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 # Some videos don't provide ttsurl but rather caption_tracks and
1273 # caption_translation_languages (e.g. 20LmZk1hakA)
1274 caption_tracks = args['caption_tracks']
1275 caption_translation_languages = args['caption_translation_languages']
1276 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1277 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1278 caption_qs = compat_parse_qs(parsed_caption_url.query)
1281 for lang in caption_translation_languages.split(','):
1282 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1283 sub_lang = lang_qs.get('lc', [None])[0]
1287 for ext in self._SUBTITLE_FORMATS:
1289 'tlang': [sub_lang],
1292 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1293 query=compat_urllib_parse_urlencode(caption_qs, True)))
1294 sub_formats.append({
1298 sub_lang_list[sub_lang] = sub_formats
1299 return sub_lang_list
1300 # An extractor error can be raise by the download process if there are
1301 # no automatic captions but there are subtitles
1302 except (KeyError, ExtractorError):
1303 self._downloader.report_warning(err_msg)
1306 def _mark_watched(self, video_id, video_info):
1307 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1308 if not playback_url:
1310 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1311 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1313 # cpn generation algorithm is reverse engineered from base.js.
1314 # In fact it works even with dummy cpn.
1315 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1316 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1322 playback_url = compat_urlparse.urlunparse(
1323 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1325 self._download_webpage(
1326 playback_url, video_id, 'Marking watched',
1327 'Unable to mark watched', fatal=False)
1330 def extract_id(cls, url):
1331 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1333 raise ExtractorError('Invalid URL: %s' % url)
1334 video_id = mobj.group(2)
1337 def _extract_annotations(self, video_id):
1338 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1339 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1342 def _extract_chapters(description, duration):
1345 chapter_lines = re.findall(
1346 r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1348 if not chapter_lines:
1351 for next_num, (chapter_line, time_point) in enumerate(
1352 chapter_lines, start=1):
1353 start_time = parse_duration(time_point)
1354 if start_time is None:
1356 end_time = (duration if next_num == len(chapter_lines)
1357 else parse_duration(chapter_lines[next_num][1]))
1358 if end_time is None:
1360 chapter_title = re.sub(
1361 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1362 chapter_title = re.sub(r'\s+', ' ', chapter_title)
1364 'start_time': start_time,
1365 'end_time': end_time,
1366 'title': chapter_title,
1370 def _real_extract(self, url):
1371 url, smuggled_data = unsmuggle_url(url, {})
1374 'http' if self._downloader.params.get('prefer_insecure', False)
1379 parsed_url = compat_urllib_parse_urlparse(url)
1380 for component in [parsed_url.fragment, parsed_url.query]:
1381 query = compat_parse_qs(component)
1382 if start_time is None and 't' in query:
1383 start_time = parse_duration(query['t'][0])
1384 if start_time is None and 'start' in query:
1385 start_time = parse_duration(query['start'][0])
1386 if end_time is None and 'end' in query:
1387 end_time = parse_duration(query['end'][0])
1389 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1390 mobj = re.search(self._NEXT_URL_RE, url)
1392 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1393 video_id = self.extract_id(url)
1396 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1397 video_webpage = self._download_webpage(url, video_id)
1399 # Attempt to extract SWF player URL
1400 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1401 if mobj is not None:
1402 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1408 def add_dash_mpd(video_info):
1409 dash_mpd = video_info.get('dashmpd')
1410 if dash_mpd and dash_mpd[0] not in dash_mpds:
1411 dash_mpds.append(dash_mpd[0])
1414 embed_webpage = None
1416 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1418 # We simulate the access to the video from www.youtube.com/v/{video_id}
1419 # this can be viewed without login into Youtube
1420 url = proto + '://www.youtube.com/embed/%s' % video_id
1421 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1422 data = compat_urllib_parse_urlencode({
1423 'video_id': video_id,
1424 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1425 'sts': self._search_regex(
1426 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1428 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1429 video_info_webpage = self._download_webpage(
1430 video_info_url, video_id,
1431 note='Refetching age-gated info webpage',
1432 errnote='unable to download video info webpage')
1433 video_info = compat_parse_qs(video_info_webpage)
1434 add_dash_mpd(video_info)
1439 # Try looking directly into the video webpage
1440 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1442 args = ytplayer_config['args']
1443 if args.get('url_encoded_fmt_stream_map'):
1444 # Convert to the same format returned by compat_parse_qs
1445 video_info = dict((k, [v]) for k, v in args.items())
1446 add_dash_mpd(video_info)
1447 # Rental video is not rented but preview is available (e.g.
1448 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1449 # https://github.com/rg3/youtube-dl/issues/10532)
1450 if not video_info and args.get('ypc_vid'):
1451 return self.url_result(
1452 args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1453 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1455 sts = ytplayer_config.get('sts')
1456 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1457 # We also try looking in get_video_info since it may contain different dashmpd
1458 # URL that points to a DASH manifest with possibly different itag set (some itags
1459 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1460 # manifest pointed by get_video_info's dashmpd).
1461 # The general idea is to take a union of itags of both DASH manifests (for example
1462 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1463 self.report_video_info_webpage_download(video_id)
1464 for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
1466 'video_id': video_id,
1476 video_info_webpage = self._download_webpage(
1477 '%s://www.youtube.com/get_video_info' % proto,
1478 video_id, note=False,
1479 errnote='unable to download video info webpage',
1480 fatal=False, query=query)
1481 if not video_info_webpage:
1483 get_video_info = compat_parse_qs(video_info_webpage)
1484 add_dash_mpd(get_video_info)
1486 video_info = get_video_info
1487 if 'token' in get_video_info:
1488 # Different get_video_info requests may report different results, e.g.
1489 # some may report video unavailability, but some may serve it without
1490 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1491 # the original webpage as well as el=info and el=embedded get_video_info
1492 # requests report video unavailability due to geo restriction while
1493 # el=detailpage succeeds and returns valid data). This is probably
1494 # due to YouTube measures against IP ranges of hosting providers.
1495 # Working around by preferring the first succeeded video_info containing
1496 # the token if no such video_info yet was found.
1497 if 'token' not in video_info:
1498 video_info = get_video_info
1500 if 'token' not in video_info:
1501 if 'reason' in video_info:
1502 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1503 regions_allowed = self._html_search_meta(
1504 'regionsAllowed', video_webpage, default=None)
1505 countries = regions_allowed.split(',') if regions_allowed else None
1506 self.raise_geo_restricted(
1507 msg=video_info['reason'][0], countries=countries)
1508 raise ExtractorError(
1509 'YouTube said: %s' % video_info['reason'][0],
1510 expected=True, video_id=video_id)
1512 raise ExtractorError(
1513 '"token" parameter not in video info for unknown reason',
1517 if 'title' in video_info:
1518 video_title = video_info['title'][0]
1520 self._downloader.report_warning('Unable to extract video title')
1524 description_original = video_description = get_element_by_id("eow-description", video_webpage)
1525 if video_description:
1526 description_original = video_description = re.sub(r'''(?x)
1528 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1529 (?:title|href)="([^"]+)"\s+
1530 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1534 ''', r'\1', video_description)
1535 video_description = clean_html(video_description)
1537 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1539 video_description = unescapeHTML(fd_mobj.group(1))
1541 video_description = ''
1543 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1544 if not self._downloader.params.get('noplaylist'):
1547 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1548 for feed in multifeed_metadata_list.split(','):
1549 # Unquote should take place before split on comma (,) since textual
1550 # fields may contain comma as well (see
1551 # https://github.com/rg3/youtube-dl/issues/8536)
1552 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1554 '_type': 'url_transparent',
1555 'ie_key': 'Youtube',
1557 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1558 {'force_singlefeed': True}),
1559 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1561 feed_ids.append(feed_data['id'][0])
1563 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1564 % (', '.join(feed_ids), video_id))
1565 return self.playlist_result(entries, video_id, video_title, video_description)
1566 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1568 if 'view_count' in video_info:
1569 view_count = int(video_info['view_count'][0])
1573 # Check for "rental" videos
1574 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1575 raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
1577 # Start extracting information
1578 self.report_information_extraction(video_id)
1581 if 'author' not in video_info:
1582 raise ExtractorError('Unable to extract uploader name')
1583 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1586 video_uploader_id = None
1587 video_uploader_url = None
1589 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1591 if mobj is not None:
1592 video_uploader_id = mobj.group('uploader_id')
1593 video_uploader_url = mobj.group('uploader_url')
1595 self._downloader.report_warning('unable to extract uploader nickname')
1598 # We try first to get a high quality image:
1599 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1600 video_webpage, re.DOTALL)
1601 if m_thumb is not None:
1602 video_thumbnail = m_thumb.group(1)
1603 elif 'thumbnail_url' not in video_info:
1604 self._downloader.report_warning('unable to extract video thumbnail')
1605 video_thumbnail = None
1606 else: # don't panic if we can't find it
1607 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1610 upload_date = self._html_search_meta(
1611 'datePublished', video_webpage, 'upload date', default=None)
1613 upload_date = self._search_regex(
1614 [r'(?s)id="eow-date.*?>(.*?)</span>',
1615 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1616 video_webpage, 'upload date', default=None)
1618 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1619 upload_date = unified_strdate(upload_date)
1621 video_license = self._html_search_regex(
1622 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1623 video_webpage, 'license', default=None)
1625 m_music = re.search(
1626 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1629 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1630 video_creator = clean_html(m_music.group('creator'))
1632 video_alt_title = video_creator = None
1634 m_episode = re.search(
1635 r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
1638 series = m_episode.group('series')
1639 season_number = int(m_episode.group('season'))
1640 episode_number = int(m_episode.group('episode'))
1642 series = season_number = episode_number = None
1644 m_cat_container = self._search_regex(
1645 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1646 video_webpage, 'categories', default=None)
1648 category = self._html_search_regex(
1649 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1651 video_categories = None if category is None else [category]
1653 video_categories = None
1656 unescapeHTML(m.group('content'))
1657 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1659 def _extract_count(count_name):
1660 return str_to_int(self._search_regex(
1661 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1662 % re.escape(count_name),
1663 video_webpage, count_name, default=None))
1665 like_count = _extract_count('like')
1666 dislike_count = _extract_count('dislike')
1669 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1670 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1672 video_duration = try_get(
1673 video_info, lambda x: int_or_none(x['length_seconds'][0]))
1674 if not video_duration:
1675 video_duration = parse_duration(self._html_search_meta(
1676 'duration', video_webpage, 'video duration'))
1679 video_annotations = None
1680 if self._downloader.params.get('writeannotations', False):
1681 video_annotations = self._extract_annotations(video_id)
1683 chapters = self._extract_chapters(description_original, video_duration)
1685 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1686 self.report_rtmp_download()
1688 'format_id': '_rtmp',
1690 'url': video_info['conn'][0],
1691 'player_url': player_url,
1693 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1694 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1695 if 'rtmpe%3Dyes' in encoded_url_map:
1696 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1698 fmt_list = video_info.get('fmt_list', [''])[0]
1700 for fmt in fmt_list.split(','):
1701 spec = fmt.split('/')
1703 width_height = spec[1].split('x')
1704 if len(width_height) == 2:
1705 formats_spec[spec[0]] = {
1706 'resolution': spec[1],
1707 'width': int_or_none(width_height[0]),
1708 'height': int_or_none(width_height[1]),
1711 for url_data_str in encoded_url_map.split(','):
1712 url_data = compat_parse_qs(url_data_str)
1713 if 'itag' not in url_data or 'url' not in url_data:
1715 format_id = url_data['itag'][0]
1716 url = url_data['url'][0]
1718 if 'sig' in url_data:
1719 url += '&signature=' + url_data['sig'][0]
1720 elif 's' in url_data:
1721 encrypted_sig = url_data['s'][0]
1722 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1724 jsplayer_url_json = self._search_regex(
1726 embed_webpage if age_gate else video_webpage,
1727 'JS player URL (1)', default=None)
1728 if not jsplayer_url_json and not age_gate:
1729 # We need the embed website after all
1730 if embed_webpage is None:
1731 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1732 embed_webpage = self._download_webpage(
1733 embed_url, video_id, 'Downloading embed webpage')
1734 jsplayer_url_json = self._search_regex(
1735 ASSETS_RE, embed_webpage, 'JS player URL')
1737 player_url = json.loads(jsplayer_url_json)
1738 if player_url is None:
1739 player_url_json = self._search_regex(
1740 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1741 video_webpage, 'age gate player URL')
1742 player_url = json.loads(player_url_json)
1744 if self._downloader.params.get('verbose'):
1745 if player_url is None:
1746 player_version = 'unknown'
1747 player_desc = 'unknown'
1749 if player_url.endswith('swf'):
1750 player_version = self._search_regex(
1751 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1752 'flash player', fatal=False)
1753 player_desc = 'flash player %s' % player_version
1755 player_version = self._search_regex(
1756 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1757 r'(?:www|player)-([^/]+)(?:/[a-z]{2}_[A-Z]{2})?/base\.js'],
1759 'html5 player', fatal=False)
1760 player_desc = 'html5 player %s' % player_version
1762 parts_sizes = self._signature_cache_id(encrypted_sig)
1763 self.to_screen('{%s} signature length %s, %s' %
1764 (format_id, parts_sizes, player_desc))
1766 signature = self._decrypt_signature(
1767 encrypted_sig, video_id, player_url, age_gate)
1768 url += '&signature=' + signature
1769 if 'ratebypass' not in url:
1770 url += '&ratebypass=yes'
1773 'format_id': format_id,
1775 'player_url': player_url,
1777 if format_id in self._formats:
1778 dct.update(self._formats[format_id])
1779 if format_id in formats_spec:
1780 dct.update(formats_spec[format_id])
1782 # Some itags are not included in DASH manifest thus corresponding formats will
1783 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1784 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1785 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1786 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1789 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1790 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1793 'fps': int_or_none(url_data.get('fps', [None])[0]),
1794 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1796 for key, value in more_fields.items():
1799 type_ = url_data.get('type', [None])[0]
1801 type_split = type_.split(';')
1802 kind_ext = type_split[0].split('/')
1803 if len(kind_ext) == 2:
1805 dct['ext'] = mimetype2ext(type_split[0])
1806 if kind in ('audio', 'video'):
1808 for mobj in re.finditer(
1809 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1810 if mobj.group('key') == 'codecs':
1811 codecs = mobj.group('val')
1814 dct.update(parse_codecs(codecs))
1816 elif video_info.get('hlsvp'):
1817 manifest_url = video_info['hlsvp'][0]
1819 m3u8_formats = self._extract_m3u8_formats(
1820 manifest_url, video_id, 'mp4', fatal=False)
1821 for a_format in m3u8_formats:
1822 itag = self._search_regex(
1823 r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
1825 a_format['format_id'] = itag
1826 if itag in self._formats:
1827 dct = self._formats[itag].copy()
1828 dct.update(a_format)
1830 a_format['player_url'] = player_url
1831 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1832 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1833 formats.append(a_format)
1835 unavailable_message = self._html_search_regex(
1836 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1837 video_webpage, 'unavailable message', default=None)
1838 if unavailable_message:
1839 raise ExtractorError(unavailable_message, expected=True)
1840 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1842 # Look for the DASH manifest
1843 if self._downloader.params.get('youtube_include_dash_manifest', True):
1844 dash_mpd_fatal = True
1845 for mpd_url in dash_mpds:
1848 def decrypt_sig(mobj):
1850 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1851 return '/signature/%s' % dec_s
1853 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1855 for df in self._extract_mpd_formats(
1856 mpd_url, video_id, fatal=dash_mpd_fatal,
1857 formats_dict=self._formats):
1858 # Do not overwrite DASH format found in some previous DASH manifest
1859 if df['format_id'] not in dash_formats:
1860 dash_formats[df['format_id']] = df
1861 # Additional DASH manifests may end up in HTTP Error 403 therefore
1862 # allow them to fail without bug report message if we already have
1863 # some DASH manifest succeeded. This is temporary workaround to reduce
1864 # burst of bug reports until we figure out the reason and whether it
1865 # can be fixed at all.
1866 dash_mpd_fatal = False
1867 except (ExtractorError, KeyError) as e:
1868 self.report_warning(
1869 'Skipping DASH manifest: %r' % e, video_id)
1871 # Remove the formats we found through non-DASH, they
1872 # contain less info and it can be wrong, because we use
1873 # fixed values (for example the resolution). See
1874 # https://github.com/rg3/youtube-dl/issues/5774 for an
1876 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1877 formats.extend(dash_formats.values())
1879 # Check for malformed aspect ratio
1880 stretched_m = re.search(
1881 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1884 w = float(stretched_m.group('w'))
1885 h = float(stretched_m.group('h'))
1886 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1887 # We will only process correct ratios.
1891 if f.get('vcodec') != 'none':
1892 f['stretched_ratio'] = ratio
1894 self._sort_formats(formats)
1896 self.mark_watched(video_id, video_info)
1900 'uploader': video_uploader,
1901 'uploader_id': video_uploader_id,
1902 'uploader_url': video_uploader_url,
1903 'upload_date': upload_date,
1904 'license': video_license,
1905 'creator': video_creator,
1906 'title': video_title,
1907 'alt_title': video_alt_title,
1908 'thumbnail': video_thumbnail,
1909 'description': video_description,
1910 'categories': video_categories,
1912 'subtitles': video_subtitles,
1913 'automatic_captions': automatic_captions,
1914 'duration': video_duration,
1915 'age_limit': 18 if age_gate else 0,
1916 'annotations': video_annotations,
1917 'chapters': chapters,
1918 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1919 'view_count': view_count,
1920 'like_count': like_count,
1921 'dislike_count': dislike_count,
1922 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1925 'start_time': start_time,
1926 'end_time': end_time,
1928 'season_number': season_number,
1929 'episode_number': episode_number,
1933 class YoutubeSharedVideoIE(InfoExtractor):
1934 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1935 IE_NAME = 'youtube:shared'
1938 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1940 'id': 'uPDB5I9wfp8',
1942 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1943 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1944 'upload_date': '20160219',
1945 'uploader': 'Pocoyo - Português (BR)',
1946 'uploader_id': 'PocoyoBrazil',
1948 'add_ie': ['Youtube'],
1950 # There are already too many Youtube downloads
1951 'skip_download': True,
1955 def _real_extract(self, url):
1956 video_id = self._match_id(url)
1958 webpage = self._download_webpage(url, video_id)
1960 real_video_id = self._html_search_meta(
1961 'videoId', webpage, 'YouTube video id', fatal=True)
1963 return self.url_result(real_video_id, YoutubeIE.ie_key())
1966 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1967 IE_DESC = 'YouTube.com playlists'
1968 _VALID_URL = r"""(?x)(?:
1974 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
1975 \? (?:.*?[&;])*? (?:p|a|list)=
1978 youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
1981 (?:PL|LL|EC|UU|FL|RD|UL|TL)?[0-9A-Za-z-_]{10,}
1982 # Top tracks, they can also include dots
1988 )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
1989 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
1990 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1991 IE_NAME = 'youtube:playlist'
1993 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1995 'title': 'ytdl test PL',
1996 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1998 'playlist_count': 3,
2000 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2002 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2003 'title': 'YDL_Empty_List',
2005 'playlist_count': 0,
2006 'skip': 'This playlist is private',
2008 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2009 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2011 'title': '29C3: Not my department',
2012 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2014 'playlist_count': 95,
2016 'note': 'issue #673',
2017 'url': 'PLBB231211A4F62143',
2019 'title': '[OLD]Team Fortress 2 (Class-based LP)',
2020 'id': 'PLBB231211A4F62143',
2022 'playlist_mincount': 26,
2024 'note': 'Large playlist',
2025 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2027 'title': 'Uploads from Cauchemar',
2028 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2030 'playlist_mincount': 799,
2032 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2034 'title': 'YDL_safe_search',
2035 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2037 'playlist_count': 2,
2038 'skip': 'This playlist is private',
2041 'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2042 'playlist_count': 4,
2045 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2048 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2049 'playlist_mincount': 485,
2051 'title': '2017 華語最新單曲 (2/24更新)',
2052 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2055 'note': 'Embedded SWF player',
2056 'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2057 'playlist_count': 4,
2060 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2063 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2064 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2066 'title': 'Uploads from Interstellar Movie',
2067 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2069 'playlist_mincount': 21,
2071 # Playlist URL that does not actually serve a playlist
2072 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2074 'id': 'FqZTN594JQw',
2076 'title': "Smiley's People 01 detective, Adventure Series, Action",
2077 'uploader': 'STREEM',
2078 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2079 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2080 'upload_date': '20150526',
2081 'license': 'Standard YouTube License',
2082 'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2083 'categories': ['People & Blogs'],
2086 'dislike_count': int,
2089 'skip_download': True,
2091 'add_ie': [YoutubeIE.ie_key()],
2093 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2095 'id': 'yeWKywCrFtk',
2097 'title': 'Small Scale Baler and Braiding Rugs',
2098 'uploader': 'Backus-Page House Museum',
2099 'uploader_id': 'backuspagemuseum',
2100 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2101 'upload_date': '20161008',
2102 'license': 'Standard YouTube License',
2103 'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2104 'categories': ['Nonprofits & Activism'],
2107 'dislike_count': int,
2111 'skip_download': True,
2114 'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2115 'only_matching': True,
2117 'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2118 'only_matching': True,
2121 def _real_initialize(self):
2124 def _extract_mix(self, playlist_id):
2125 # The mixes are generated from a single video
2126 # the id of the playlist is just 'RD' + video_id
2128 last_id = playlist_id[-11:]
2129 for n in itertools.count(1):
2130 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2131 webpage = self._download_webpage(
2132 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2133 new_ids = orderedSet(re.findall(
2134 r'''(?xs)data-video-username=".*?".*?
2135 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
2137 # Fetch new pages until all the videos are repeated, it seems that
2138 # there are always 51 unique videos.
2139 new_ids = [_id for _id in new_ids if _id not in ids]
2145 url_results = self._ids_to_results(ids)
2147 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2149 search_title('playlist-title') or
2150 search_title('title long-title') or
2151 search_title('title'))
2152 title = clean_html(title_span)
2154 return self.playlist_result(url_results, playlist_id, title)
2156 def _extract_playlist(self, playlist_id):
2157 url = self._TEMPLATE_URL % playlist_id
2158 page = self._download_webpage(url, playlist_id)
2160 # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
2161 for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2162 match = match.strip()
2163 # Check if the playlist exists or is private
2164 mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2166 reason = mobj.group('reason')
2167 message = 'This playlist %s' % reason
2168 if 'private' in reason:
2169 message += ', use --username or --netrc to access it'
2171 raise ExtractorError(message, expected=True)
2172 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2173 raise ExtractorError(
2174 'Invalid parameters. Maybe URL is incorrect.',
2176 elif re.match(r'[^<]*Choose your language[^<]*', match):
2179 self.report_warning('Youtube gives an alert message: ' + match)
2181 playlist_title = self._html_search_regex(
2182 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2183 page, 'title', default=None)
2187 if not playlist_title:
2189 # Some playlist URLs don't actually serve a playlist (e.g.
2190 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2191 next(self._entries(page, playlist_id))
2192 except StopIteration:
2195 return has_videos, self.playlist_result(
2196 self._entries(page, playlist_id), playlist_id, playlist_title)
2198 def _check_download_just_video(self, url, playlist_id):
2199 # Check if it's a video-specific URL
2200 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2201 video_id = query_dict.get('v', [None])[0] or self._search_regex(
2202 r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2203 'video id', default=None)
2205 if self._downloader.params.get('noplaylist'):
2206 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2207 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2209 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2210 return video_id, None
2213 def _real_extract(self, url):
2214 # Extract playlist id
2215 mobj = re.match(self._VALID_URL, url)
2217 raise ExtractorError('Invalid URL: %s' % url)
2218 playlist_id = mobj.group(1) or mobj.group(2)
2220 video_id, video = self._check_download_just_video(url, playlist_id)
2224 if playlist_id.startswith(('RD', 'UL', 'PU')):
2225 # Mixes require a custom extraction process
2226 return self._extract_mix(playlist_id)
2228 has_videos, playlist = self._extract_playlist(playlist_id)
2229 if has_videos or not video_id:
2232 # Some playlist URLs don't actually serve a playlist (see
2233 # https://github.com/rg3/youtube-dl/issues/10537).
2234 # Fallback to plain video extraction if there is a video id
2235 # along with playlist id.
2236 return self.url_result(video_id, 'Youtube', video_id=video_id)
2239 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2240 IE_DESC = 'YouTube.com channels'
2241 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
2242 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2243 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2244 IE_NAME = 'youtube:channel'
2246 'note': 'paginated channel',
2247 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2248 'playlist_mincount': 91,
2250 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2251 'title': 'Uploads from lex will',
2254 'note': 'Age restricted channel',
2255 # from https://www.youtube.com/user/DeusExOfficial
2256 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2257 'playlist_mincount': 64,
2259 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2260 'title': 'Uploads from Deus Ex',
2265 def suitable(cls, url):
2266 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2267 else super(YoutubeChannelIE, cls).suitable(url))
2269 def _build_template_url(self, url, channel_id):
2270 return self._TEMPLATE_URL % channel_id
2272 def _real_extract(self, url):
2273 channel_id = self._match_id(url)
2275 url = self._build_template_url(url, channel_id)
2277 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2278 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2279 # otherwise fallback on channel by page extraction
2280 channel_page = self._download_webpage(
2281 url + '?view=57', channel_id,
2282 'Downloading channel page', fatal=False)
2283 if channel_page is False:
2284 channel_playlist_id = False
2286 channel_playlist_id = self._html_search_meta(
2287 'channelId', channel_page, 'channel id', default=None)
2288 if not channel_playlist_id:
2289 channel_url = self._html_search_meta(
2290 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2291 channel_page, 'channel url', default=None)
2293 channel_playlist_id = self._search_regex(
2294 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2295 channel_url, 'channel id', default=None)
2296 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2297 playlist_id = 'UU' + channel_playlist_id[2:]
2298 return self.url_result(
2299 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2301 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2302 autogenerated = re.search(r'''(?x)
2304 channel-header-autogenerated-label|
2305 yt-channel-title-autogenerated
2306 )[^"]*"''', channel_page) is not None
2309 # The videos are contained in a single page
2310 # the ajax pages can't be used, they are empty
2313 video_id, 'Youtube', video_id=video_id,
2314 video_title=video_title)
2315 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2316 return self.playlist_result(entries, channel_id)
2319 next(self._entries(channel_page, channel_id))
2320 except StopIteration:
2321 alert_message = self._html_search_regex(
2322 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2323 channel_page, 'alert', default=None, group='alert')
2325 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2327 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2330 class YoutubeUserIE(YoutubeChannelIE):
2331 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2332 _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_-]+)'
2333 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2334 IE_NAME = 'youtube:user'
2337 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2338 'playlist_mincount': 320,
2340 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2341 'title': 'Uploads from The Linux Foundation',
2344 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2345 # but not https://www.youtube.com/user/12minuteathlete/videos
2346 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2347 'playlist_mincount': 249,
2349 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2350 'title': 'Uploads from 12 Minute Athlete',
2353 'url': 'ytuser:phihag',
2354 'only_matching': True,
2356 'url': 'https://www.youtube.com/c/gametrailers',
2357 'only_matching': True,
2359 'url': 'https://www.youtube.com/gametrailers',
2360 'only_matching': True,
2362 # This channel is not available, geo restricted to JP
2363 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2364 'only_matching': True,
2368 def suitable(cls, url):
2369 # Don't return True if the url can be extracted with other youtube
2370 # extractor, the regex would is too permissive and it would match.
2371 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2372 if any(ie.suitable(url) for ie in other_yt_ies):
2375 return super(YoutubeUserIE, cls).suitable(url)
2377 def _build_template_url(self, url, channel_id):
2378 mobj = re.match(self._VALID_URL, url)
2379 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2382 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2383 IE_DESC = 'YouTube.com live streams'
2384 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2385 IE_NAME = 'youtube:live'
2388 'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2390 'id': 'a48o2S1cPoo',
2392 'title': 'The Young Turks - Live Main Show',
2393 'uploader': 'The Young Turks',
2394 'uploader_id': 'TheYoungTurks',
2395 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2396 'upload_date': '20150715',
2397 'license': 'Standard YouTube License',
2398 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2399 'categories': ['News & Politics'],
2400 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2402 'dislike_count': int,
2405 'skip_download': True,
2408 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2409 'only_matching': True,
2411 'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2412 'only_matching': True,
2414 'url': 'https://www.youtube.com/TheYoungTurks/live',
2415 'only_matching': True,
2418 def _real_extract(self, url):
2419 mobj = re.match(self._VALID_URL, url)
2420 channel_id = mobj.group('id')
2421 base_url = mobj.group('base_url')
2422 webpage = self._download_webpage(url, channel_id, fatal=False)
2424 page_type = self._og_search_property(
2425 'type', webpage, 'page type', default=None)
2426 video_id = self._html_search_meta(
2427 'videoId', webpage, 'video id', default=None)
2428 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2429 return self.url_result(video_id, YoutubeIE.ie_key())
2430 return self.url_result(base_url)
2433 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2434 IE_DESC = 'YouTube.com user/channel playlists'
2435 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2436 IE_NAME = 'youtube:playlists'
2439 'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2440 'playlist_mincount': 4,
2442 'id': 'ThirstForScience',
2443 'title': 'Thirst for Science',
2446 # with "Load more" button
2447 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2448 'playlist_mincount': 70,
2451 'title': 'Игорь Клейнер',
2454 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2455 'playlist_mincount': 17,
2457 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2458 'title': 'Chem Player',
2463 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2464 IE_DESC = 'YouTube.com searches'
2465 # there doesn't appear to be a real limit, for example if you search for
2466 # 'python' you get more than 8.000.000 results
2467 _MAX_RESULTS = float('inf')
2468 IE_NAME = 'youtube:search'
2469 _SEARCH_KEY = 'ytsearch'
2470 _EXTRA_QUERY_ARGS = {}
2473 def _get_n_results(self, query, n):
2474 """Get a specified number of results for a query"""
2480 'search_query': query.encode('utf-8'),
2482 url_query.update(self._EXTRA_QUERY_ARGS)
2483 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2485 for pagenum in itertools.count(1):
2486 data = self._download_json(
2487 result_url, video_id='query "%s"' % query,
2488 note='Downloading page %s' % pagenum,
2489 errnote='Unable to download API page',
2490 query={'spf': 'navigate'})
2491 html_content = data[1]['body']['content']
2493 if 'class="search-message' in html_content:
2494 raise ExtractorError(
2495 '[youtube] No video results', expected=True)
2497 new_videos = self._ids_to_results(orderedSet(re.findall(
2498 r'href="/watch\?v=(.{11})', html_content)))
2499 videos += new_videos
2500 if not new_videos or len(videos) > limit:
2502 next_link = self._html_search_regex(
2503 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2504 html_content, 'next link', default=None)
2505 if next_link is None:
2507 result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2511 return self.playlist_result(videos, query)
2514 class YoutubeSearchDateIE(YoutubeSearchIE):
2515 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2516 _SEARCH_KEY = 'ytsearchdate'
2517 IE_DESC = 'YouTube.com searches, newest videos first'
2518 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2521 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2522 IE_DESC = 'YouTube.com search URLs'
2523 IE_NAME = 'youtube:search_url'
2524 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2525 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2527 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2528 'playlist_mincount': 5,
2530 'title': 'youtube-dl test video',
2533 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2534 'only_matching': True,
2537 def _real_extract(self, url):
2538 mobj = re.match(self._VALID_URL, url)
2539 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2540 webpage = self._download_webpage(url, query)
2541 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2544 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2545 IE_DESC = 'YouTube.com (multi-season) shows'
2546 _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
2547 IE_NAME = 'youtube:show'
2549 'url': 'https://www.youtube.com/show/airdisasters',
2550 'playlist_mincount': 5,
2552 'id': 'airdisasters',
2553 'title': 'Air Disasters',
2557 def _real_extract(self, url):
2558 playlist_id = self._match_id(url)
2559 return super(YoutubeShowIE, self)._real_extract(
2560 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2563 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2565 Base class for feed extractors
2566 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2568 _LOGIN_REQUIRED = True
2572 return 'youtube:%s' % self._FEED_NAME
2574 def _real_initialize(self):
2577 def _real_extract(self, url):
2578 page = self._download_webpage(
2579 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2581 # The extraction process is the same as for playlists, but the regex
2582 # for the video ids doesn't contain an index
2584 more_widget_html = content_html = page
2585 for page_num in itertools.count(1):
2586 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2588 # 'recommended' feed has infinite 'load more' and each new portion spins
2589 # the same videos in (sometimes) slightly different order, so we'll check
2590 # for unicity and break when portion has no new videos
2591 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2597 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2601 more = self._download_json(
2602 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2603 'Downloading page #%s' % page_num,
2604 transform_source=uppercase_escape)
2605 content_html = more['content_html']
2606 more_widget_html = more['load_more_widget_html']
2608 return self.playlist_result(
2609 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2612 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2613 IE_NAME = 'youtube:watchlater'
2614 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2615 _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2618 'url': 'https://www.youtube.com/playlist?list=WL',
2619 'only_matching': True,
2621 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2622 'only_matching': True,
2625 def _real_extract(self, url):
2626 _, video = self._check_download_just_video(url, 'WL')
2629 _, playlist = self._extract_playlist('WL')
2633 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2634 IE_NAME = 'youtube:favorites'
2635 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2636 _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2637 _LOGIN_REQUIRED = True
2639 def _real_extract(self, url):
2640 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2641 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2642 return self.url_result(playlist_id, 'YoutubePlaylist')
2645 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2646 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2647 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2648 _FEED_NAME = 'recommended'
2649 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2652 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2653 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2654 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2655 _FEED_NAME = 'subscriptions'
2656 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2659 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2660 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2661 _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
2662 _FEED_NAME = 'history'
2663 _PLAYLIST_TITLE = 'Youtube History'
2666 class YoutubeTruncatedURLIE(InfoExtractor):
2667 IE_NAME = 'youtube:truncated_url'
2668 IE_DESC = False # Do not list
2669 _VALID_URL = r'''(?x)
2671 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2674 annotation_id=annotation_[^&]+|
2680 attribution_link\?a=[^&]+
2686 'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
2687 'only_matching': True,
2689 'url': 'https://www.youtube.com/watch?',
2690 'only_matching': True,
2692 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2693 'only_matching': True,
2695 'url': 'https://www.youtube.com/watch?feature=foo',
2696 'only_matching': True,
2698 'url': 'https://www.youtube.com/watch?hl=en-GB',
2699 'only_matching': True,
2701 'url': 'https://www.youtube.com/watch?t=2372',
2702 'only_matching': True,
2705 def _real_extract(self, url):
2706 raise ExtractorError(
2707 'Did you forget to quote the URL? Remember that & is a meta '
2708 'character in most shells, so you want to put the URL in quotes, '
2710 '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2711 ' or simply youtube-dl BaW_jenozKc .',
2715 class YoutubeTruncatedIDIE(InfoExtractor):
2716 IE_NAME = 'youtube:truncated_id'
2717 IE_DESC = False # Do not list
2718 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2721 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2722 'only_matching': True,
2725 def _real_extract(self, url):
2726 video_id = self._match_id(url)
2727 raise ExtractorError(
2728 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),