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 (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
32 get_element_by_attribute,
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
57 _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
58 _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
59 _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
61 _NETRC_MACHINE = 'youtube'
62 # If True it will raise an error if no login info is provided
63 _LOGIN_REQUIRED = False
65 _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
67 def _set_language(self):
69 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
70 # YouTube sets the expire time to about two months
71 expire_time=time.time() + 2 * 30 * 24 * 3600)
73 def _ids_to_results(self, ids):
75 self.url_result(vid_id, 'Youtube', video_id=vid_id)
80 Attempt to log in to YouTube.
81 True is returned if successful or skipped.
82 False is returned if login failed.
84 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
86 (username, password) = self._get_login_info()
87 # No authentication to be performed
89 if self._LOGIN_REQUIRED:
90 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
93 login_page = self._download_webpage(
94 self._LOGIN_URL, None,
95 note='Downloading login page',
96 errnote='unable to fetch login page', fatal=False)
97 if login_page is False:
100 login_form = self._hidden_inputs(login_page)
102 def req(url, f_req, note, errnote):
103 data = login_form.copy()
106 'checkConnection': 'youtube',
107 'checkedDomains': 'youtube',
109 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
110 'f.req': json.dumps(f_req),
111 'flowName': 'GlifWebSignIn',
112 'flowEntry': 'ServiceLogin',
114 return self._download_json(
115 url, None, note=note, errnote=errnote,
116 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
118 data=urlencode_postdata(data), headers={
119 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
120 'Google-Accounts-XSRF': 1,
124 self._downloader.report_warning(message)
128 None, [], None, 'US', None, None, 2, False, True,
132 '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',
134 1, [None, None, []], None, None, None, True
139 lookup_results = req(
140 self._LOOKUP_URL, lookup_req,
141 'Looking up account info', 'Unable to look up account info')
143 if lookup_results is False:
146 user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
148 warn('Unable to extract user hash')
153 None, 1, None, [1, None, None, None, [password, None, True]],
155 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],
156 1, [None, None, []], None, None, None, True
159 challenge_results = req(
160 self._CHALLENGE_URL, challenge_req,
161 'Logging in', 'Unable to log in')
163 if challenge_results is False:
166 login_res = try_get(challenge_results, lambda x: x[0][5], list)
168 login_msg = try_get(login_res, lambda x: x[5], compat_str)
170 'Unable to login: %s' % 'Invalid password'
171 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
174 res = try_get(challenge_results, lambda x: x[0][-1], list)
176 warn('Unable to extract result entry')
179 tfa = try_get(res, lambda x: x[0][0], list)
181 tfa_str = try_get(tfa, lambda x: x[2], compat_str)
182 if tfa_str == 'TWO_STEP_VERIFICATION':
183 # SEND_SUCCESS - TFA code has been successfully sent to phone
184 # QUOTA_EXCEEDED - reached the limit of TFA codes
185 status = try_get(tfa, lambda x: x[5], compat_str)
186 if status == 'QUOTA_EXCEEDED':
187 warn('Exceeded the limit of TFA codes, try later')
190 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
192 warn('Unable to extract TL')
195 tfa_code = self._get_tfa_info('2-step verification code')
199 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
200 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
203 tfa_code = remove_start(tfa_code, 'G-')
206 user_hash, None, 2, None,
208 9, None, None, None, None, None, None, None,
209 [None, tfa_code, True, 2]
213 self._TFA_URL.format(tl), tfa_req,
214 'Submitting TFA code', 'Unable to submit TFA code')
216 if tfa_results is False:
219 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
221 tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
223 'Unable to finish TFA: %s' % 'Invalid TFA code'
224 if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
227 check_cookie_url = try_get(
228 tfa_results, lambda x: x[0][-1][2], compat_str)
230 check_cookie_url = try_get(res, lambda x: x[2], compat_str)
232 if not check_cookie_url:
233 warn('Unable to extract CheckCookie URL')
236 check_cookie_results = self._download_webpage(
237 check_cookie_url, None, 'Checking cookie', fatal=False)
239 if check_cookie_results is False:
242 if 'https://myaccount.google.com/' not in check_cookie_results:
243 warn('Unable to log in')
248 def _real_initialize(self):
249 if self._downloader is None:
252 if not self._login():
256 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
257 # Extract entries from page with "Load more" button
258 def _entries(self, page, playlist_id):
259 more_widget_html = content_html = page
260 for page_num in itertools.count(1):
261 for entry in self._process_page(content_html):
264 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
268 more = self._download_json(
269 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
270 'Downloading page #%s' % page_num,
271 transform_source=uppercase_escape)
272 content_html = more['content_html']
273 if not content_html.strip():
274 # Some webpages show a "Load more" button but they don't
277 more_widget_html = more['load_more_widget_html']
280 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
281 def _process_page(self, content):
282 for video_id, video_title in self.extract_videos_from_page(content):
283 yield self.url_result(video_id, 'Youtube', video_id, video_title)
285 def extract_videos_from_page(self, page):
288 for mobj in re.finditer(self._VIDEO_RE, page):
289 # The link with index 0 is not the first video of the playlist (not sure if still actual)
290 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
292 video_id = mobj.group('id')
293 video_title = unescapeHTML(mobj.group('title'))
295 video_title = video_title.strip()
297 idx = ids_in_page.index(video_id)
298 if video_title and not titles_in_page[idx]:
299 titles_in_page[idx] = video_title
301 ids_in_page.append(video_id)
302 titles_in_page.append(video_title)
303 return zip(ids_in_page, titles_in_page)
306 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
307 def _process_page(self, content):
308 for playlist_id in orderedSet(re.findall(
309 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
311 yield self.url_result(
312 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
314 def _real_extract(self, url):
315 playlist_id = self._match_id(url)
316 webpage = self._download_webpage(url, playlist_id)
317 title = self._og_search_title(webpage, fatal=False)
318 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
321 class YoutubeIE(YoutubeBaseInfoExtractor):
322 IE_DESC = 'YouTube.com'
323 _VALID_URL = r"""(?x)^
325 (?:https?://|//) # http(s):// or protocol-independent URL
326 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
327 (?:www\.)?deturl\.com/www\.youtube\.com/|
328 (?:www\.)?pwnyoutube\.com/|
329 (?:www\.)?yourepeat\.com/|
330 tube\.majestyc\.net/|
331 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
332 (?:.*?\#/)? # handle anchor (#/) redirect urls
333 (?: # the various things that can precede the ID:
334 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
335 |(?: # or the v= param in all its forms
336 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
337 (?:\?|\#!?) # the params delimiter ? or # or #!
338 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
343 youtu\.be| # just youtu.be/xxxx
344 vid\.plus| # or vid.plus/xxxx
345 zwearz\.com/watch| # or zwearz.com/watch/xxxx
347 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
349 )? # all until now is optional -> you can pass the naked ID
350 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
353 %(playlist_id)s| # combined list/video URLs are handled by the playlist IE
354 WL # WL are handled by the watch later IE
357 (?(1).+)? # if we found the ID, everything can follow
358 $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
359 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
361 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
362 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
363 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
364 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
365 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
366 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
367 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
368 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
369 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
370 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
371 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
372 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
373 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
374 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
375 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
376 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
377 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
378 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
382 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
383 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
384 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
385 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
386 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
387 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
388 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
390 # Apple HTTP Live Streaming
391 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
392 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
393 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
394 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
395 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
396 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
397 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
398 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
401 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
402 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
403 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
404 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
405 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
406 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
407 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
408 '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
409 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
410 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
411 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
412 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
415 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
416 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
417 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
418 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
419 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
420 '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
421 '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
424 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
425 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
426 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
427 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
428 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
429 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
430 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
431 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
432 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
433 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
434 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
435 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
436 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
437 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
438 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
439 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
440 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
441 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
442 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
443 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
444 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
445 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
448 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
449 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
451 # Dash webm audio with opus inside
452 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
453 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
454 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
457 '_rtmp': {'protocol': 'rtmp'},
459 _SUBTITLE_FORMATS = ('ttml', 'vtt')
466 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
470 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
471 'uploader': 'Philipp Hagemeister',
472 'uploader_id': 'phihag',
473 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
474 'upload_date': '20121002',
475 'license': 'Standard YouTube License',
476 '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 .',
477 'categories': ['Science & Technology'],
478 'tags': ['youtube-dl'],
481 'dislike_count': int,
487 'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
488 'note': 'Test generic use_cipher_signature video (#897)',
492 'upload_date': '20120506',
493 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
494 'alt_title': 'I Love It (feat. Charli XCX)',
495 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
496 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
497 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
498 'iconic ep', 'iconic', 'love', 'it'],
500 'uploader': 'Icona Pop',
501 'uploader_id': 'IconaPop',
502 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
503 'license': 'Standard YouTube License',
504 'creator': 'Icona Pop',
508 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
509 'note': 'Test VEVO video with age protection (#956)',
513 'upload_date': '20130703',
514 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
515 'alt_title': 'Tunnel Vision',
516 'description': 'md5:64249768eec3bc4276236606ea996373',
518 'uploader': 'justintimberlakeVEVO',
519 'uploader_id': 'justintimberlakeVEVO',
520 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
521 'license': 'Standard YouTube License',
522 'creator': 'Justin Timberlake',
527 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
528 'note': 'Embed-only video (#1746)',
532 'upload_date': '20120608',
533 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
534 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
535 'uploader': 'SET India',
536 'uploader_id': 'setindia',
537 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
538 'license': 'Standard YouTube License',
543 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
544 'note': 'Use the first video ID in the URL',
548 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
549 'uploader': 'Philipp Hagemeister',
550 'uploader_id': 'phihag',
551 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
552 'upload_date': '20121002',
553 'license': 'Standard YouTube License',
554 '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 .',
555 'categories': ['Science & Technology'],
556 'tags': ['youtube-dl'],
559 'dislike_count': int,
562 'skip_download': True,
566 'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
567 'note': '256k DASH audio (format 141) via DASH manifest',
571 'upload_date': '20121002',
572 'uploader_id': '8KVIDEO',
573 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
575 'uploader': '8KVIDEO',
576 'license': 'Standard YouTube License',
577 'title': 'UHDTV TEST 8K VIDEO.mp4'
580 'youtube_include_dash_manifest': True,
583 'skip': 'format 141 not served anymore',
585 # DASH manifest with encrypted signature
587 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
591 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
592 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
594 'uploader': 'AfrojackVEVO',
595 'uploader_id': 'AfrojackVEVO',
596 'upload_date': '20131011',
597 'license': 'Standard YouTube License',
600 'youtube_include_dash_manifest': True,
601 'format': '141/bestaudio[ext=m4a]',
604 # JS player signature function name containing $
606 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
610 'title': 'Taylor Swift - Shake It Off',
611 'alt_title': 'Shake It Off',
612 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
614 'uploader': 'TaylorSwiftVEVO',
615 'uploader_id': 'TaylorSwiftVEVO',
616 'upload_date': '20140818',
617 'license': 'Standard YouTube License',
618 'creator': 'Taylor Swift',
621 'youtube_include_dash_manifest': True,
622 'format': '141/bestaudio[ext=m4a]',
627 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
632 'upload_date': '20100909',
633 'uploader': 'The Amazing Atheist',
634 'uploader_id': 'TheAmazingAtheist',
635 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
636 'license': 'Standard YouTube License',
637 'title': 'Burning Everyone\'s Koran',
638 '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',
641 # Normal age-gate video (No vevo, embed allowed)
643 'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
647 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
648 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
650 'uploader': 'The Witcher',
651 'uploader_id': 'WitcherGame',
652 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
653 'upload_date': '20140605',
654 'license': 'Standard YouTube License',
658 # Age-gate video with encrypted signature
660 'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
664 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
665 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
667 'uploader': 'LloydVEVO',
668 'uploader_id': 'LloydVEVO',
669 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
670 'upload_date': '20110629',
671 'license': 'Standard YouTube License',
675 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
677 'url': '__2ABJjxzNo',
682 'upload_date': '20100430',
683 'uploader_id': 'deadmau5',
684 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
685 'creator': 'deadmau5',
686 'description': 'md5:12c56784b8032162bb936a5f76d55360',
687 'uploader': 'deadmau5',
688 'license': 'Standard YouTube License',
689 'title': 'Deadmau5 - Some Chords (HD)',
690 'alt_title': 'Some Chords',
692 'expected_warnings': [
693 'DASH manifest missing',
696 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
698 'url': 'lqQg6PlCWgI',
703 'upload_date': '20150827',
704 'uploader_id': 'olympic',
705 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
706 'license': 'Standard YouTube License',
707 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
708 'uploader': 'Olympic',
709 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
712 'skip_download': 'requires avconv',
717 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
721 'stretched_ratio': 16 / 9.,
723 'upload_date': '20110310',
724 'uploader_id': 'AllenMeow',
725 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
726 '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',
728 'license': 'Standard YouTube License',
729 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
732 # url_encoded_fmt_stream_map is empty string
734 'url': 'qEJwOuvDf7I',
738 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
740 'upload_date': '20150404',
741 'uploader_id': 'spbelect',
742 'uploader': 'Наблюдатели Петербурга',
745 'skip_download': 'requires avconv',
747 'skip': 'This live event has ended.',
749 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
751 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
755 'title': 'md5:7b81415841e02ecd4313668cde88737a',
756 'description': 'md5:116377fd2963b81ec4ce64b542173306',
758 'upload_date': '20150625',
759 'uploader_id': 'dorappi2000',
760 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
761 'uploader': 'dorappi2000',
762 'license': 'Standard YouTube License',
763 'formats': 'mincount:32',
766 # DASH manifest with segment_list
768 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
769 'md5': '8ce563a1d667b599d21064e982ab9e31',
773 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
774 'uploader': 'Airtek',
775 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
776 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
777 'license': 'Standard YouTube License',
778 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
781 'youtube_include_dash_manifest': True,
782 'format': '135', # bestvideo
784 'skip': 'This live event has ended.',
787 # Multifeed videos (multiple cameras), URL is for Main Camera
788 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
791 'title': 'teamPGP: Rocket League Noob Stream',
792 'description': 'md5:dc7872fb300e143831327f1bae3af010',
798 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
799 'description': 'md5:dc7872fb300e143831327f1bae3af010',
801 'upload_date': '20150721',
802 'uploader': 'Beer Games Beer',
803 'uploader_id': 'beergamesbeer',
804 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
805 'license': 'Standard YouTube License',
811 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
812 'description': 'md5:dc7872fb300e143831327f1bae3af010',
814 'upload_date': '20150721',
815 'uploader': 'Beer Games Beer',
816 'uploader_id': 'beergamesbeer',
817 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
818 'license': 'Standard YouTube License',
824 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
825 'description': 'md5:dc7872fb300e143831327f1bae3af010',
827 'upload_date': '20150721',
828 'uploader': 'Beer Games Beer',
829 'uploader_id': 'beergamesbeer',
830 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
831 'license': 'Standard YouTube License',
837 'title': 'teamPGP: Rocket League Noob Stream (zim)',
838 'description': 'md5:dc7872fb300e143831327f1bae3af010',
840 'upload_date': '20150721',
841 'uploader': 'Beer Games Beer',
842 'uploader_id': 'beergamesbeer',
843 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
844 'license': 'Standard YouTube License',
848 'skip_download': True,
852 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
853 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
856 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
859 'skip': 'Not multifeed anymore',
862 'url': 'https://vid.plus/FlRa-iH7PGw',
863 'only_matching': True,
866 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
867 'only_matching': True,
870 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
871 # Also tests cut-off URL expansion in video description (see
872 # https://github.com/rg3/youtube-dl/issues/1892,
873 # https://github.com/rg3/youtube-dl/issues/8164)
874 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
878 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
879 'alt_title': 'Dark Walk',
880 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
882 'upload_date': '20151119',
883 'uploader_id': 'IronSoulElf',
884 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
885 'uploader': 'IronSoulElf',
886 'license': 'Standard YouTube License',
887 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
890 'skip_download': True,
894 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
895 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
896 'only_matching': True,
899 # Video with yt:stretch=17:0
900 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
904 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
905 'description': 'md5:ee18a25c350637c8faff806845bddee9',
906 'upload_date': '20151107',
907 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
908 'uploader': 'CH GAMER DROID',
911 'skip_download': True,
913 'skip': 'This video does not exist.',
916 # Video licensed under Creative Commons
917 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
921 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
922 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
924 'upload_date': '20150127',
925 'uploader_id': 'BerkmanCenter',
926 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
927 'uploader': 'The Berkman Klein Center for Internet & Society',
928 'license': 'Creative Commons Attribution license (reuse allowed)',
931 'skip_download': True,
935 # Channel-like uploader_url
936 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
940 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
941 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
943 'upload_date': '20151119',
944 'uploader': 'Bernie 2016',
945 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
946 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
947 'license': 'Creative Commons Attribution license (reuse allowed)',
950 'skip_download': True,
954 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
955 'only_matching': True,
958 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
959 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
960 'only_matching': True,
963 # Rental video preview
964 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
968 'title': 'Piku - Trailer',
969 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
970 'upload_date': '20150811',
971 'uploader': 'FlixMatrix',
972 'uploader_id': 'FlixMatrixKaravan',
973 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
974 'license': 'Standard YouTube License',
977 'skip_download': True,
981 # YouTube Red video with episode data
982 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
986 'title': 'Isolation - Mind Field (Ep 1)',
987 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
989 'upload_date': '20170118',
990 'uploader': 'Vsauce',
991 'uploader_id': 'Vsauce',
992 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
993 'license': 'Standard YouTube License',
994 'series': 'Mind Field',
999 'skip_download': True,
1001 'expected_warnings': [
1002 'Skipping DASH manifest',
1007 'url': '1t24XAntNCY',
1008 'only_matching': True,
1011 # geo restricted to JP
1012 'url': 'sJL6WA-aGkQ',
1013 'only_matching': True,
1016 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1017 'only_matching': True,
1021 def __init__(self, *args, **kwargs):
1022 super(YoutubeIE, self).__init__(*args, **kwargs)
1023 self._player_cache = {}
1025 def report_video_info_webpage_download(self, video_id):
1026 """Report attempt to download video info webpage."""
1027 self.to_screen('%s: Downloading video info webpage' % video_id)
1029 def report_information_extraction(self, video_id):
1030 """Report attempt to extract video information."""
1031 self.to_screen('%s: Extracting video information' % video_id)
1033 def report_unavailable_format(self, video_id, format):
1034 """Report extracted video URL."""
1035 self.to_screen('%s: Format %s not available' % (video_id, format))
1037 def report_rtmp_download(self):
1038 """Indicate the download will use the RTMP protocol."""
1039 self.to_screen('RTMP download detected')
1041 def _signature_cache_id(self, example_sig):
1042 """ Return a string representation of a signature """
1043 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1045 def _extract_signature_function(self, video_id, player_url, example_sig):
1047 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1050 raise ExtractorError('Cannot identify player %r' % player_url)
1051 player_type = id_m.group('ext')
1052 player_id = id_m.group('id')
1054 # Read from filesystem cache
1055 func_id = '%s_%s_%s' % (
1056 player_type, player_id, self._signature_cache_id(example_sig))
1057 assert os.path.basename(func_id) == func_id
1059 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1060 if cache_spec is not None:
1061 return lambda s: ''.join(s[i] for i in cache_spec)
1064 'Downloading player %s' % player_url
1065 if self._downloader.params.get('verbose') else
1066 'Downloading %s player %s' % (player_type, player_id)
1068 if player_type == 'js':
1069 code = self._download_webpage(
1070 player_url, video_id,
1072 errnote='Download of %s failed' % player_url)
1073 res = self._parse_sig_js(code)
1074 elif player_type == 'swf':
1075 urlh = self._request_webpage(
1076 player_url, video_id,
1078 errnote='Download of %s failed' % player_url)
1080 res = self._parse_sig_swf(code)
1082 assert False, 'Invalid player type %r' % player_type
1084 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1085 cache_res = res(test_string)
1086 cache_spec = [ord(c) for c in cache_res]
1088 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1091 def _print_sig_code(self, func, example_sig):
1092 def gen_sig_code(idxs):
1093 def _genslice(start, end, step):
1094 starts = '' if start == 0 else str(start)
1095 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1096 steps = '' if step == 1 else (':%d' % step)
1097 return 's[%s%s%s]' % (starts, ends, steps)
1100 # Quelch pyflakes warnings - start will be set when step is set
1101 start = '(Never used)'
1102 for i, prev in zip(idxs[1:], idxs[:-1]):
1103 if step is not None:
1104 if i - prev == step:
1106 yield _genslice(start, prev, step)
1109 if i - prev in [-1, 1]:
1114 yield 's[%d]' % prev
1118 yield _genslice(start, i, step)
1120 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1121 cache_res = func(test_string)
1122 cache_spec = [ord(c) for c in cache_res]
1123 expr_code = ' + '.join(gen_sig_code(cache_spec))
1124 signature_id_tuple = '(%s)' % (
1125 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1126 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1127 ' return %s\n') % (signature_id_tuple, expr_code)
1128 self.to_screen('Extracted signature function:\n' + code)
1130 def _parse_sig_js(self, jscode):
1131 funcname = self._search_regex(
1132 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1133 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1134 jscode, 'Initial JS player signature function name', group='sig')
1136 jsi = JSInterpreter(jscode)
1137 initial_function = jsi.extract_function(funcname)
1138 return lambda s: initial_function([s])
1140 def _parse_sig_swf(self, file_contents):
1141 swfi = SWFInterpreter(file_contents)
1142 TARGET_CLASSNAME = 'SignatureDecipher'
1143 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1144 initial_function = swfi.extract_function(searched_class, 'decipher')
1145 return lambda s: initial_function([s])
1147 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1148 """Turn the encrypted s field into a working signature"""
1150 if player_url is None:
1151 raise ExtractorError('Cannot decrypt signature without player_url')
1153 if player_url.startswith('//'):
1154 player_url = 'https:' + player_url
1155 elif not re.match(r'https?://', player_url):
1156 player_url = compat_urlparse.urljoin(
1157 'https://www.youtube.com', player_url)
1159 player_id = (player_url, self._signature_cache_id(s))
1160 if player_id not in self._player_cache:
1161 func = self._extract_signature_function(
1162 video_id, player_url, s
1164 self._player_cache[player_id] = func
1165 func = self._player_cache[player_id]
1166 if self._downloader.params.get('youtube_print_sig_code'):
1167 self._print_sig_code(func, s)
1169 except Exception as e:
1170 tb = traceback.format_exc()
1171 raise ExtractorError(
1172 'Signature extraction failed: ' + tb, cause=e)
1174 def _get_subtitles(self, video_id, webpage):
1176 subs_doc = self._download_xml(
1177 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1178 video_id, note=False)
1179 except ExtractorError as err:
1180 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1184 for track in subs_doc.findall('track'):
1185 lang = track.attrib['lang_code']
1186 if lang in sub_lang_list:
1189 for ext in self._SUBTITLE_FORMATS:
1190 params = compat_urllib_parse_urlencode({
1194 'name': track.attrib['name'].encode('utf-8'),
1196 sub_formats.append({
1197 'url': 'https://www.youtube.com/api/timedtext?' + params,
1200 sub_lang_list[lang] = sub_formats
1201 if not sub_lang_list:
1202 self._downloader.report_warning('video doesn\'t have subtitles')
1204 return sub_lang_list
1206 def _get_ytplayer_config(self, video_id, webpage):
1208 # User data may contain arbitrary character sequences that may affect
1209 # JSON extraction with regex, e.g. when '};' is contained the second
1210 # regex won't capture the whole JSON. Yet working around by trying more
1211 # concrete regex first keeping in mind proper quoted string handling
1212 # to be implemented in future that will replace this workaround (see
1213 # https://github.com/rg3/youtube-dl/issues/7468,
1214 # https://github.com/rg3/youtube-dl/pull/7599)
1215 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1216 r';ytplayer\.config\s*=\s*({.+?});',
1218 config = self._search_regex(
1219 patterns, webpage, 'ytplayer.config', default=None)
1221 return self._parse_json(
1222 uppercase_escape(config), video_id, fatal=False)
1224 def _get_automatic_captions(self, video_id, webpage):
1225 """We need the webpage for getting the captions url, pass it as an
1226 argument to speed up the process."""
1227 self.to_screen('%s: Looking for automatic captions' % video_id)
1228 player_config = self._get_ytplayer_config(video_id, webpage)
1229 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1230 if not player_config:
1231 self._downloader.report_warning(err_msg)
1234 args = player_config['args']
1235 caption_url = args.get('ttsurl')
1237 timestamp = args['timestamp']
1238 # We get the available subtitles
1239 list_params = compat_urllib_parse_urlencode({
1244 list_url = caption_url + '&' + list_params
1245 caption_list = self._download_xml(list_url, video_id)
1246 original_lang_node = caption_list.find('track')
1247 if original_lang_node is None:
1248 self._downloader.report_warning('Video doesn\'t have automatic captions')
1250 original_lang = original_lang_node.attrib['lang_code']
1251 caption_kind = original_lang_node.attrib.get('kind', '')
1254 for lang_node in caption_list.findall('target'):
1255 sub_lang = lang_node.attrib['lang_code']
1257 for ext in self._SUBTITLE_FORMATS:
1258 params = compat_urllib_parse_urlencode({
1259 'lang': original_lang,
1263 'kind': caption_kind,
1265 sub_formats.append({
1266 'url': caption_url + '&' + params,
1269 sub_lang_list[sub_lang] = sub_formats
1270 return sub_lang_list
1272 def make_captions(sub_url, sub_langs):
1273 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1274 caption_qs = compat_parse_qs(parsed_sub_url.query)
1276 for sub_lang in sub_langs:
1278 for ext in self._SUBTITLE_FORMATS:
1280 'tlang': [sub_lang],
1283 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1284 query=compat_urllib_parse_urlencode(caption_qs, True)))
1285 sub_formats.append({
1289 captions[sub_lang] = sub_formats
1292 # New captions format as of 22.06.2017
1293 player_response = args.get('player_response')
1294 if player_response and isinstance(player_response, compat_str):
1295 player_response = self._parse_json(
1296 player_response, video_id, fatal=False)
1298 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1299 base_url = renderer['captionTracks'][0]['baseUrl']
1301 for lang in renderer['translationLanguages']:
1302 lang_code = lang.get('languageCode')
1304 sub_lang_list.append(lang_code)
1305 return make_captions(base_url, sub_lang_list)
1307 # Some videos don't provide ttsurl but rather caption_tracks and
1308 # caption_translation_languages (e.g. 20LmZk1hakA)
1309 # Does not used anymore as of 22.06.2017
1310 caption_tracks = args['caption_tracks']
1311 caption_translation_languages = args['caption_translation_languages']
1312 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1314 for lang in caption_translation_languages.split(','):
1315 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1316 sub_lang = lang_qs.get('lc', [None])[0]
1318 sub_lang_list.append(sub_lang)
1319 return make_captions(caption_url, sub_lang_list)
1320 # An extractor error can be raise by the download process if there are
1321 # no automatic captions but there are subtitles
1322 except (KeyError, IndexError, ExtractorError):
1323 self._downloader.report_warning(err_msg)
1326 def _mark_watched(self, video_id, video_info):
1327 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1328 if not playback_url:
1330 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1331 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1333 # cpn generation algorithm is reverse engineered from base.js.
1334 # In fact it works even with dummy cpn.
1335 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1336 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1342 playback_url = compat_urlparse.urlunparse(
1343 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1345 self._download_webpage(
1346 playback_url, video_id, 'Marking watched',
1347 'Unable to mark watched', fatal=False)
1350 def extract_id(cls, url):
1351 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1353 raise ExtractorError('Invalid URL: %s' % url)
1354 video_id = mobj.group(2)
1357 def _extract_annotations(self, video_id):
1358 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1359 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1362 def _extract_chapters(description, duration):
1365 chapter_lines = re.findall(
1366 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1368 if not chapter_lines:
1371 for next_num, (chapter_line, time_point) in enumerate(
1372 chapter_lines, start=1):
1373 start_time = parse_duration(time_point)
1374 if start_time is None:
1376 if start_time > duration:
1378 end_time = (duration if next_num == len(chapter_lines)
1379 else parse_duration(chapter_lines[next_num][1]))
1380 if end_time is None:
1382 if end_time > duration:
1384 if start_time > end_time:
1386 chapter_title = re.sub(
1387 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1388 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1390 'start_time': start_time,
1391 'end_time': end_time,
1392 'title': chapter_title,
1396 def _real_extract(self, url):
1397 url, smuggled_data = unsmuggle_url(url, {})
1400 'http' if self._downloader.params.get('prefer_insecure', False)
1405 parsed_url = compat_urllib_parse_urlparse(url)
1406 for component in [parsed_url.fragment, parsed_url.query]:
1407 query = compat_parse_qs(component)
1408 if start_time is None and 't' in query:
1409 start_time = parse_duration(query['t'][0])
1410 if start_time is None and 'start' in query:
1411 start_time = parse_duration(query['start'][0])
1412 if end_time is None and 'end' in query:
1413 end_time = parse_duration(query['end'][0])
1415 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1416 mobj = re.search(self._NEXT_URL_RE, url)
1418 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1419 video_id = self.extract_id(url)
1422 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1423 video_webpage = self._download_webpage(url, video_id)
1425 # Attempt to extract SWF player URL
1426 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1427 if mobj is not None:
1428 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1434 def add_dash_mpd(video_info):
1435 dash_mpd = video_info.get('dashmpd')
1436 if dash_mpd and dash_mpd[0] not in dash_mpds:
1437 dash_mpds.append(dash_mpd[0])
1440 embed_webpage = None
1442 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1444 # We simulate the access to the video from www.youtube.com/v/{video_id}
1445 # this can be viewed without login into Youtube
1446 url = proto + '://www.youtube.com/embed/%s' % video_id
1447 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1448 data = compat_urllib_parse_urlencode({
1449 'video_id': video_id,
1450 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1451 'sts': self._search_regex(
1452 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1454 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1455 video_info_webpage = self._download_webpage(
1456 video_info_url, video_id,
1457 note='Refetching age-gated info webpage',
1458 errnote='unable to download video info webpage')
1459 video_info = compat_parse_qs(video_info_webpage)
1460 add_dash_mpd(video_info)
1465 # Try looking directly into the video webpage
1466 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1468 args = ytplayer_config['args']
1469 if args.get('url_encoded_fmt_stream_map'):
1470 # Convert to the same format returned by compat_parse_qs
1471 video_info = dict((k, [v]) for k, v in args.items())
1472 add_dash_mpd(video_info)
1473 # Rental video is not rented but preview is available (e.g.
1474 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1475 # https://github.com/rg3/youtube-dl/issues/10532)
1476 if not video_info and args.get('ypc_vid'):
1477 return self.url_result(
1478 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1479 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1481 sts = ytplayer_config.get('sts')
1482 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1483 # We also try looking in get_video_info since it may contain different dashmpd
1484 # URL that points to a DASH manifest with possibly different itag set (some itags
1485 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1486 # manifest pointed by get_video_info's dashmpd).
1487 # The general idea is to take a union of itags of both DASH manifests (for example
1488 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1489 self.report_video_info_webpage_download(video_id)
1490 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1492 'video_id': video_id,
1502 video_info_webpage = self._download_webpage(
1503 '%s://www.youtube.com/get_video_info' % proto,
1504 video_id, note=False,
1505 errnote='unable to download video info webpage',
1506 fatal=False, query=query)
1507 if not video_info_webpage:
1509 get_video_info = compat_parse_qs(video_info_webpage)
1510 add_dash_mpd(get_video_info)
1512 video_info = get_video_info
1513 if 'token' in get_video_info:
1514 # Different get_video_info requests may report different results, e.g.
1515 # some may report video unavailability, but some may serve it without
1516 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1517 # the original webpage as well as el=info and el=embedded get_video_info
1518 # requests report video unavailability due to geo restriction while
1519 # el=detailpage succeeds and returns valid data). This is probably
1520 # due to YouTube measures against IP ranges of hosting providers.
1521 # Working around by preferring the first succeeded video_info containing
1522 # the token if no such video_info yet was found.
1523 if 'token' not in video_info:
1524 video_info = get_video_info
1526 if 'token' not in video_info:
1527 if 'reason' in video_info:
1528 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1529 regions_allowed = self._html_search_meta(
1530 'regionsAllowed', video_webpage, default=None)
1531 countries = regions_allowed.split(',') if regions_allowed else None
1532 self.raise_geo_restricted(
1533 msg=video_info['reason'][0], countries=countries)
1534 raise ExtractorError(
1535 'YouTube said: %s' % video_info['reason'][0],
1536 expected=True, video_id=video_id)
1538 raise ExtractorError(
1539 '"token" parameter not in video info for unknown reason',
1543 if 'title' in video_info:
1544 video_title = video_info['title'][0]
1546 self._downloader.report_warning('Unable to extract video title')
1550 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1551 if video_description:
1552 description_original = video_description = re.sub(r'''(?x)
1554 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1555 (?:title|href)="([^"]+)"\s+
1556 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1560 ''', r'\1', video_description)
1561 video_description = clean_html(video_description)
1563 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1565 video_description = unescapeHTML(fd_mobj.group(1))
1567 video_description = ''
1569 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1570 if not self._downloader.params.get('noplaylist'):
1573 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1574 for feed in multifeed_metadata_list.split(','):
1575 # Unquote should take place before split on comma (,) since textual
1576 # fields may contain comma as well (see
1577 # https://github.com/rg3/youtube-dl/issues/8536)
1578 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1580 '_type': 'url_transparent',
1581 'ie_key': 'Youtube',
1583 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1584 {'force_singlefeed': True}),
1585 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1587 feed_ids.append(feed_data['id'][0])
1589 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1590 % (', '.join(feed_ids), video_id))
1591 return self.playlist_result(entries, video_id, video_title, video_description)
1592 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1594 if 'view_count' in video_info:
1595 view_count = int(video_info['view_count'][0])
1599 # Check for "rental" videos
1600 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1601 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1603 # Start extracting information
1604 self.report_information_extraction(video_id)
1607 if 'author' not in video_info:
1608 raise ExtractorError('Unable to extract uploader name')
1609 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1612 video_uploader_id = None
1613 video_uploader_url = None
1615 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1617 if mobj is not None:
1618 video_uploader_id = mobj.group('uploader_id')
1619 video_uploader_url = mobj.group('uploader_url')
1621 self._downloader.report_warning('unable to extract uploader nickname')
1624 # We try first to get a high quality image:
1625 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1626 video_webpage, re.DOTALL)
1627 if m_thumb is not None:
1628 video_thumbnail = m_thumb.group(1)
1629 elif 'thumbnail_url' not in video_info:
1630 self._downloader.report_warning('unable to extract video thumbnail')
1631 video_thumbnail = None
1632 else: # don't panic if we can't find it
1633 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1636 upload_date = self._html_search_meta(
1637 'datePublished', video_webpage, 'upload date', default=None)
1639 upload_date = self._search_regex(
1640 [r'(?s)id="eow-date.*?>(.*?)</span>',
1641 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1642 video_webpage, 'upload date', default=None)
1644 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1645 upload_date = unified_strdate(upload_date)
1647 video_license = self._html_search_regex(
1648 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1649 video_webpage, 'license', default=None)
1651 m_music = re.search(
1652 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1655 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1656 video_creator = clean_html(m_music.group('creator'))
1658 video_alt_title = video_creator = None
1660 m_episode = re.search(
1661 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1664 series = m_episode.group('series')
1665 season_number = int(m_episode.group('season'))
1666 episode_number = int(m_episode.group('episode'))
1668 series = season_number = episode_number = None
1670 m_cat_container = self._search_regex(
1671 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1672 video_webpage, 'categories', default=None)
1674 category = self._html_search_regex(
1675 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1677 video_categories = None if category is None else [category]
1679 video_categories = None
1682 unescapeHTML(m.group('content'))
1683 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1685 def _extract_count(count_name):
1686 return str_to_int(self._search_regex(
1687 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1688 % re.escape(count_name),
1689 video_webpage, count_name, default=None))
1691 like_count = _extract_count('like')
1692 dislike_count = _extract_count('dislike')
1695 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1696 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1698 video_duration = try_get(
1699 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1700 if not video_duration:
1701 video_duration = parse_duration(self._html_search_meta(
1702 'duration', video_webpage, 'video duration'))
1705 video_annotations = None
1706 if self._downloader.params.get('writeannotations', False):
1707 video_annotations = self._extract_annotations(video_id)
1709 chapters = self._extract_chapters(description_original, video_duration)
1711 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1712 self.report_rtmp_download()
1714 'format_id': '_rtmp',
1716 'url': video_info['conn'][0],
1717 'player_url': player_url,
1719 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1720 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1721 if 'rtmpe%3Dyes' in encoded_url_map:
1722 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1724 fmt_list = video_info.get('fmt_list', [''])[0]
1726 for fmt in fmt_list.split(','):
1727 spec = fmt.split('/')
1729 width_height = spec[1].split('x')
1730 if len(width_height) == 2:
1731 formats_spec[spec[0]] = {
1732 'resolution': spec[1],
1733 'width': int_or_none(width_height[0]),
1734 'height': int_or_none(width_height[1]),
1737 for url_data_str in encoded_url_map.split(','):
1738 url_data = compat_parse_qs(url_data_str)
1739 if 'itag' not in url_data or 'url' not in url_data:
1741 format_id = url_data['itag'][0]
1742 url = url_data['url'][0]
1744 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1745 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1746 jsplayer_url_json = self._search_regex(
1748 embed_webpage if age_gate else video_webpage,
1749 'JS player URL (1)', default=None)
1750 if not jsplayer_url_json and not age_gate:
1751 # We need the embed website after all
1752 if embed_webpage is None:
1753 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1754 embed_webpage = self._download_webpage(
1755 embed_url, video_id, 'Downloading embed webpage')
1756 jsplayer_url_json = self._search_regex(
1757 ASSETS_RE, embed_webpage, 'JS player URL')
1759 player_url = json.loads(jsplayer_url_json)
1760 if player_url is None:
1761 player_url_json = self._search_regex(
1762 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1763 video_webpage, 'age gate player URL')
1764 player_url = json.loads(player_url_json)
1766 if 'sig' in url_data:
1767 url += '&signature=' + url_data['sig'][0]
1768 elif 's' in url_data:
1769 encrypted_sig = url_data['s'][0]
1771 if self._downloader.params.get('verbose'):
1772 if player_url is None:
1773 player_version = 'unknown'
1774 player_desc = 'unknown'
1776 if player_url.endswith('swf'):
1777 player_version = self._search_regex(
1778 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1779 'flash player', fatal=False)
1780 player_desc = 'flash player %s' % player_version
1782 player_version = self._search_regex(
1783 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1784 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1786 'html5 player', fatal=False)
1787 player_desc = 'html5 player %s' % player_version
1789 parts_sizes = self._signature_cache_id(encrypted_sig)
1790 self.to_screen('{%s} signature length %s, %s' %
1791 (format_id, parts_sizes, player_desc))
1793 signature = self._decrypt_signature(
1794 encrypted_sig, video_id, player_url, age_gate)
1795 url += '&signature=' + signature
1796 if 'ratebypass' not in url:
1797 url += '&ratebypass=yes'
1800 'format_id': format_id,
1802 'player_url': player_url,
1804 if format_id in self._formats:
1805 dct.update(self._formats[format_id])
1806 if format_id in formats_spec:
1807 dct.update(formats_spec[format_id])
1809 # Some itags are not included in DASH manifest thus corresponding formats will
1810 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1811 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1812 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1813 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1816 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1817 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1820 'fps': int_or_none(url_data.get('fps', [None])[0]),
1821 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1823 for key, value in more_fields.items():
1826 type_ = url_data.get('type', [None])[0]
1828 type_split = type_.split(';')
1829 kind_ext = type_split[0].split('/')
1830 if len(kind_ext) == 2:
1832 dct['ext'] = mimetype2ext(type_split[0])
1833 if kind in ('audio', 'video'):
1835 for mobj in re.finditer(
1836 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1837 if mobj.group('key') == 'codecs':
1838 codecs = mobj.group('val')
1841 dct.update(parse_codecs(codecs))
1843 elif video_info.get('hlsvp'):
1844 manifest_url = video_info['hlsvp'][0]
1846 m3u8_formats = self._extract_m3u8_formats(
1847 manifest_url, video_id, 'mp4', fatal=False)
1848 for a_format in m3u8_formats:
1849 itag = self._search_regex(
1850 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1852 a_format['format_id'] = itag
1853 if itag in self._formats:
1854 dct = self._formats[itag].copy()
1855 dct.update(a_format)
1857 a_format['player_url'] = player_url
1858 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1859 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1860 formats.append(a_format)
1862 unavailable_message = self._html_search_regex(
1863 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1864 video_webpage, 'unavailable message', default=None)
1865 if unavailable_message:
1866 raise ExtractorError(unavailable_message, expected=True)
1867 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1869 # Look for the DASH manifest
1870 if self._downloader.params.get('youtube_include_dash_manifest', True):
1871 dash_mpd_fatal = True
1872 for mpd_url in dash_mpds:
1875 def decrypt_sig(mobj):
1877 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1878 return '/signature/%s' % dec_s
1880 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1882 for df in self._extract_mpd_formats(
1883 mpd_url, video_id, fatal=dash_mpd_fatal,
1884 formats_dict=self._formats):
1885 # Do not overwrite DASH format found in some previous DASH manifest
1886 if df['format_id'] not in dash_formats:
1887 dash_formats[df['format_id']] = df
1888 # Additional DASH manifests may end up in HTTP Error 403 therefore
1889 # allow them to fail without bug report message if we already have
1890 # some DASH manifest succeeded. This is temporary workaround to reduce
1891 # burst of bug reports until we figure out the reason and whether it
1892 # can be fixed at all.
1893 dash_mpd_fatal = False
1894 except (ExtractorError, KeyError) as e:
1895 self.report_warning(
1896 'Skipping DASH manifest: %r' % e, video_id)
1898 # Remove the formats we found through non-DASH, they
1899 # contain less info and it can be wrong, because we use
1900 # fixed values (for example the resolution). See
1901 # https://github.com/rg3/youtube-dl/issues/5774 for an
1903 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1904 formats.extend(dash_formats.values())
1906 # Check for malformed aspect ratio
1907 stretched_m = re.search(
1908 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1911 w = float(stretched_m.group('w'))
1912 h = float(stretched_m.group('h'))
1913 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1914 # We will only process correct ratios.
1918 if f.get('vcodec') != 'none':
1919 f['stretched_ratio'] = ratio
1921 self._sort_formats(formats)
1923 self.mark_watched(video_id, video_info)
1927 'uploader': video_uploader,
1928 'uploader_id': video_uploader_id,
1929 'uploader_url': video_uploader_url,
1930 'upload_date': upload_date,
1931 'license': video_license,
1932 'creator': video_creator,
1933 'title': video_title,
1934 'alt_title': video_alt_title,
1935 'thumbnail': video_thumbnail,
1936 'description': video_description,
1937 'categories': video_categories,
1939 'subtitles': video_subtitles,
1940 'automatic_captions': automatic_captions,
1941 'duration': video_duration,
1942 'age_limit': 18 if age_gate else 0,
1943 'annotations': video_annotations,
1944 'chapters': chapters,
1945 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1946 'view_count': view_count,
1947 'like_count': like_count,
1948 'dislike_count': dislike_count,
1949 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1952 'start_time': start_time,
1953 'end_time': end_time,
1955 'season_number': season_number,
1956 'episode_number': episode_number,
1960 class YoutubeSharedVideoIE(InfoExtractor):
1961 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1962 IE_NAME = 'youtube:shared'
1965 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1967 'id': 'uPDB5I9wfp8',
1969 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1970 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1971 'upload_date': '20160219',
1972 'uploader': 'Pocoyo - Português (BR)',
1973 'uploader_id': 'PocoyoBrazil',
1975 'add_ie': ['Youtube'],
1977 # There are already too many Youtube downloads
1978 'skip_download': True,
1982 def _real_extract(self, url):
1983 video_id = self._match_id(url)
1985 webpage = self._download_webpage(url, video_id)
1987 real_video_id = self._html_search_meta(
1988 'videoId', webpage, 'YouTube video id', fatal=True)
1990 return self.url_result(real_video_id, YoutubeIE.ie_key())
1993 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1994 IE_DESC = 'YouTube.com playlists'
1995 _VALID_URL = r"""(?x)(?:
2001 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2002 \? (?:.*?[&;])*? (?:p|a|list)=
2005 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2008 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2009 # Top tracks, they can also include dots
2015 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2016 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
2017 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2018 IE_NAME = 'youtube:playlist'
2020 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2022 'title': 'ytdl test PL',
2023 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2025 'playlist_count': 3,
2027 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2029 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2030 'title': 'YDL_Empty_List',
2032 'playlist_count': 0,
2033 'skip': 'This playlist is private',
2035 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2036 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2038 'title': '29C3: Not my department',
2039 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2041 'playlist_count': 95,
2043 'note': 'issue #673',
2044 'url': 'PLBB231211A4F62143',
2046 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2047 'id': 'PLBB231211A4F62143',
2049 'playlist_mincount': 26,
2051 'note': 'Large playlist',
2052 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2054 'title': 'Uploads from Cauchemar',
2055 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2057 'playlist_mincount': 799,
2059 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2061 'title': 'YDL_safe_search',
2062 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2064 'playlist_count': 2,
2065 'skip': 'This playlist is private',
2068 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2069 'playlist_count': 4,
2072 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2075 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2076 'playlist_mincount': 485,
2078 'title': '2017 華語最新單曲 (2/24更新)',
2079 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2082 'note': 'Embedded SWF player',
2083 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2084 'playlist_count': 4,
2087 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2090 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2091 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2093 'title': 'Uploads from Interstellar Movie',
2094 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2096 'playlist_mincount': 21,
2098 # Playlist URL that does not actually serve a playlist
2099 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2101 'id': 'FqZTN594JQw',
2103 'title': "Smiley's People 01 detective, Adventure Series, Action",
2104 'uploader': 'STREEM',
2105 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2106 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2107 'upload_date': '20150526',
2108 'license': 'Standard YouTube License',
2109 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2110 'categories': ['People & Blogs'],
2113 'dislike_count': int,
2116 'skip_download': True,
2118 'add_ie': [YoutubeIE.ie_key()],
2120 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2122 'id': 'yeWKywCrFtk',
2124 'title': 'Small Scale Baler and Braiding Rugs',
2125 'uploader': 'Backus-Page House Museum',
2126 'uploader_id': 'backuspagemuseum',
2127 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2128 'upload_date': '20161008',
2129 'license': 'Standard YouTube License',
2130 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2131 'categories': ['Nonprofits & Activism'],
2134 'dislike_count': int,
2138 'skip_download': True,
2141 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2142 'only_matching': True,
2144 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2145 'only_matching': True,
2148 def _real_initialize(self):
2151 def _extract_mix(self, playlist_id):
2152 # The mixes are generated from a single video
2153 # the id of the playlist is just 'RD' + video_id
2155 last_id = playlist_id[-11:]
2156 for n in itertools.count(1):
2157 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2158 webpage = self._download_webpage(
2159 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2160 new_ids = orderedSet(re.findall(
2161 r'''(?xs)data-video-username=".*?".*?
2162 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2164 # Fetch new pages until all the videos are repeated, it seems that
2165 # there are always 51 unique videos.
2166 new_ids = [_id for _id in new_ids if _id not in ids]
2172 url_results = self._ids_to_results(ids)
2174 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2176 search_title('playlist-title') or
2177 search_title('title long-title') or
2178 search_title('title'))
2179 title = clean_html(title_span)
2181 return self.playlist_result(url_results, playlist_id, title)
2183 def _extract_playlist(self, playlist_id):
2184 url = self._TEMPLATE_URL % playlist_id
2185 page = self._download_webpage(url, playlist_id)
2187 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2188 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2189 match = match.strip()
2190 # Check if the playlist exists or is private
2191 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2193 reason = mobj.group('reason')
2194 message = 'This playlist %s' % reason
2195 if 'private' in reason:
2196 message += ', use --username or --netrc to access it'
2198 raise ExtractorError(message, expected=True)
2199 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2200 raise ExtractorError(
2201 'Invalid parameters. Maybe URL is incorrect.',
2203 elif re.match(r'[^<]*Choose your language[^<]*', match):
2206 self.report_warning('Youtube gives an alert message: ' + match)
2208 playlist_title = self._html_search_regex(
2209 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2210 page, 'title', default=None)
2214 if not playlist_title:
2216 # Some playlist URLs don't actually serve a playlist (e.g.
2217 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2218 next(self._entries(page, playlist_id))
2219 except StopIteration:
2222 return has_videos, self.playlist_result(
2223 self._entries(page, playlist_id), playlist_id, playlist_title)
2225 def _check_download_just_video(self, url, playlist_id):
2226 # Check if it's a video-specific URL
2227 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2228 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2229 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2230 'video id', default=None)
2232 if self._downloader.params.get('noplaylist'):
2233 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2234 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2236 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2237 return video_id, None
2240 def _real_extract(self, url):
2241 # Extract playlist id
2242 mobj = re.match(self._VALID_URL, url)
2244 raise ExtractorError('Invalid URL: %s' % url)
2245 playlist_id = mobj.group(1) or mobj.group(2)
2247 video_id, video = self._check_download_just_video(url, playlist_id)
2251 if playlist_id.startswith(('RD', 'UL', 'PU')):
2252 # Mixes require a custom extraction process
2253 return self._extract_mix(playlist_id)
2255 has_videos, playlist = self._extract_playlist(playlist_id)
2256 if has_videos or not video_id:
2259 # Some playlist URLs don't actually serve a playlist (see
2260 # https://github.com/rg3/youtube-dl/issues/10537).
2261 # Fallback to plain video extraction if there is a video id
2262 # along with playlist id.
2263 return self.url_result(video_id, 'Youtube', video_id=video_id)
2266 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2267 IE_DESC = 'YouTube.com channels'
2268 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2269 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2270 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2271 IE_NAME = 'youtube:channel'
2273 'note': 'paginated channel',
2274 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2275 'playlist_mincount': 91,
2277 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2278 'title': 'Uploads from lex will',
2281 'note': 'Age restricted channel',
2282 # from https://www.youtube.com/user/DeusExOfficial
2283 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2284 'playlist_mincount': 64,
2286 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2287 'title': 'Uploads from Deus Ex',
2292 def suitable(cls, url):
2293 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2294 else super(YoutubeChannelIE, cls).suitable(url))
2296 def _build_template_url(self, url, channel_id):
2297 return self._TEMPLATE_URL % channel_id
2299 def _real_extract(self, url):
2300 channel_id = self._match_id(url)
2302 url = self._build_template_url(url, channel_id)
2304 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2305 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2306 # otherwise fallback on channel by page extraction
2307 channel_page = self._download_webpage(
2308 url + '?view=57', channel_id,
2309 'Downloading channel page', fatal=False)
2310 if channel_page is False:
2311 channel_playlist_id = False
2313 channel_playlist_id = self._html_search_meta(
2314 'channelId', channel_page, 'channel id', default=None)
2315 if not channel_playlist_id:
2316 channel_url = self._html_search_meta(
2317 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2318 channel_page, 'channel url', default=None)
2320 channel_playlist_id = self._search_regex(
2321 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2322 channel_url, 'channel id', default=None)
2323 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2324 playlist_id = 'UU' + channel_playlist_id[2:]
2325 return self.url_result(
2326 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2328 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2329 autogenerated = re.search(r'''(?x)
2331 channel-header-autogenerated-label|
2332 yt-channel-title-autogenerated
2333 )[^"]*"''', channel_page) is not None
2336 # The videos are contained in a single page
2337 # the ajax pages can't be used, they are empty
2340 video_id, 'Youtube', video_id=video_id,
2341 video_title=video_title)
2342 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2343 return self.playlist_result(entries, channel_id)
2346 next(self._entries(channel_page, channel_id))
2347 except StopIteration:
2348 alert_message = self._html_search_regex(
2349 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2350 channel_page, 'alert', default=None, group='alert')
2352 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2354 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2357 class YoutubeUserIE(YoutubeChannelIE):
2358 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2359 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2360 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2361 IE_NAME = 'youtube:user'
2364 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2365 'playlist_mincount': 320,
2367 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2368 'title': 'Uploads from The Linux Foundation',
2371 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2372 # but not https://www.youtube.com/user/12minuteathlete/videos
2373 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2374 'playlist_mincount': 249,
2376 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2377 'title': 'Uploads from 12 Minute Athlete',
2380 'url': 'ytuser:phihag',
2381 'only_matching': True,
2383 'url': 'https://www.youtube.com/c/gametrailers',
2384 'only_matching': True,
2386 'url': 'https://www.youtube.com/gametrailers',
2387 'only_matching': True,
2389 # This channel is not available, geo restricted to JP
2390 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2391 'only_matching': True,
2395 def suitable(cls, url):
2396 # Don't return True if the url can be extracted with other youtube
2397 # extractor, the regex would is too permissive and it would match.
2398 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2399 if any(ie.suitable(url) for ie in other_yt_ies):
2402 return super(YoutubeUserIE, cls).suitable(url)
2404 def _build_template_url(self, url, channel_id):
2405 mobj = re.match(self._VALID_URL, url)
2406 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2409 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2410 IE_DESC = 'YouTube.com live streams'
2411 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2412 IE_NAME = 'youtube:live'
2415 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2417 'id': 'a48o2S1cPoo',
2419 'title': 'The Young Turks - Live Main Show',
2420 'uploader': 'The Young Turks',
2421 'uploader_id': 'TheYoungTurks',
2422 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2423 'upload_date': '20150715',
2424 'license': 'Standard YouTube License',
2425 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2426 'categories': ['News & Politics'],
2427 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2429 'dislike_count': int,
2432 'skip_download': True,
2435 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2436 'only_matching': True,
2438 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2439 'only_matching': True,
2441 'url': 'https://www.youtube.com/TheYoungTurks/live',
2442 'only_matching': True,
2445 def _real_extract(self, url):
2446 mobj = re.match(self._VALID_URL, url)
2447 channel_id = mobj.group('id')
2448 base_url = mobj.group('base_url')
2449 webpage = self._download_webpage(url, channel_id, fatal=False)
2451 page_type = self._og_search_property(
2452 'type', webpage, 'page type', default=None)
2453 video_id = self._html_search_meta(
2454 'videoId', webpage, 'video id', default=None)
2455 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2456 return self.url_result(video_id, YoutubeIE.ie_key())
2457 return self.url_result(base_url)
2460 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2461 IE_DESC = 'YouTube.com user/channel playlists'
2462 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2463 IE_NAME = 'youtube:playlists'
2466 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2467 'playlist_mincount': 4,
2469 'id': 'ThirstForScience',
2470 'title': 'Thirst for Science',
2473 # with "Load more" button
2474 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2475 'playlist_mincount': 70,
2478 'title': 'Игорь Клейнер',
2481 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2482 'playlist_mincount': 17,
2484 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2485 'title': 'Chem Player',
2490 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2491 IE_DESC = 'YouTube.com searches'
2492 # there doesn't appear to be a real limit, for example if you search for
2493 # 'python' you get more than 8.000.000 results
2494 _MAX_RESULTS = float('inf')
2495 IE_NAME = 'youtube:search'
2496 _SEARCH_KEY = 'ytsearch'
2497 _EXTRA_QUERY_ARGS = {}
2500 def _get_n_results(self, query, n):
2501 """Get a specified number of results for a query"""
2507 'search_query': query.encode('utf-8'),
2509 url_query.update(self._EXTRA_QUERY_ARGS)
2510 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2512 for pagenum in itertools.count(1):
2513 data = self._download_json(
2514 result_url, video_id='query "%s"' % query,
2515 note='Downloading page %s' % pagenum,
2516 errnote='Unable to download API page',
2517 query={'spf': 'navigate'})
2518 html_content = data[1]['body']['content']
2520 if 'class="search-message' in html_content:
2521 raise ExtractorError(
2522 '[youtube] No video results', expected=True)
2524 new_videos = self._ids_to_results(orderedSet(re.findall(
2525 r'href="/watch\?v=(.{11})', html_content)))
2526 videos += new_videos
2527 if not new_videos or len(videos) > limit:
2529 next_link = self._html_search_regex(
2530 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2531 html_content, 'next link', default=None)
2532 if next_link is None:
2534 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2538 return self.playlist_result(videos, query)
2541 class YoutubeSearchDateIE(YoutubeSearchIE):
2542 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2543 _SEARCH_KEY = 'ytsearchdate'
2544 IE_DESC = 'YouTube.com searches, newest videos first'
2545 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2548 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2549 IE_DESC = 'YouTube.com search URLs'
2550 IE_NAME = 'youtube:search_url'
2551 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2552 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2554 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2555 'playlist_mincount': 5,
2557 'title': 'youtube-dl test video',
2560 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2561 'only_matching': True,
2564 def _real_extract(self, url):
2565 mobj = re.match(self._VALID_URL, url)
2566 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2567 webpage = self._download_webpage(url, query)
2568 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2571 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2572 IE_DESC = 'YouTube.com (multi-season) shows'
2573 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2574 IE_NAME = 'youtube:show'
2576 'url': 'https://www.youtube.com/show/airdisasters',
2577 'playlist_mincount': 5,
2579 'id': 'airdisasters',
2580 'title': 'Air Disasters',
2584 def _real_extract(self, url):
2585 playlist_id = self._match_id(url)
2586 return super(YoutubeShowIE, self)._real_extract(
2587 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2590 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2592 Base class for feed extractors
2593 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2595 _LOGIN_REQUIRED = True
2599 return 'youtube:%s' % self._FEED_NAME
2601 def _real_initialize(self):
2604 def _real_extract(self, url):
2605 page = self._download_webpage(
2606 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2608 # The extraction process is the same as for playlists, but the regex
2609 # for the video ids doesn't contain an index
2611 more_widget_html = content_html = page
2612 for page_num in itertools.count(1):
2613 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2615 # 'recommended' feed has infinite 'load more' and each new portion spins
2616 # the same videos in (sometimes) slightly different order, so we'll check
2617 # for unicity and break when portion has no new videos
2618 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2624 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2628 more = self._download_json(
2629 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2630 'Downloading page #%s' % page_num,
2631 transform_source=uppercase_escape)
2632 content_html = more['content_html']
2633 more_widget_html = more['load_more_widget_html']
2635 return self.playlist_result(
2636 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2639 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2640 IE_NAME = 'youtube:watchlater'
2641 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2642 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2645 'url': 'https://www.youtube.com/playlist?list=WL',
2646 'only_matching': True,
2648 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2649 'only_matching': True,
2652 def _real_extract(self, url):
2653 _, video = self._check_download_just_video(url, 'WL')
2656 _, playlist = self._extract_playlist('WL')
2660 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2661 IE_NAME = 'youtube:favorites'
2662 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2663 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2664 _LOGIN_REQUIRED = True
2666 def _real_extract(self, url):
2667 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2668 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2669 return self.url_result(playlist_id, 'YoutubePlaylist')
2672 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2673 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2674 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2675 _FEED_NAME = 'recommended'
2676 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2679 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2680 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2681 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2682 _FEED_NAME = 'subscriptions'
2683 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2686 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2687 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2688 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2689 _FEED_NAME = 'history'
2690 _PLAYLIST_TITLE = 'Youtube History'
2693 class YoutubeTruncatedURLIE(InfoExtractor):
2694 IE_NAME = 'youtube:truncated_url'
2695 IE_DESC = False # Do not list
2696 _VALID_URL = r'''(?x)
2698 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2701 annotation_id=annotation_[^&]+|
2707 attribution_link\?a=[^&]+
2713 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2714 'only_matching': True,
2716 'url': 'https://www.youtube.com/watch?',
2717 'only_matching': True,
2719 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2720 'only_matching': True,
2722 'url': 'https://www.youtube.com/watch?feature=foo',
2723 'only_matching': True,
2725 'url': 'https://www.youtube.com/watch?hl=en-GB',
2726 'only_matching': True,
2728 'url': 'https://www.youtube.com/watch?t=2372',
2729 'only_matching': True,
2732 def _real_extract(self, url):
2733 raise ExtractorError(
2734 'Did you forget to quote the URL? Remember that & is a meta '
2735 'character in most shells, so you want to put the URL in quotes, '
2737 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2738 ' or simply youtube-dl BaW_jenozKc .',
2742 class YoutubeTruncatedIDIE(InfoExtractor):
2743 IE_NAME = 'youtube:truncated_id'
2744 IE_DESC = False # Do not list
2745 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2748 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2749 'only_matching': True,
2752 def _real_extract(self, url):
2753 video_id = self._match_id(url)
2754 raise ExtractorError(
2755 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),