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)
676 # YouTube Red ad is not captured for creator
678 'url': '__2ABJjxzNo',
683 'upload_date': '20100430',
684 'uploader_id': 'deadmau5',
685 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
686 'creator': 'deadmau5',
687 'description': 'md5:12c56784b8032162bb936a5f76d55360',
688 'uploader': 'deadmau5',
689 'license': 'Standard YouTube License',
690 'title': 'Deadmau5 - Some Chords (HD)',
691 'alt_title': 'Some Chords',
693 'expected_warnings': [
694 'DASH manifest missing',
697 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
699 'url': 'lqQg6PlCWgI',
704 'upload_date': '20150827',
705 'uploader_id': 'olympic',
706 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
707 'license': 'Standard YouTube License',
708 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
709 'uploader': 'Olympic',
710 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
713 'skip_download': 'requires avconv',
718 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
722 'stretched_ratio': 16 / 9.,
724 'upload_date': '20110310',
725 'uploader_id': 'AllenMeow',
726 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
727 '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',
729 'license': 'Standard YouTube License',
730 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
733 # url_encoded_fmt_stream_map is empty string
735 'url': 'qEJwOuvDf7I',
739 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
741 'upload_date': '20150404',
742 'uploader_id': 'spbelect',
743 'uploader': 'Наблюдатели Петербурга',
746 'skip_download': 'requires avconv',
748 'skip': 'This live event has ended.',
750 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
752 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
756 'title': 'md5:7b81415841e02ecd4313668cde88737a',
757 'description': 'md5:116377fd2963b81ec4ce64b542173306',
759 'upload_date': '20150625',
760 'uploader_id': 'dorappi2000',
761 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
762 'uploader': 'dorappi2000',
763 'license': 'Standard YouTube License',
764 'formats': 'mincount:32',
767 # DASH manifest with segment_list
769 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
770 'md5': '8ce563a1d667b599d21064e982ab9e31',
774 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
775 'uploader': 'Airtek',
776 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
777 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
778 'license': 'Standard YouTube License',
779 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
782 'youtube_include_dash_manifest': True,
783 'format': '135', # bestvideo
785 'skip': 'This live event has ended.',
788 # Multifeed videos (multiple cameras), URL is for Main Camera
789 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
792 'title': 'teamPGP: Rocket League Noob Stream',
793 'description': 'md5:dc7872fb300e143831327f1bae3af010',
799 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
800 'description': 'md5:dc7872fb300e143831327f1bae3af010',
802 'upload_date': '20150721',
803 'uploader': 'Beer Games Beer',
804 'uploader_id': 'beergamesbeer',
805 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
806 'license': 'Standard YouTube License',
812 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
813 'description': 'md5:dc7872fb300e143831327f1bae3af010',
815 'upload_date': '20150721',
816 'uploader': 'Beer Games Beer',
817 'uploader_id': 'beergamesbeer',
818 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
819 'license': 'Standard YouTube License',
825 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
826 'description': 'md5:dc7872fb300e143831327f1bae3af010',
828 'upload_date': '20150721',
829 'uploader': 'Beer Games Beer',
830 'uploader_id': 'beergamesbeer',
831 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
832 'license': 'Standard YouTube License',
838 'title': 'teamPGP: Rocket League Noob Stream (zim)',
839 'description': 'md5:dc7872fb300e143831327f1bae3af010',
841 'upload_date': '20150721',
842 'uploader': 'Beer Games Beer',
843 'uploader_id': 'beergamesbeer',
844 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
845 'license': 'Standard YouTube License',
849 'skip_download': True,
853 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
854 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
857 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
860 'skip': 'Not multifeed anymore',
863 'url': 'https://vid.plus/FlRa-iH7PGw',
864 'only_matching': True,
867 'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
868 'only_matching': True,
871 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
872 # Also tests cut-off URL expansion in video description (see
873 # https://github.com/rg3/youtube-dl/issues/1892,
874 # https://github.com/rg3/youtube-dl/issues/8164)
875 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
879 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
880 'alt_title': 'Dark Walk',
881 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
883 'upload_date': '20151119',
884 'uploader_id': 'IronSoulElf',
885 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
886 'uploader': 'IronSoulElf',
887 'license': 'Standard YouTube License',
888 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
891 'skip_download': True,
895 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
896 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
897 'only_matching': True,
900 # Video with yt:stretch=17:0
901 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
905 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
906 'description': 'md5:ee18a25c350637c8faff806845bddee9',
907 'upload_date': '20151107',
908 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
909 'uploader': 'CH GAMER DROID',
912 'skip_download': True,
914 'skip': 'This video does not exist.',
917 # Video licensed under Creative Commons
918 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
922 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
923 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
925 'upload_date': '20150127',
926 'uploader_id': 'BerkmanCenter',
927 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
928 'uploader': 'The Berkman Klein Center for Internet & Society',
929 'license': 'Creative Commons Attribution license (reuse allowed)',
932 'skip_download': True,
936 # Channel-like uploader_url
937 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
941 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
942 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
944 'upload_date': '20151119',
945 'uploader': 'Bernie 2016',
946 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
947 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
948 'license': 'Creative Commons Attribution license (reuse allowed)',
951 'skip_download': True,
955 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
956 'only_matching': True,
959 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
960 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
961 'only_matching': True,
964 # Rental video preview
965 'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
969 'title': 'Piku - Trailer',
970 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
971 'upload_date': '20150811',
972 'uploader': 'FlixMatrix',
973 'uploader_id': 'FlixMatrixKaravan',
974 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
975 'license': 'Standard YouTube License',
978 'skip_download': True,
982 # YouTube Red video with episode data
983 'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
987 'title': 'Isolation - Mind Field (Ep 1)',
988 'description': 'md5:8013b7ddea787342608f63a13ddc9492',
990 'upload_date': '20170118',
991 'uploader': 'Vsauce',
992 'uploader_id': 'Vsauce',
993 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
994 'license': 'Standard YouTube License',
995 'series': 'Mind Field',
1000 'skip_download': True,
1002 'expected_warnings': [
1003 'Skipping DASH manifest',
1008 'url': '1t24XAntNCY',
1009 'only_matching': True,
1012 # geo restricted to JP
1013 'url': 'sJL6WA-aGkQ',
1014 'only_matching': True,
1017 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1018 'only_matching': True,
1022 def __init__(self, *args, **kwargs):
1023 super(YoutubeIE, self).__init__(*args, **kwargs)
1024 self._player_cache = {}
1026 def report_video_info_webpage_download(self, video_id):
1027 """Report attempt to download video info webpage."""
1028 self.to_screen('%s: Downloading video info webpage' % video_id)
1030 def report_information_extraction(self, video_id):
1031 """Report attempt to extract video information."""
1032 self.to_screen('%s: Extracting video information' % video_id)
1034 def report_unavailable_format(self, video_id, format):
1035 """Report extracted video URL."""
1036 self.to_screen('%s: Format %s not available' % (video_id, format))
1038 def report_rtmp_download(self):
1039 """Indicate the download will use the RTMP protocol."""
1040 self.to_screen('RTMP download detected')
1042 def _signature_cache_id(self, example_sig):
1043 """ Return a string representation of a signature """
1044 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1046 def _extract_signature_function(self, video_id, player_url, example_sig):
1048 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1051 raise ExtractorError('Cannot identify player %r' % player_url)
1052 player_type = id_m.group('ext')
1053 player_id = id_m.group('id')
1055 # Read from filesystem cache
1056 func_id = '%s_%s_%s' % (
1057 player_type, player_id, self._signature_cache_id(example_sig))
1058 assert os.path.basename(func_id) == func_id
1060 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1061 if cache_spec is not None:
1062 return lambda s: ''.join(s[i] for i in cache_spec)
1065 'Downloading player %s' % player_url
1066 if self._downloader.params.get('verbose') else
1067 'Downloading %s player %s' % (player_type, player_id)
1069 if player_type == 'js':
1070 code = self._download_webpage(
1071 player_url, video_id,
1073 errnote='Download of %s failed' % player_url)
1074 res = self._parse_sig_js(code)
1075 elif player_type == 'swf':
1076 urlh = self._request_webpage(
1077 player_url, video_id,
1079 errnote='Download of %s failed' % player_url)
1081 res = self._parse_sig_swf(code)
1083 assert False, 'Invalid player type %r' % player_type
1085 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1086 cache_res = res(test_string)
1087 cache_spec = [ord(c) for c in cache_res]
1089 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1092 def _print_sig_code(self, func, example_sig):
1093 def gen_sig_code(idxs):
1094 def _genslice(start, end, step):
1095 starts = '' if start == 0 else str(start)
1096 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1097 steps = '' if step == 1 else (':%d' % step)
1098 return 's[%s%s%s]' % (starts, ends, steps)
1101 # Quelch pyflakes warnings - start will be set when step is set
1102 start = '(Never used)'
1103 for i, prev in zip(idxs[1:], idxs[:-1]):
1104 if step is not None:
1105 if i - prev == step:
1107 yield _genslice(start, prev, step)
1110 if i - prev in [-1, 1]:
1115 yield 's[%d]' % prev
1119 yield _genslice(start, i, step)
1121 test_string = ''.join(map(compat_chr, range(len(example_sig))))
1122 cache_res = func(test_string)
1123 cache_spec = [ord(c) for c in cache_res]
1124 expr_code = ' + '.join(gen_sig_code(cache_spec))
1125 signature_id_tuple = '(%s)' % (
1126 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1127 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1128 ' return %s\n') % (signature_id_tuple, expr_code)
1129 self.to_screen('Extracted signature function:\n' + code)
1131 def _parse_sig_js(self, jscode):
1132 funcname = self._search_regex(
1133 (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1134 r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\('),
1135 jscode, 'Initial JS player signature function name', group='sig')
1137 jsi = JSInterpreter(jscode)
1138 initial_function = jsi.extract_function(funcname)
1139 return lambda s: initial_function([s])
1141 def _parse_sig_swf(self, file_contents):
1142 swfi = SWFInterpreter(file_contents)
1143 TARGET_CLASSNAME = 'SignatureDecipher'
1144 searched_class = swfi.extract_class(TARGET_CLASSNAME)
1145 initial_function = swfi.extract_function(searched_class, 'decipher')
1146 return lambda s: initial_function([s])
1148 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1149 """Turn the encrypted s field into a working signature"""
1151 if player_url is None:
1152 raise ExtractorError('Cannot decrypt signature without player_url')
1154 if player_url.startswith('//'):
1155 player_url = 'https:' + player_url
1156 elif not re.match(r'https?://', player_url):
1157 player_url = compat_urlparse.urljoin(
1158 'https://www.youtube.com', player_url)
1160 player_id = (player_url, self._signature_cache_id(s))
1161 if player_id not in self._player_cache:
1162 func = self._extract_signature_function(
1163 video_id, player_url, s
1165 self._player_cache[player_id] = func
1166 func = self._player_cache[player_id]
1167 if self._downloader.params.get('youtube_print_sig_code'):
1168 self._print_sig_code(func, s)
1170 except Exception as e:
1171 tb = traceback.format_exc()
1172 raise ExtractorError(
1173 'Signature extraction failed: ' + tb, cause=e)
1175 def _get_subtitles(self, video_id, webpage):
1177 subs_doc = self._download_xml(
1178 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1179 video_id, note=False)
1180 except ExtractorError as err:
1181 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1185 for track in subs_doc.findall('track'):
1186 lang = track.attrib['lang_code']
1187 if lang in sub_lang_list:
1190 for ext in self._SUBTITLE_FORMATS:
1191 params = compat_urllib_parse_urlencode({
1195 'name': track.attrib['name'].encode('utf-8'),
1197 sub_formats.append({
1198 'url': 'https://www.youtube.com/api/timedtext?' + params,
1201 sub_lang_list[lang] = sub_formats
1202 if not sub_lang_list:
1203 self._downloader.report_warning('video doesn\'t have subtitles')
1205 return sub_lang_list
1207 def _get_ytplayer_config(self, video_id, webpage):
1209 # User data may contain arbitrary character sequences that may affect
1210 # JSON extraction with regex, e.g. when '};' is contained the second
1211 # regex won't capture the whole JSON. Yet working around by trying more
1212 # concrete regex first keeping in mind proper quoted string handling
1213 # to be implemented in future that will replace this workaround (see
1214 # https://github.com/rg3/youtube-dl/issues/7468,
1215 # https://github.com/rg3/youtube-dl/pull/7599)
1216 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1217 r';ytplayer\.config\s*=\s*({.+?});',
1219 config = self._search_regex(
1220 patterns, webpage, 'ytplayer.config', default=None)
1222 return self._parse_json(
1223 uppercase_escape(config), video_id, fatal=False)
1225 def _get_automatic_captions(self, video_id, webpage):
1226 """We need the webpage for getting the captions url, pass it as an
1227 argument to speed up the process."""
1228 self.to_screen('%s: Looking for automatic captions' % video_id)
1229 player_config = self._get_ytplayer_config(video_id, webpage)
1230 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1231 if not player_config:
1232 self._downloader.report_warning(err_msg)
1235 args = player_config['args']
1236 caption_url = args.get('ttsurl')
1238 timestamp = args['timestamp']
1239 # We get the available subtitles
1240 list_params = compat_urllib_parse_urlencode({
1245 list_url = caption_url + '&' + list_params
1246 caption_list = self._download_xml(list_url, video_id)
1247 original_lang_node = caption_list.find('track')
1248 if original_lang_node is None:
1249 self._downloader.report_warning('Video doesn\'t have automatic captions')
1251 original_lang = original_lang_node.attrib['lang_code']
1252 caption_kind = original_lang_node.attrib.get('kind', '')
1255 for lang_node in caption_list.findall('target'):
1256 sub_lang = lang_node.attrib['lang_code']
1258 for ext in self._SUBTITLE_FORMATS:
1259 params = compat_urllib_parse_urlencode({
1260 'lang': original_lang,
1264 'kind': caption_kind,
1266 sub_formats.append({
1267 'url': caption_url + '&' + params,
1270 sub_lang_list[sub_lang] = sub_formats
1271 return sub_lang_list
1273 def make_captions(sub_url, sub_langs):
1274 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1275 caption_qs = compat_parse_qs(parsed_sub_url.query)
1277 for sub_lang in sub_langs:
1279 for ext in self._SUBTITLE_FORMATS:
1281 'tlang': [sub_lang],
1284 sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1285 query=compat_urllib_parse_urlencode(caption_qs, True)))
1286 sub_formats.append({
1290 captions[sub_lang] = sub_formats
1293 # New captions format as of 22.06.2017
1294 player_response = args.get('player_response')
1295 if player_response and isinstance(player_response, compat_str):
1296 player_response = self._parse_json(
1297 player_response, video_id, fatal=False)
1299 renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1300 base_url = renderer['captionTracks'][0]['baseUrl']
1302 for lang in renderer['translationLanguages']:
1303 lang_code = lang.get('languageCode')
1305 sub_lang_list.append(lang_code)
1306 return make_captions(base_url, sub_lang_list)
1308 # Some videos don't provide ttsurl but rather caption_tracks and
1309 # caption_translation_languages (e.g. 20LmZk1hakA)
1310 # Does not used anymore as of 22.06.2017
1311 caption_tracks = args['caption_tracks']
1312 caption_translation_languages = args['caption_translation_languages']
1313 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1315 for lang in caption_translation_languages.split(','):
1316 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1317 sub_lang = lang_qs.get('lc', [None])[0]
1319 sub_lang_list.append(sub_lang)
1320 return make_captions(caption_url, sub_lang_list)
1321 # An extractor error can be raise by the download process if there are
1322 # no automatic captions but there are subtitles
1323 except (KeyError, IndexError, ExtractorError):
1324 self._downloader.report_warning(err_msg)
1327 def _mark_watched(self, video_id, video_info):
1328 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1329 if not playback_url:
1331 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1332 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1334 # cpn generation algorithm is reverse engineered from base.js.
1335 # In fact it works even with dummy cpn.
1336 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1337 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1343 playback_url = compat_urlparse.urlunparse(
1344 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1346 self._download_webpage(
1347 playback_url, video_id, 'Marking watched',
1348 'Unable to mark watched', fatal=False)
1351 def extract_id(cls, url):
1352 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1354 raise ExtractorError('Invalid URL: %s' % url)
1355 video_id = mobj.group(2)
1358 def _extract_annotations(self, video_id):
1359 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1360 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1363 def _extract_chapters(description, duration):
1366 chapter_lines = re.findall(
1367 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1369 if not chapter_lines:
1372 for next_num, (chapter_line, time_point) in enumerate(
1373 chapter_lines, start=1):
1374 start_time = parse_duration(time_point)
1375 if start_time is None:
1377 if start_time > duration:
1379 end_time = (duration if next_num == len(chapter_lines)
1380 else parse_duration(chapter_lines[next_num][1]))
1381 if end_time is None:
1383 if end_time > duration:
1385 if start_time > end_time:
1387 chapter_title = re.sub(
1388 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1389 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1391 'start_time': start_time,
1392 'end_time': end_time,
1393 'title': chapter_title,
1397 def _real_extract(self, url):
1398 url, smuggled_data = unsmuggle_url(url, {})
1401 'http' if self._downloader.params.get('prefer_insecure', False)
1406 parsed_url = compat_urllib_parse_urlparse(url)
1407 for component in [parsed_url.fragment, parsed_url.query]:
1408 query = compat_parse_qs(component)
1409 if start_time is None and 't' in query:
1410 start_time = parse_duration(query['t'][0])
1411 if start_time is None and 'start' in query:
1412 start_time = parse_duration(query['start'][0])
1413 if end_time is None and 'end' in query:
1414 end_time = parse_duration(query['end'][0])
1416 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1417 mobj = re.search(self._NEXT_URL_RE, url)
1419 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1420 video_id = self.extract_id(url)
1423 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1424 video_webpage = self._download_webpage(url, video_id)
1426 # Attempt to extract SWF player URL
1427 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1428 if mobj is not None:
1429 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1435 def add_dash_mpd(video_info):
1436 dash_mpd = video_info.get('dashmpd')
1437 if dash_mpd and dash_mpd[0] not in dash_mpds:
1438 dash_mpds.append(dash_mpd[0])
1441 embed_webpage = None
1443 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1445 # We simulate the access to the video from www.youtube.com/v/{video_id}
1446 # this can be viewed without login into Youtube
1447 url = proto + '://www.youtube.com/embed/%s' % video_id
1448 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1449 data = compat_urllib_parse_urlencode({
1450 'video_id': video_id,
1451 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1452 'sts': self._search_regex(
1453 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1455 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1456 video_info_webpage = self._download_webpage(
1457 video_info_url, video_id,
1458 note='Refetching age-gated info webpage',
1459 errnote='unable to download video info webpage')
1460 video_info = compat_parse_qs(video_info_webpage)
1461 add_dash_mpd(video_info)
1466 # Try looking directly into the video webpage
1467 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1469 args = ytplayer_config['args']
1470 if args.get('url_encoded_fmt_stream_map'):
1471 # Convert to the same format returned by compat_parse_qs
1472 video_info = dict((k, [v]) for k, v in args.items())
1473 add_dash_mpd(video_info)
1474 # Rental video is not rented but preview is available (e.g.
1475 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1476 # https://github.com/rg3/youtube-dl/issues/10532)
1477 if not video_info and args.get('ypc_vid'):
1478 return self.url_result(
1479 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1480 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1482 sts = ytplayer_config.get('sts')
1483 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1484 # We also try looking in get_video_info since it may contain different dashmpd
1485 # URL that points to a DASH manifest with possibly different itag set (some itags
1486 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1487 # manifest pointed by get_video_info's dashmpd).
1488 # The general idea is to take a union of itags of both DASH manifests (for example
1489 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1490 self.report_video_info_webpage_download(video_id)
1491 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1493 'video_id': video_id,
1503 video_info_webpage = self._download_webpage(
1504 '%s://www.youtube.com/get_video_info' % proto,
1505 video_id, note=False,
1506 errnote='unable to download video info webpage',
1507 fatal=False, query=query)
1508 if not video_info_webpage:
1510 get_video_info = compat_parse_qs(video_info_webpage)
1511 add_dash_mpd(get_video_info)
1513 video_info = get_video_info
1514 if 'token' in get_video_info:
1515 # Different get_video_info requests may report different results, e.g.
1516 # some may report video unavailability, but some may serve it without
1517 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1518 # the original webpage as well as el=info and el=embedded get_video_info
1519 # requests report video unavailability due to geo restriction while
1520 # el=detailpage succeeds and returns valid data). This is probably
1521 # due to YouTube measures against IP ranges of hosting providers.
1522 # Working around by preferring the first succeeded video_info containing
1523 # the token if no such video_info yet was found.
1524 if 'token' not in video_info:
1525 video_info = get_video_info
1527 if 'token' not in video_info:
1528 if 'reason' in video_info:
1529 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1530 regions_allowed = self._html_search_meta(
1531 'regionsAllowed', video_webpage, default=None)
1532 countries = regions_allowed.split(',') if regions_allowed else None
1533 self.raise_geo_restricted(
1534 msg=video_info['reason'][0], countries=countries)
1535 raise ExtractorError(
1536 'YouTube said: %s' % video_info['reason'][0],
1537 expected=True, video_id=video_id)
1539 raise ExtractorError(
1540 '"token" parameter not in video info for unknown reason',
1544 if 'title' in video_info:
1545 video_title = video_info['title'][0]
1547 self._downloader.report_warning('Unable to extract video title')
1551 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1552 if video_description:
1553 description_original = video_description = re.sub(r'''(?x)
1555 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1556 (?:title|href)="([^"]+)"\s+
1557 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1561 ''', r'\1', video_description)
1562 video_description = clean_html(video_description)
1564 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1566 video_description = unescapeHTML(fd_mobj.group(1))
1568 video_description = ''
1570 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1571 if not self._downloader.params.get('noplaylist'):
1574 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1575 for feed in multifeed_metadata_list.split(','):
1576 # Unquote should take place before split on comma (,) since textual
1577 # fields may contain comma as well (see
1578 # https://github.com/rg3/youtube-dl/issues/8536)
1579 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1581 '_type': 'url_transparent',
1582 'ie_key': 'Youtube',
1584 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1585 {'force_singlefeed': True}),
1586 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1588 feed_ids.append(feed_data['id'][0])
1590 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1591 % (', '.join(feed_ids), video_id))
1592 return self.playlist_result(entries, video_id, video_title, video_description)
1593 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1595 if 'view_count' in video_info:
1596 view_count = int(video_info['view_count'][0])
1600 # Check for "rental" videos
1601 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1602 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1604 # Start extracting information
1605 self.report_information_extraction(video_id)
1608 if 'author' not in video_info:
1609 raise ExtractorError('Unable to extract uploader name')
1610 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1613 video_uploader_id = None
1614 video_uploader_url = None
1616 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1618 if mobj is not None:
1619 video_uploader_id = mobj.group('uploader_id')
1620 video_uploader_url = mobj.group('uploader_url')
1622 self._downloader.report_warning('unable to extract uploader nickname')
1625 # We try first to get a high quality image:
1626 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1627 video_webpage, re.DOTALL)
1628 if m_thumb is not None:
1629 video_thumbnail = m_thumb.group(1)
1630 elif 'thumbnail_url' not in video_info:
1631 self._downloader.report_warning('unable to extract video thumbnail')
1632 video_thumbnail = None
1633 else: # don't panic if we can't find it
1634 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1637 upload_date = self._html_search_meta(
1638 'datePublished', video_webpage, 'upload date', default=None)
1640 upload_date = self._search_regex(
1641 [r'(?s)id="eow-date.*?>(.*?)</span>',
1642 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1643 video_webpage, 'upload date', default=None)
1645 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1646 upload_date = unified_strdate(upload_date)
1648 video_license = self._html_search_regex(
1649 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1650 video_webpage, 'license', default=None)
1652 m_music = re.search(
1654 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
1662 \bhref=["\']/red[^>]*>| # drop possible
1663 >\s*Listen ad-free with YouTube Red # YouTube Red ad
1670 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1671 video_creator = clean_html(m_music.group('creator'))
1673 video_alt_title = video_creator = None
1675 m_episode = re.search(
1676 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1679 series = m_episode.group('series')
1680 season_number = int(m_episode.group('season'))
1681 episode_number = int(m_episode.group('episode'))
1683 series = season_number = episode_number = None
1685 m_cat_container = self._search_regex(
1686 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1687 video_webpage, 'categories', default=None)
1689 category = self._html_search_regex(
1690 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1692 video_categories = None if category is None else [category]
1694 video_categories = None
1697 unescapeHTML(m.group('content'))
1698 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1700 def _extract_count(count_name):
1701 return str_to_int(self._search_regex(
1702 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1703 % re.escape(count_name),
1704 video_webpage, count_name, default=None))
1706 like_count = _extract_count('like')
1707 dislike_count = _extract_count('dislike')
1710 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1711 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1713 video_duration = try_get(
1714 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1715 if not video_duration:
1716 video_duration = parse_duration(self._html_search_meta(
1717 'duration', video_webpage, 'video duration'))
1720 video_annotations = None
1721 if self._downloader.params.get('writeannotations', False):
1722 video_annotations = self._extract_annotations(video_id)
1724 chapters = self._extract_chapters(description_original, video_duration)
1726 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1727 self.report_rtmp_download()
1729 'format_id': '_rtmp',
1731 'url': video_info['conn'][0],
1732 'player_url': player_url,
1734 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1735 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1736 if 'rtmpe%3Dyes' in encoded_url_map:
1737 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1739 fmt_list = video_info.get('fmt_list', [''])[0]
1741 for fmt in fmt_list.split(','):
1742 spec = fmt.split('/')
1744 width_height = spec[1].split('x')
1745 if len(width_height) == 2:
1746 formats_spec[spec[0]] = {
1747 'resolution': spec[1],
1748 'width': int_or_none(width_height[0]),
1749 'height': int_or_none(width_height[1]),
1752 for url_data_str in encoded_url_map.split(','):
1753 url_data = compat_parse_qs(url_data_str)
1754 if 'itag' not in url_data or 'url' not in url_data:
1756 format_id = url_data['itag'][0]
1757 url = url_data['url'][0]
1759 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1760 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1761 jsplayer_url_json = self._search_regex(
1763 embed_webpage if age_gate else video_webpage,
1764 'JS player URL (1)', default=None)
1765 if not jsplayer_url_json and not age_gate:
1766 # We need the embed website after all
1767 if embed_webpage is None:
1768 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1769 embed_webpage = self._download_webpage(
1770 embed_url, video_id, 'Downloading embed webpage')
1771 jsplayer_url_json = self._search_regex(
1772 ASSETS_RE, embed_webpage, 'JS player URL')
1774 player_url = json.loads(jsplayer_url_json)
1775 if player_url is None:
1776 player_url_json = self._search_regex(
1777 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1778 video_webpage, 'age gate player URL')
1779 player_url = json.loads(player_url_json)
1781 if 'sig' in url_data:
1782 url += '&signature=' + url_data['sig'][0]
1783 elif 's' in url_data:
1784 encrypted_sig = url_data['s'][0]
1786 if self._downloader.params.get('verbose'):
1787 if player_url is None:
1788 player_version = 'unknown'
1789 player_desc = 'unknown'
1791 if player_url.endswith('swf'):
1792 player_version = self._search_regex(
1793 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1794 'flash player', fatal=False)
1795 player_desc = 'flash player %s' % player_version
1797 player_version = self._search_regex(
1798 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1799 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1801 'html5 player', fatal=False)
1802 player_desc = 'html5 player %s' % player_version
1804 parts_sizes = self._signature_cache_id(encrypted_sig)
1805 self.to_screen('{%s} signature length %s, %s' %
1806 (format_id, parts_sizes, player_desc))
1808 signature = self._decrypt_signature(
1809 encrypted_sig, video_id, player_url, age_gate)
1810 url += '&signature=' + signature
1811 if 'ratebypass' not in url:
1812 url += '&ratebypass=yes'
1815 'format_id': format_id,
1817 'player_url': player_url,
1819 if format_id in self._formats:
1820 dct.update(self._formats[format_id])
1821 if format_id in formats_spec:
1822 dct.update(formats_spec[format_id])
1824 # Some itags are not included in DASH manifest thus corresponding formats will
1825 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1826 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1827 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1828 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1831 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1832 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1835 'fps': int_or_none(url_data.get('fps', [None])[0]),
1836 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1838 for key, value in more_fields.items():
1841 type_ = url_data.get('type', [None])[0]
1843 type_split = type_.split(';')
1844 kind_ext = type_split[0].split('/')
1845 if len(kind_ext) == 2:
1847 dct['ext'] = mimetype2ext(type_split[0])
1848 if kind in ('audio', 'video'):
1850 for mobj in re.finditer(
1851 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1852 if mobj.group('key') == 'codecs':
1853 codecs = mobj.group('val')
1856 dct.update(parse_codecs(codecs))
1858 elif video_info.get('hlsvp'):
1859 manifest_url = video_info['hlsvp'][0]
1861 m3u8_formats = self._extract_m3u8_formats(
1862 manifest_url, video_id, 'mp4', fatal=False)
1863 for a_format in m3u8_formats:
1864 itag = self._search_regex(
1865 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1867 a_format['format_id'] = itag
1868 if itag in self._formats:
1869 dct = self._formats[itag].copy()
1870 dct.update(a_format)
1872 a_format['player_url'] = player_url
1873 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1874 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1875 formats.append(a_format)
1877 unavailable_message = self._html_search_regex(
1878 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1879 video_webpage, 'unavailable message', default=None)
1880 if unavailable_message:
1881 raise ExtractorError(unavailable_message, expected=True)
1882 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1884 # Look for the DASH manifest
1885 if self._downloader.params.get('youtube_include_dash_manifest', True):
1886 dash_mpd_fatal = True
1887 for mpd_url in dash_mpds:
1890 def decrypt_sig(mobj):
1892 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1893 return '/signature/%s' % dec_s
1895 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1897 for df in self._extract_mpd_formats(
1898 mpd_url, video_id, fatal=dash_mpd_fatal,
1899 formats_dict=self._formats):
1900 # Do not overwrite DASH format found in some previous DASH manifest
1901 if df['format_id'] not in dash_formats:
1902 dash_formats[df['format_id']] = df
1903 # Additional DASH manifests may end up in HTTP Error 403 therefore
1904 # allow them to fail without bug report message if we already have
1905 # some DASH manifest succeeded. This is temporary workaround to reduce
1906 # burst of bug reports until we figure out the reason and whether it
1907 # can be fixed at all.
1908 dash_mpd_fatal = False
1909 except (ExtractorError, KeyError) as e:
1910 self.report_warning(
1911 'Skipping DASH manifest: %r' % e, video_id)
1913 # Remove the formats we found through non-DASH, they
1914 # contain less info and it can be wrong, because we use
1915 # fixed values (for example the resolution). See
1916 # https://github.com/rg3/youtube-dl/issues/5774 for an
1918 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1919 formats.extend(dash_formats.values())
1921 # Check for malformed aspect ratio
1922 stretched_m = re.search(
1923 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1926 w = float(stretched_m.group('w'))
1927 h = float(stretched_m.group('h'))
1928 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1929 # We will only process correct ratios.
1933 if f.get('vcodec') != 'none':
1934 f['stretched_ratio'] = ratio
1936 self._sort_formats(formats)
1938 self.mark_watched(video_id, video_info)
1942 'uploader': video_uploader,
1943 'uploader_id': video_uploader_id,
1944 'uploader_url': video_uploader_url,
1945 'upload_date': upload_date,
1946 'license': video_license,
1947 'creator': video_creator,
1948 'title': video_title,
1949 'alt_title': video_alt_title,
1950 'thumbnail': video_thumbnail,
1951 'description': video_description,
1952 'categories': video_categories,
1954 'subtitles': video_subtitles,
1955 'automatic_captions': automatic_captions,
1956 'duration': video_duration,
1957 'age_limit': 18 if age_gate else 0,
1958 'annotations': video_annotations,
1959 'chapters': chapters,
1960 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1961 'view_count': view_count,
1962 'like_count': like_count,
1963 'dislike_count': dislike_count,
1964 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1967 'start_time': start_time,
1968 'end_time': end_time,
1970 'season_number': season_number,
1971 'episode_number': episode_number,
1975 class YoutubeSharedVideoIE(InfoExtractor):
1976 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1977 IE_NAME = 'youtube:shared'
1980 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1982 'id': 'uPDB5I9wfp8',
1984 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1985 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1986 'upload_date': '20160219',
1987 'uploader': 'Pocoyo - Português (BR)',
1988 'uploader_id': 'PocoyoBrazil',
1990 'add_ie': ['Youtube'],
1992 # There are already too many Youtube downloads
1993 'skip_download': True,
1997 def _real_extract(self, url):
1998 video_id = self._match_id(url)
2000 webpage = self._download_webpage(url, video_id)
2002 real_video_id = self._html_search_meta(
2003 'videoId', webpage, 'YouTube video id', fatal=True)
2005 return self.url_result(real_video_id, YoutubeIE.ie_key())
2008 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2009 IE_DESC = 'YouTube.com playlists'
2010 _VALID_URL = r"""(?x)(?:
2016 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2017 \? (?:.*?[&;])*? (?:p|a|list)=
2020 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2023 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
2024 # Top tracks, they can also include dots
2030 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2031 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
2032 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2033 IE_NAME = 'youtube:playlist'
2035 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2037 'title': 'ytdl test PL',
2038 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2040 'playlist_count': 3,
2042 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2044 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2045 'title': 'YDL_Empty_List',
2047 'playlist_count': 0,
2048 'skip': 'This playlist is private',
2050 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2051 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2053 'title': '29C3: Not my department',
2054 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2056 'playlist_count': 95,
2058 'note': 'issue #673',
2059 'url': 'PLBB231211A4F62143',
2061 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2062 'id': 'PLBB231211A4F62143',
2064 'playlist_mincount': 26,
2066 'note': 'Large playlist',
2067 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2069 'title': 'Uploads from Cauchemar',
2070 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2072 'playlist_mincount': 799,
2074 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2076 'title': 'YDL_safe_search',
2077 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2079 'playlist_count': 2,
2080 'skip': 'This playlist is private',
2083 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2084 'playlist_count': 4,
2087 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2090 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2091 'playlist_mincount': 485,
2093 'title': '2017 華語最新單曲 (2/24更新)',
2094 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2097 'note': 'Embedded SWF player',
2098 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2099 'playlist_count': 4,
2102 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2105 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2106 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2108 'title': 'Uploads from Interstellar Movie',
2109 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2111 'playlist_mincount': 21,
2113 # Playlist URL that does not actually serve a playlist
2114 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2116 'id': 'FqZTN594JQw',
2118 'title': "Smiley's People 01 detective, Adventure Series, Action",
2119 'uploader': 'STREEM',
2120 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2121 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2122 'upload_date': '20150526',
2123 'license': 'Standard YouTube License',
2124 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2125 'categories': ['People & Blogs'],
2128 'dislike_count': int,
2131 'skip_download': True,
2133 'add_ie': [YoutubeIE.ie_key()],
2135 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2137 'id': 'yeWKywCrFtk',
2139 'title': 'Small Scale Baler and Braiding Rugs',
2140 'uploader': 'Backus-Page House Museum',
2141 'uploader_id': 'backuspagemuseum',
2142 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2143 'upload_date': '20161008',
2144 'license': 'Standard YouTube License',
2145 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2146 'categories': ['Nonprofits & Activism'],
2149 'dislike_count': int,
2153 'skip_download': True,
2156 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2157 'only_matching': True,
2159 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2160 'only_matching': True,
2163 def _real_initialize(self):
2166 def _extract_mix(self, playlist_id):
2167 # The mixes are generated from a single video
2168 # the id of the playlist is just 'RD' + video_id
2170 last_id = playlist_id[-11:]
2171 for n in itertools.count(1):
2172 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2173 webpage = self._download_webpage(
2174 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2175 new_ids = orderedSet(re.findall(
2176 r'''(?xs)data-video-username=".*?".*?
2177 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2179 # Fetch new pages until all the videos are repeated, it seems that
2180 # there are always 51 unique videos.
2181 new_ids = [_id for _id in new_ids if _id not in ids]
2187 url_results = self._ids_to_results(ids)
2189 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2191 search_title('playlist-title') or
2192 search_title('title long-title') or
2193 search_title('title'))
2194 title = clean_html(title_span)
2196 return self.playlist_result(url_results, playlist_id, title)
2198 def _extract_playlist(self, playlist_id):
2199 url = self._TEMPLATE_URL % playlist_id
2200 page = self._download_webpage(url, playlist_id)
2202 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2203 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2204 match = match.strip()
2205 # Check if the playlist exists or is private
2206 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2208 reason = mobj.group('reason')
2209 message = 'This playlist %s' % reason
2210 if 'private' in reason:
2211 message += ', use --username or --netrc to access it'
2213 raise ExtractorError(message, expected=True)
2214 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2215 raise ExtractorError(
2216 'Invalid parameters. Maybe URL is incorrect.',
2218 elif re.match(r'[^<]*Choose your language[^<]*', match):
2221 self.report_warning('Youtube gives an alert message: ' + match)
2223 playlist_title = self._html_search_regex(
2224 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2225 page, 'title', default=None)
2229 if not playlist_title:
2231 # Some playlist URLs don't actually serve a playlist (e.g.
2232 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2233 next(self._entries(page, playlist_id))
2234 except StopIteration:
2237 return has_videos, self.playlist_result(
2238 self._entries(page, playlist_id), playlist_id, playlist_title)
2240 def _check_download_just_video(self, url, playlist_id):
2241 # Check if it's a video-specific URL
2242 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2243 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2244 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2245 'video id', default=None)
2247 if self._downloader.params.get('noplaylist'):
2248 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2249 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2251 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2252 return video_id, None
2255 def _real_extract(self, url):
2256 # Extract playlist id
2257 mobj = re.match(self._VALID_URL, url)
2259 raise ExtractorError('Invalid URL: %s' % url)
2260 playlist_id = mobj.group(1) or mobj.group(2)
2262 video_id, video = self._check_download_just_video(url, playlist_id)
2266 if playlist_id.startswith(('RD', 'UL', 'PU')):
2267 # Mixes require a custom extraction process
2268 return self._extract_mix(playlist_id)
2270 has_videos, playlist = self._extract_playlist(playlist_id)
2271 if has_videos or not video_id:
2274 # Some playlist URLs don't actually serve a playlist (see
2275 # https://github.com/rg3/youtube-dl/issues/10537).
2276 # Fallback to plain video extraction if there is a video id
2277 # along with playlist id.
2278 return self.url_result(video_id, 'Youtube', video_id=video_id)
2281 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2282 IE_DESC = 'YouTube.com channels'
2283 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2284 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2285 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2286 IE_NAME = 'youtube:channel'
2288 'note': 'paginated channel',
2289 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2290 'playlist_mincount': 91,
2292 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2293 'title': 'Uploads from lex will',
2296 'note': 'Age restricted channel',
2297 # from https://www.youtube.com/user/DeusExOfficial
2298 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2299 'playlist_mincount': 64,
2301 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2302 'title': 'Uploads from Deus Ex',
2307 def suitable(cls, url):
2308 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2309 else super(YoutubeChannelIE, cls).suitable(url))
2311 def _build_template_url(self, url, channel_id):
2312 return self._TEMPLATE_URL % channel_id
2314 def _real_extract(self, url):
2315 channel_id = self._match_id(url)
2317 url = self._build_template_url(url, channel_id)
2319 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2320 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2321 # otherwise fallback on channel by page extraction
2322 channel_page = self._download_webpage(
2323 url + '?view=57', channel_id,
2324 'Downloading channel page', fatal=False)
2325 if channel_page is False:
2326 channel_playlist_id = False
2328 channel_playlist_id = self._html_search_meta(
2329 'channelId', channel_page, 'channel id', default=None)
2330 if not channel_playlist_id:
2331 channel_url = self._html_search_meta(
2332 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2333 channel_page, 'channel url', default=None)
2335 channel_playlist_id = self._search_regex(
2336 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2337 channel_url, 'channel id', default=None)
2338 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2339 playlist_id = 'UU' + channel_playlist_id[2:]
2340 return self.url_result(
2341 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2343 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2344 autogenerated = re.search(r'''(?x)
2346 channel-header-autogenerated-label|
2347 yt-channel-title-autogenerated
2348 )[^"]*"''', channel_page) is not None
2351 # The videos are contained in a single page
2352 # the ajax pages can't be used, they are empty
2355 video_id, 'Youtube', video_id=video_id,
2356 video_title=video_title)
2357 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2358 return self.playlist_result(entries, channel_id)
2361 next(self._entries(channel_page, channel_id))
2362 except StopIteration:
2363 alert_message = self._html_search_regex(
2364 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2365 channel_page, 'alert', default=None, group='alert')
2367 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2369 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2372 class YoutubeUserIE(YoutubeChannelIE):
2373 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2374 _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_-]+)'
2375 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2376 IE_NAME = 'youtube:user'
2379 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2380 'playlist_mincount': 320,
2382 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2383 'title': 'Uploads from The Linux Foundation',
2386 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2387 # but not https://www.youtube.com/user/12minuteathlete/videos
2388 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2389 'playlist_mincount': 249,
2391 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2392 'title': 'Uploads from 12 Minute Athlete',
2395 'url': 'ytuser:phihag',
2396 'only_matching': True,
2398 'url': 'https://www.youtube.com/c/gametrailers',
2399 'only_matching': True,
2401 'url': 'https://www.youtube.com/gametrailers',
2402 'only_matching': True,
2404 # This channel is not available, geo restricted to JP
2405 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2406 'only_matching': True,
2410 def suitable(cls, url):
2411 # Don't return True if the url can be extracted with other youtube
2412 # extractor, the regex would is too permissive and it would match.
2413 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2414 if any(ie.suitable(url) for ie in other_yt_ies):
2417 return super(YoutubeUserIE, cls).suitable(url)
2419 def _build_template_url(self, url, channel_id):
2420 mobj = re.match(self._VALID_URL, url)
2421 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2424 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2425 IE_DESC = 'YouTube.com live streams'
2426 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2427 IE_NAME = 'youtube:live'
2430 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2432 'id': 'a48o2S1cPoo',
2434 'title': 'The Young Turks - Live Main Show',
2435 'uploader': 'The Young Turks',
2436 'uploader_id': 'TheYoungTurks',
2437 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2438 'upload_date': '20150715',
2439 'license': 'Standard YouTube License',
2440 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2441 'categories': ['News & Politics'],
2442 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2444 'dislike_count': int,
2447 'skip_download': True,
2450 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2451 'only_matching': True,
2453 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2454 'only_matching': True,
2456 'url': 'https://www.youtube.com/TheYoungTurks/live',
2457 'only_matching': True,
2460 def _real_extract(self, url):
2461 mobj = re.match(self._VALID_URL, url)
2462 channel_id = mobj.group('id')
2463 base_url = mobj.group('base_url')
2464 webpage = self._download_webpage(url, channel_id, fatal=False)
2466 page_type = self._og_search_property(
2467 'type', webpage, 'page type', default=None)
2468 video_id = self._html_search_meta(
2469 'videoId', webpage, 'video id', default=None)
2470 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2471 return self.url_result(video_id, YoutubeIE.ie_key())
2472 return self.url_result(base_url)
2475 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2476 IE_DESC = 'YouTube.com user/channel playlists'
2477 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2478 IE_NAME = 'youtube:playlists'
2481 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2482 'playlist_mincount': 4,
2484 'id': 'ThirstForScience',
2485 'title': 'Thirst for Science',
2488 # with "Load more" button
2489 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2490 'playlist_mincount': 70,
2493 'title': 'Игорь Клейнер',
2496 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2497 'playlist_mincount': 17,
2499 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2500 'title': 'Chem Player',
2505 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2506 IE_DESC = 'YouTube.com searches'
2507 # there doesn't appear to be a real limit, for example if you search for
2508 # 'python' you get more than 8.000.000 results
2509 _MAX_RESULTS = float('inf')
2510 IE_NAME = 'youtube:search'
2511 _SEARCH_KEY = 'ytsearch'
2512 _EXTRA_QUERY_ARGS = {}
2515 def _get_n_results(self, query, n):
2516 """Get a specified number of results for a query"""
2522 'search_query': query.encode('utf-8'),
2524 url_query.update(self._EXTRA_QUERY_ARGS)
2525 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2527 for pagenum in itertools.count(1):
2528 data = self._download_json(
2529 result_url, video_id='query "%s"' % query,
2530 note='Downloading page %s' % pagenum,
2531 errnote='Unable to download API page',
2532 query={'spf': 'navigate'})
2533 html_content = data[1]['body']['content']
2535 if 'class="search-message' in html_content:
2536 raise ExtractorError(
2537 '[youtube] No video results', expected=True)
2539 new_videos = self._ids_to_results(orderedSet(re.findall(
2540 r'href="/watch\?v=(.{11})', html_content)))
2541 videos += new_videos
2542 if not new_videos or len(videos) > limit:
2544 next_link = self._html_search_regex(
2545 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2546 html_content, 'next link', default=None)
2547 if next_link is None:
2549 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2553 return self.playlist_result(videos, query)
2556 class YoutubeSearchDateIE(YoutubeSearchIE):
2557 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2558 _SEARCH_KEY = 'ytsearchdate'
2559 IE_DESC = 'YouTube.com searches, newest videos first'
2560 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2563 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2564 IE_DESC = 'YouTube.com search URLs'
2565 IE_NAME = 'youtube:search_url'
2566 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2567 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2569 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2570 'playlist_mincount': 5,
2572 'title': 'youtube-dl test video',
2575 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2576 'only_matching': True,
2579 def _real_extract(self, url):
2580 mobj = re.match(self._VALID_URL, url)
2581 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2582 webpage = self._download_webpage(url, query)
2583 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2586 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2587 IE_DESC = 'YouTube.com (multi-season) shows'
2588 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2589 IE_NAME = 'youtube:show'
2591 'url': 'https://www.youtube.com/show/airdisasters',
2592 'playlist_mincount': 5,
2594 'id': 'airdisasters',
2595 'title': 'Air Disasters',
2599 def _real_extract(self, url):
2600 playlist_id = self._match_id(url)
2601 return super(YoutubeShowIE, self)._real_extract(
2602 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2605 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2607 Base class for feed extractors
2608 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2610 _LOGIN_REQUIRED = True
2614 return 'youtube:%s' % self._FEED_NAME
2616 def _real_initialize(self):
2619 def _real_extract(self, url):
2620 page = self._download_webpage(
2621 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2623 # The extraction process is the same as for playlists, but the regex
2624 # for the video ids doesn't contain an index
2626 more_widget_html = content_html = page
2627 for page_num in itertools.count(1):
2628 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2630 # 'recommended' feed has infinite 'load more' and each new portion spins
2631 # the same videos in (sometimes) slightly different order, so we'll check
2632 # for unicity and break when portion has no new videos
2633 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2639 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2643 more = self._download_json(
2644 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2645 'Downloading page #%s' % page_num,
2646 transform_source=uppercase_escape)
2647 content_html = more['content_html']
2648 more_widget_html = more['load_more_widget_html']
2650 return self.playlist_result(
2651 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2654 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2655 IE_NAME = 'youtube:watchlater'
2656 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2657 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2660 'url': 'https://www.youtube.com/playlist?list=WL',
2661 'only_matching': True,
2663 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2664 'only_matching': True,
2667 def _real_extract(self, url):
2668 _, video = self._check_download_just_video(url, 'WL')
2671 _, playlist = self._extract_playlist('WL')
2675 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2676 IE_NAME = 'youtube:favorites'
2677 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2678 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2679 _LOGIN_REQUIRED = True
2681 def _real_extract(self, url):
2682 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2683 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2684 return self.url_result(playlist_id, 'YoutubePlaylist')
2687 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2688 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2689 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2690 _FEED_NAME = 'recommended'
2691 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2694 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2695 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2696 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2697 _FEED_NAME = 'subscriptions'
2698 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2701 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2702 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2703 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2704 _FEED_NAME = 'history'
2705 _PLAYLIST_TITLE = 'Youtube History'
2708 class YoutubeTruncatedURLIE(InfoExtractor):
2709 IE_NAME = 'youtube:truncated_url'
2710 IE_DESC = False # Do not list
2711 _VALID_URL = r'''(?x)
2713 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2716 annotation_id=annotation_[^&]+|
2722 attribution_link\?a=[^&]+
2728 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2729 'only_matching': True,
2731 'url': 'https://www.youtube.com/watch?',
2732 'only_matching': True,
2734 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2735 'only_matching': True,
2737 'url': 'https://www.youtube.com/watch?feature=foo',
2738 'only_matching': True,
2740 'url': 'https://www.youtube.com/watch?hl=en-GB',
2741 'only_matching': True,
2743 'url': 'https://www.youtube.com/watch?t=2372',
2744 'only_matching': True,
2747 def _real_extract(self, url):
2748 raise ExtractorError(
2749 'Did you forget to quote the URL? Remember that & is a meta '
2750 'character in most shells, so you want to put the URL in quotes, '
2752 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2753 ' or simply youtube-dl BaW_jenozKc .',
2757 class YoutubeTruncatedIDIE(InfoExtractor):
2758 IE_NAME = 'youtube:truncated_id'
2759 IE_DESC = False # Do not list
2760 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2763 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2764 'only_matching': True,
2767 def _real_extract(self, url):
2768 video_id = self._match_id(url)
2769 raise ExtractorError(
2770 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),