 'url': 'ytuser:phihag',
2494 'only_matching': True,
2496 'url': 'https://www.youtube.com/c/gametrailers',
2497 'only_matching': True,
2499 'url': 'https://www.youtube.com/gametrailers',
2500 'only_matching': True,
2502 # This channel is not available, geo restricted to JP
2503 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2504 'only_matching': True,
2508 def suitable(cls, url):
2509 # Don't return True if the url can be extracted with other youtube
2510 # extractor, the regex would is too permissive and it would match.
2511 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2512 if any(ie.suitable(url) for ie in other_yt_ies):
2515 return super(YoutubeUserIE, cls).suitable(url)
2517 def _build_template_url(self, url, channel_id):
2518 mobj = re.match(self._VALID_URL, url)
2519 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2522 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2523 IE_DESC = 'YouTube.com live streams'
2524 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2525 IE_NAME = 'youtube:live'
2528 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2530 'id': 'a48o2S1cPoo',
2532 'title': 'The Young Turks - Live Main Show',
2533 'uploader': 'The Young Turks',
2534 'uploader_id': 'TheYoungTurks',
2535 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2536 'upload_date': '20150715',
2537 'license': 'Standard YouTube License',
2538 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2539 'categories': ['News & Politics'],
2540 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2542 'dislike_count': int,
2545 'skip_download': True,
2548 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2549 'only_matching': True,
2551 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2552 'only_matching': True,
2554 'url': 'https://www.youtube.com/TheYoungTurks/live',
2555 'only_matching': True,
2558 def _real_extract(self, url):
2559 mobj = re.match(self._VALID_URL, url)
2560 channel_id = mobj.group('id')
2561 base_url = mobj.group('base_url')
2562 webpage = self._download_webpage(url, channel_id, fatal=False)
2564 page_type = self._og_search_property(
2565 'type', webpage, 'page type', default='')
2566 video_id = self._html_search_meta(
2567 'videoId', webpage, 'video id', default=None)
2568 if page_type.startswith('video') and video_id and re.match(
2569 r'^[0-9A-Za-z_-]{11}$', video_id):
2570 return self.url_result(video_id, YoutubeIE.ie_key())
2571 return self.url_result(base_url)
2574 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2575 IE_DESC = 'YouTube.com user/channel playlists'
2576 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2577 IE_NAME = 'youtube:playlists'
2580 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2581 'playlist_mincount': 4,
2583 'id': 'ThirstForScience',
2584 'title': 'Thirst for Science',
2587 # with "Load more" button
2588 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2589 'playlist_mincount': 70,
2592 'title': 'Игорь Клейнер',
2595 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2596 'playlist_mincount': 17,
2598 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2599 'title': 'Chem Player',
2604 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2605 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2608 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2609 IE_DESC = 'YouTube.com searches'
2610 # there doesn't appear to be a real limit, for example if you search for
2611 # 'python' you get more than 8.000.000 results
2612 _MAX_RESULTS = float('inf')
2613 IE_NAME = 'youtube:search'
2614 _SEARCH_KEY = 'ytsearch'
2615 _EXTRA_QUERY_ARGS = {}
2618 def _get_n_results(self, query, n):
2619 """Get a specified number of results for a query"""
2625 'search_query': query.encode('utf-8'),
2627 url_query.update(self._EXTRA_QUERY_ARGS)
2628 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2630 for pagenum in itertools.count(1):
2631 data = self._download_json(
2632 result_url, video_id='query "%s"' % query,
2633 note='Downloading page %s' % pagenum,
2634 errnote='Unable to download API page',
2635 query={'spf': 'navigate'})
2636 html_content = data[1]['body']['content']
2638 if 'class="search-message' in html_content:
2639 raise ExtractorError(
2640 '[youtube] No video results', expected=True)
2642 new_videos = list(self._process_page(html_content))
2643 videos += new_videos
2644 if not new_videos or len(videos) > limit:
2646 next_link = self._html_search_regex(
2647 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2648 html_content, 'next link', default=None)
2649 if next_link is None:
2651 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2655 return self.playlist_result(videos, query)
2658 class YoutubeSearchDateIE(YoutubeSearchIE):
2659 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2660 _SEARCH_KEY = 'ytsearchdate'
2661 IE_DESC = 'YouTube.com searches, newest videos first'
2662 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2665 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2666 IE_DESC = 'YouTube.com search URLs'
2667 IE_NAME = 'youtube:search_url'
2668 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2670 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2671 'playlist_mincount': 5,
2673 'title': 'youtube-dl test video',
2676 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2677 'only_matching': True,
2680 def _real_extract(self, url):
2681 mobj = re.match(self._VALID_URL, url)
2682 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2683 webpage = self._download_webpage(url, query)
2684 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2687 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2688 IE_DESC = 'YouTube.com (multi-season) shows'
2689 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2690 IE_NAME = 'youtube:show'
2692 'url': 'https://www.youtube.com/show/airdisasters',
2693 'playlist_mincount': 5,
2695 'id': 'airdisasters',
2696 'title': 'Air Disasters',
2700 def _real_extract(self, url):
2701 playlist_id = self._match_id(url)
2702 return super(YoutubeShowIE, self)._real_extract(
2703 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2706 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2708 Base class for feed extractors
2709 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2711 _LOGIN_REQUIRED = True
2715 return 'youtube:%s' % self._FEED_NAME
2717 def _real_initialize(self):
2720 def _entries(self, page):
2721 # The extraction process is the same as for playlists, but the regex
2722 # for the video ids doesn't contain an index
2724 more_widget_html = content_html = page
2725 for page_num in itertools.count(1):
2726 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2728 # 'recommended' feed has infinite 'load more' and each new portion spins
2729 # the same videos in (sometimes) slightly different order, so we'll check
2730 # for unicity and break when portion has no new videos
2731 new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
2737 for entry in self._ids_to_results(new_ids):
2740 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2744 more = self._download_json(
2745 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2746 'Downloading page #%s' % page_num,
2747 transform_source=uppercase_escape)
2748 content_html = more['content_html']
2749 more_widget_html = more['load_more_widget_html']
2751 def _real_extract(self, url):
2752 page = self._download_webpage(
2753 'https://www.youtube.com/feed/%s' % self._FEED_NAME,
2754 self._PLAYLIST_TITLE)
2755 return self.playlist_result(
2756 self._entries(page), playlist_title=self._PLAYLIST_TITLE)
2759 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2760 IE_NAME = 'youtube:watchlater'
2761 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2762 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2765 'url': 'https://www.youtube.com/playlist?list=WL',
2766 'only_matching': True,
2768 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2769 'only_matching': True,
2772 def _real_extract(self, url):
2773 _, video = self._check_download_just_video(url, 'WL')
2776 _, playlist = self._extract_playlist('WL')
2780 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2781 IE_NAME = 'youtube:favorites'
2782 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2783 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2784 _LOGIN_REQUIRED = True
2786 def _real_extract(self, url):
2787 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2788 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2789 return self.url_result(playlist_id, 'YoutubePlaylist')
2792 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2793 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2794 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2795 _FEED_NAME = 'recommended'
2796 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2799 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2800 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2801 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2802 _FEED_NAME = 'subscriptions'
2803 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2806 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2807 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2808 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2809 _FEED_NAME = 'history'
2810 _PLAYLIST_TITLE = 'Youtube History'
2813 class YoutubeTruncatedURLIE(InfoExtractor):
2814 IE_NAME = 'youtube:truncated_url'
2815 IE_DESC = False # Do not list
2816 _VALID_URL = r'''(?x)
2818 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2821 annotation_id=annotation_[^&]+|
2827 attribution_link\?a=[^&]+
2833 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2834 'only_matching': True,
2836 'url': 'https://www.youtube.com/watch?',
2837 'only_matching': True,
2839 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2840 'only_matching': True,
2842 'url': 'https://www.youtube.com/watch?feature=foo',
2843 'only_matching': True,
2845 'url': 'https://www.youtube.com/watch?hl=en-GB',
2846 'only_matching': True,
2848 'url': 'https://www.youtube.com/watch?t=2372',
2849 'only_matching': True,
2852 def _real_extract(self, url):
2853 raise ExtractorError(
2854 'Did you forget to quote the URL? Remember that & is a meta '
2855 'character in most shells, so you want to put the URL in quotes, '
2857 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2858 ' or simply youtube-dl BaW_jenozKc .',
2862 class YoutubeTruncatedIDIE(InfoExtractor):
2863 IE_NAME = 'youtube:truncated_id'
2864 IE_DESC = False # Do not list
2865 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2868 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2869 'only_matching': True,
2872 def _real_extract(self, url):
2873 video_id = self._match_id(url)
2874 raise ExtractorError(
2875 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),