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'
56 _PASSWORD_CHALLENGE_URL = 'https://accounts.google.com/signin/challenge/sl/password'
57 _NETRC_MACHINE = 'youtube'
58 # If True it will raise an error if no login info is provided
59 _LOGIN_REQUIRED = False
61 def _set_language(self):
63 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
64 # YouTube sets the expire time to about two months
65 expire_time=time.time() + 2 * 30 * 24 * 3600)
67 def _ids_to_results(self, ids):
69 self.url_result(vid_id, 'Youtube', video_id=vid_id)
74 Attempt to log in to YouTube.
75 True is returned if successful or skipped.
76 False is returned if login failed.
78 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
80 (username, password) = self._get_login_info()
81 # No authentication to be performed
83 if self._LOGIN_REQUIRED:
84 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
87 login_page = self._download_webpage(
88 self._LOGIN_URL, None,
89 note='Downloading login page',
90 errnote='unable to fetch login page', fatal=False)
91 if login_page is False:
94 login_form = self._hidden_inputs(login_page)
97 'checkConnection': 'youtube',
102 login_results = self._download_webpage(
103 self._PASSWORD_CHALLENGE_URL, None,
104 note='Logging in', errnote='unable to log in', fatal=False,
105 data=urlencode_postdata(login_form))
106 if login_results is False:
109 error_msg = self._html_search_regex(
110 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
111 login_results, 'error message', default=None)
113 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
115 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
116 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
119 # TODO add SMS and phone call support - these require making a request and then prompting the user
121 if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
122 tfa_code = self._get_tfa_info('2-step verification code')
125 self._downloader.report_warning(
126 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
127 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
130 tfa_code = remove_start(tfa_code, 'G-')
132 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
134 tfa_form_strs.update({
139 tfa_data = urlencode_postdata(tfa_form_strs)
141 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
142 tfa_results = self._download_webpage(
144 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
146 if tfa_results is False:
149 if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
150 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
152 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
153 self._downloader.report_warning('unable to log in - did the page structure change?')
155 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
156 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
159 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
160 self._downloader.report_warning('unable to log in: bad username or password')
164 def _real_initialize(self):
165 if self._downloader is None:
168 if not self._login():
172 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
173 # Extract entries from page with "Load more" button
174 def _entries(self, page, playlist_id):
175 more_widget_html = content_html = page
176 for page_num in itertools.count(1):
177 for entry in self._process_page(content_html):
180 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
184 more = self._download_json(
185 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
186 'Downloading page #%s' % page_num,
187 transform_source=uppercase_escape)
188 content_html = more['content_html']
189 if not content_html.strip():
190 # Some webpages show a "Load more" button but they don't
193 more_widget_html = more['load_more_widget_html']
196 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
197 def _process_page(self, content):
198 for video_id, video_title in self.extract_videos_from_page(content):
199 yield self.url_result(video_id, 'Youtube', video_id, video_title)
201 def extract_videos_from_page(self, page):
204 for mobj in re.finditer(self._VIDEO_RE, page):
205 # The link with index 0 is not the first video of the playlist (not sure if still actual)
206 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
208 video_id = mobj.group('id')
209 video_title = unescapeHTML(mobj.group('title'))
211 video_title = video_title.strip()
213 idx = ids_in_page.index(video_id)
214 if video_title and not titles_in_page[idx]:
215 titles_in_page[idx] = video_title
217 ids_in_page.append(video_id)
218 titles_in_page.append(video_title)
219 return zip(ids_in_page, titles_in_page)
222 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
223 def _process_page(self, content):
224 for playlist_id in orderedSet(re.findall(
225 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
227 yield self.url_result(
228 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
230 def _real_extract(self, url):
231 playlist_id = self._match_id(url)
232 webpage = self._download_webpage(url, playlist_id)
233 title = self._og_search_title(webpage, fatal=False)
234 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
237 class YoutubeIE(YoutubeBaseInfoExtractor):
238 IE_DESC = 'YouTube.com'
239 _VALID_URL = r"""(?x)^
241 (?:https?://|//) # http(s):// or protocol-independent URL
242 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
243 (?:www\.)?deturl\.com/www\.youtube\.com/|
244 (?:www\.)?pwnyoutube\.com/|
245 (?:www\.)?yourepeat\.com/|
246 tube\.majestyc\.net/|
247 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
248 (?:.*?\#/)? # handle anchor (#/) redirect urls
249 (?: # the various things that can precede the ID:
250 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
251 |(?: # or the v= param in all its forms
252 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
253 (?:\?|\#!?) # the params delimiter ? or # or #!
254 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
259 youtu\.be| # just youtu.be/xxxx
260 vid\.plus| # or vid.plus/xxxx
261 zwearz\.com/watch| # or zwearz.com/watch/xxxx
263 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
265 )? # all until now is optional -> you can pass the naked ID
266 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
267 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
268 (?(1).+)? # if we found the ID, everything can follow
270 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
272 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
273 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
274 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
275 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
276 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
277 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
278 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
279 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
280 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
281 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
282 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
283 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
284 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
285 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
286 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
287 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
288 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
293 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
294 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
295 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
296 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
297 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
298 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
299 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
301 # Apple HTTP Live Streaming
302 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
303 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
304 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
305 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
306 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
307 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
308 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
309 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
312 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
313 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
314 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
315 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
316 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
317 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
318 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
319 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
320 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
321 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
322 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
325 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
326 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
327 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
328 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
329 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
332 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
333 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
334 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
335 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
336 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
337 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
339 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
340 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
341 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
342 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
343 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
344 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
345 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
346 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
348 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
350 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
351 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
352 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
356 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
357 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
359 # Dash webm audio with opus inside
360 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
361 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
362 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
365 '_rtmp': {'protocol': 'rtmp'},
367 _SUBTITLE_FORMATS = ('ttml', 'vtt')
372 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
376 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
377 'uploader': 'Philipp Hagemeister',
378 'uploader_id': 'phihag',
379 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
380 'upload_date': '20121002',
381 'license': 'Standard YouTube License',
382 '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 .',
383 'categories': ['Science & Technology'],
384 'tags': ['youtube-dl'],
386 'dislike_count': int,
392 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
393 'note': 'Test generic use_cipher_signature video (#897)',
397 'upload_date': '20120506',
398 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
399 'alt_title': 'I Love It (feat. Charli XCX)',
400 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
401 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
402 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
403 'iconic ep', 'iconic', 'love', 'it'],
404 'uploader': 'Icona Pop',
405 'uploader_id': 'IconaPop',
406 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
407 'license': 'Standard YouTube License',
408 'creator': 'Icona Pop',
412 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
413 'note': 'Test VEVO video with age protection (#956)',
417 'upload_date': '20130703',
418 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
419 'alt_title': 'Tunnel Vision',
420 'description': 'md5:64249768eec3bc4276236606ea996373',
421 'uploader': 'justintimberlakeVEVO',
422 'uploader_id': 'justintimberlakeVEVO',
423 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
424 'license': 'Standard YouTube License',
425 'creator': 'Justin Timberlake',
430 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
431 'note': 'Embed-only video (#1746)',
435 'upload_date': '20120608',
436 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
437 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
438 'uploader': 'SET India',
439 'uploader_id': 'setindia',
440 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
441 'license': 'Standard YouTube License',
446 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
447 'note': 'Use the first video ID in the URL',
451 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
452 'uploader': 'Philipp Hagemeister',
453 'uploader_id': 'phihag',
454 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
455 'upload_date': '20121002',
456 'license': 'Standard YouTube License',
457 '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 .',
458 'categories': ['Science & Technology'],
459 'tags': ['youtube-dl'],
461 'dislike_count': int,
464 'skip_download': True,
468 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
469 'note': '256k DASH audio (format 141) via DASH manifest',
473 'upload_date': '20121002',
474 'uploader_id': '8KVIDEO',
475 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
477 'uploader': '8KVIDEO',
478 'license': 'Standard YouTube License',
479 'title': 'UHDTV TEST 8K VIDEO.mp4'
482 'youtube_include_dash_manifest': True,
485 'skip': 'format 141 not served anymore',
487 # DASH manifest with encrypted signature
489 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
493 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
494 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
495 'uploader': 'AfrojackVEVO',
496 'uploader_id': 'AfrojackVEVO',
497 'upload_date': '20131011',
498 'license': 'Standard YouTube License',
501 'youtube_include_dash_manifest': True,
502 'format': '141/bestaudio[ext=m4a]',
505 # JS player signature function name containing $
507 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
511 'title': 'Taylor Swift - Shake It Off',
512 'alt_title': 'Shake It Off',
513 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
514 'uploader': 'TaylorSwiftVEVO',
515 'uploader_id': 'TaylorSwiftVEVO',
516 'upload_date': '20140818',
517 'license': 'Standard YouTube License',
518 'creator': 'Taylor Swift',
521 'youtube_include_dash_manifest': True,
522 'format': '141/bestaudio[ext=m4a]',
527 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
531 'upload_date': '20100909',
532 'uploader': 'The Amazing Atheist',
533 'uploader_id': 'TheAmazingAtheist',
534 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
535 'license': 'Standard YouTube License',
536 'title': 'Burning Everyone\'s Koran',
537 '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',
540 # Normal age-gate video (No vevo, embed allowed)
542 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
546 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
547 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
548 'uploader': 'The Witcher',
549 'uploader_id': 'WitcherGame',
550 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
551 'upload_date': '20140605',
552 'license': 'Standard YouTube License',
556 # Age-gate video with encrypted signature
558 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
562 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
563 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
564 'uploader': 'LloydVEVO',
565 'uploader_id': 'LloydVEVO',
566 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
567 'upload_date': '20110629',
568 'license': 'Standard YouTube License',
572 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
574 'url': '__2ABJjxzNo',
578 'upload_date': '20100430',
579 'uploader_id': 'deadmau5',
580 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
581 'creator': 'deadmau5',
582 'description': 'md5:12c56784b8032162bb936a5f76d55360',
583 'uploader': 'deadmau5',
584 'license': 'Standard YouTube License',
585 'title': 'Deadmau5 - Some Chords (HD)',
586 'alt_title': 'Some Chords',
588 'expected_warnings': [
589 'DASH manifest missing',
592 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
594 'url': 'lqQg6PlCWgI',
598 'upload_date': '20150827',
599 'uploader_id': 'olympic',
600 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
601 'license': 'Standard YouTube License',
602 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
603 'uploader': 'Olympic',
604 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
607 'skip_download': 'requires avconv',
612 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
616 'stretched_ratio': 16 / 9.,
617 'upload_date': '20110310',
618 'uploader_id': 'AllenMeow',
619 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
620 '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',
622 'license': 'Standard YouTube License',
623 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
626 # url_encoded_fmt_stream_map is empty string
628 'url': 'qEJwOuvDf7I',
632 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
634 'upload_date': '20150404',
635 'uploader_id': 'spbelect',
636 'uploader': 'Наблюдатели Петербурга',
639 'skip_download': 'requires avconv',
641 'skip': 'This live event has ended.',
643 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
645 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
649 'title': 'md5:7b81415841e02ecd4313668cde88737a',
650 'description': 'md5:116377fd2963b81ec4ce64b542173306',
651 'upload_date': '20150625',
652 'uploader_id': 'dorappi2000',
653 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
654 'uploader': 'dorappi2000',
655 'license': 'Standard YouTube License',
656 'formats': 'mincount:32',
659 # DASH manifest with segment_list
661 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
662 'md5': '8ce563a1d667b599d21064e982ab9e31',
666 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
667 'uploader': 'Airtek',
668 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
669 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
670 'license': 'Standard YouTube License',
671 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
674 'youtube_include_dash_manifest': True,
675 'format': '135', # bestvideo
677 'skip': 'This live event has ended.',
680 # Multifeed videos (multiple cameras), URL is for Main Camera
681 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
684 'title': 'teamPGP: Rocket League Noob Stream',
685 'description': 'md5:dc7872fb300e143831327f1bae3af010',
691 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
692 'description': 'md5:dc7872fb300e143831327f1bae3af010',
693 'upload_date': '20150721',
694 'uploader': 'Beer Games Beer',
695 'uploader_id': 'beergamesbeer',
696 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
697 'license': 'Standard YouTube License',
703 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
704 'description': 'md5:dc7872fb300e143831327f1bae3af010',
705 'upload_date': '20150721',
706 'uploader': 'Beer Games Beer',
707 'uploader_id': 'beergamesbeer',
708 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
709 'license': 'Standard YouTube License',
715 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
716 'description': 'md5:dc7872fb300e143831327f1bae3af010',
717 'upload_date': '20150721',
718 'uploader': 'Beer Games Beer',
719 'uploader_id': 'beergamesbeer',
720 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
721 'license': 'Standard YouTube License',
727 'title': 'teamPGP: Rocket League Noob Stream (zim)',
728 'description': 'md5:dc7872fb300e143831327f1bae3af010',
729 'upload_date': '20150721',
730 'uploader': 'Beer Games Beer',
731 'uploader_id': 'beergamesbeer',
732 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
733 'license': 'Standard YouTube License',
737 'skip_download': True,
741 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
742 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
745 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
748 'skip': 'Not multifeed anymore',
751 'url': 'http://vid.plus/FlRa-iH7PGw',
752 'only_matching': True,
755 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
756 'only_matching': True,
759 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
760 # Also tests cut-off URL expansion in video description (see
761 # https://github.com/rg3/youtube-dl/issues/1892,
762 # https://github.com/rg3/youtube-dl/issues/8164)
763 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
767 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
768 'alt_title': 'Dark Walk',
769 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
770 'upload_date': '20151119',
771 'uploader_id': 'IronSoulElf',
772 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
773 'uploader': 'IronSoulElf',
774 'license': 'Standard YouTube License',
775 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
778 'skip_download': True,
782 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
783 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
784 'only_matching': True,
787 # Video with yt:stretch=17:0
788 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
792 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
793 'description': 'md5:ee18a25c350637c8faff806845bddee9',
794 'upload_date': '20151107',
795 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
796 'uploader': 'CH GAMER DROID',
799 'skip_download': True,
801 'skip': 'This video does not exist.',
804 # Video licensed under Creative Commons
805 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
809 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
810 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
811 'upload_date': '20150127',
812 'uploader_id': 'BerkmanCenter',
813 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
814 'uploader': 'BerkmanCenter',
815 'license': 'Creative Commons Attribution license (reuse allowed)',
818 'skip_download': True,
822 # Channel-like uploader_url
823 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
827 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
828 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
829 'upload_date': '20151119',
830 'uploader': 'Bernie 2016',
831 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
833 'license': 'Creative Commons Attribution license (reuse allowed)',
836 'skip_download': True,
840 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
841 'only_matching': True,
844 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
845 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
846 'only_matching': True,
850 def __init__(self, *args, **kwargs):
851 super(YoutubeIE, self).__init__(*args, **kwargs)
852 self._player_cache = {}
854 def report_video_info_webpage_download(self, video_id):
855 """Report attempt to download video info webpage."""
856 self.to_screen('%s: Downloading video info webpage' % video_id)
858 def report_information_extraction(self, video_id):
859 """Report attempt to extract video information."""
860 self.to_screen('%s: Extracting video information' % video_id)
862 def report_unavailable_format(self, video_id, format):
863 """Report extracted video URL."""
864 self.to_screen('%s: Format %s not available' % (video_id, format))
866 def report_rtmp_download(self):
867 """Indicate the download will use the RTMP protocol."""
868 self.to_screen('RTMP download detected')
870 def _signature_cache_id(self, example_sig):
871 """ Return a string representation of a signature """
872 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
874 def _extract_signature_function(self, video_id, player_url, example_sig):
876 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
879 raise ExtractorError('Cannot identify player %r' % player_url)
880 player_type = id_m.group('ext')
881 player_id = id_m.group('id')
883 # Read from filesystem cache
884 func_id = '%s_%s_%s' % (
885 player_type, player_id, self._signature_cache_id(example_sig))
886 assert os.path.basename(func_id) == func_id
888 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
889 if cache_spec is not None:
890 return lambda s: ''.join(s[i] for i in cache_spec)
893 'Downloading player %s' % player_url
894 if self._downloader.params.get('verbose') else
895 'Downloading %s player %s' % (player_type, player_id)
897 if player_type == 'js':
898 code = self._download_webpage(
899 player_url, video_id,
901 errnote='Download of %s failed' % player_url)
902 res = self._parse_sig_js(code)
903 elif player_type == 'swf':
904 urlh = self._request_webpage(
905 player_url, video_id,
907 errnote='Download of %s failed' % player_url)
909 res = self._parse_sig_swf(code)
911 assert False, 'Invalid player type %r' % player_type
913 test_string = ''.join(map(compat_chr, range(len(example_sig))))
914 cache_res = res(test_string)
915 cache_spec = [ord(c) for c in cache_res]
917 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
920 def _print_sig_code(self, func, example_sig):
921 def gen_sig_code(idxs):
922 def _genslice(start, end, step):
923 starts = '' if start == 0 else str(start)
924 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
925 steps = '' if step == 1 else (':%d' % step)
926 return 's[%s%s%s]' % (starts, ends, steps)
929 # Quelch pyflakes warnings - start will be set when step is set
930 start = '(Never used)'
931 for i, prev in zip(idxs[1:], idxs[:-1]):
935 yield _genslice(start, prev, step)
938 if i - prev in [-1, 1]:
947 yield _genslice(start, i, step)
949 test_string = ''.join(map(compat_chr, range(len(example_sig))))
950 cache_res = func(test_string)
951 cache_spec = [ord(c) for c in cache_res]
952 expr_code = ' + '.join(gen_sig_code(cache_spec))
953 signature_id_tuple = '(%s)' % (
954 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
955 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
956 ' return %s\n') % (signature_id_tuple, expr_code)
957 self.to_screen('Extracted signature function:\n' + code)
959 def _parse_sig_js(self, jscode):
960 funcname = self._search_regex(
961 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
962 'Initial JS player signature function name')
964 jsi = JSInterpreter(jscode)
965 initial_function = jsi.extract_function(funcname)
966 return lambda s: initial_function([s])
968 def _parse_sig_swf(self, file_contents):
969 swfi = SWFInterpreter(file_contents)
970 TARGET_CLASSNAME = 'SignatureDecipher'
971 searched_class = swfi.extract_class(TARGET_CLASSNAME)
972 initial_function = swfi.extract_function(searched_class, 'decipher')
973 return lambda s: initial_function([s])
975 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
976 """Turn the encrypted s field into a working signature"""
978 if player_url is None:
979 raise ExtractorError('Cannot decrypt signature without player_url')
981 if player_url.startswith('//'):
982 player_url = 'https:' + player_url
984 player_id = (player_url, self._signature_cache_id(s))
985 if player_id not in self._player_cache:
986 func = self._extract_signature_function(
987 video_id, player_url, s
989 self._player_cache[player_id] = func
990 func = self._player_cache[player_id]
991 if self._downloader.params.get('youtube_print_sig_code'):
992 self._print_sig_code(func, s)
994 except Exception as e:
995 tb = traceback.format_exc()
996 raise ExtractorError(
997 'Signature extraction failed: ' + tb, cause=e)
999 def _get_subtitles(self, video_id, webpage):
1001 subs_doc = self._download_xml(
1002 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1003 video_id, note=False)
1004 except ExtractorError as err:
1005 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1009 for track in subs_doc.findall('track'):
1010 lang = track.attrib['lang_code']
1011 if lang in sub_lang_list:
1014 for ext in self._SUBTITLE_FORMATS:
1015 params = compat_urllib_parse_urlencode({
1019 'name': track.attrib['name'].encode('utf-8'),
1021 sub_formats.append({
1022 'url': 'https://www.youtube.com/api/timedtext?' + params,
1025 sub_lang_list[lang] = sub_formats
1026 if not sub_lang_list:
1027 self._downloader.report_warning('video doesn\'t have subtitles')
1029 return sub_lang_list
1031 def _get_ytplayer_config(self, video_id, webpage):
1033 # User data may contain arbitrary character sequences that may affect
1034 # JSON extraction with regex, e.g. when '};' is contained the second
1035 # regex won't capture the whole JSON. Yet working around by trying more
1036 # concrete regex first keeping in mind proper quoted string handling
1037 # to be implemented in future that will replace this workaround (see
1038 # https://github.com/rg3/youtube-dl/issues/7468,
1039 # https://github.com/rg3/youtube-dl/pull/7599)
1040 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1041 r';ytplayer\.config\s*=\s*({.+?});',
1043 config = self._search_regex(
1044 patterns, webpage, 'ytplayer.config', default=None)
1046 return self._parse_json(
1047 uppercase_escape(config), video_id, fatal=False)
1049 def _get_automatic_captions(self, video_id, webpage):
1050 """We need the webpage for getting the captions url, pass it as an
1051 argument to speed up the process."""
1052 self.to_screen('%s: Looking for automatic captions' % video_id)
1053 player_config = self._get_ytplayer_config(video_id, webpage)
1054 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1055 if not player_config:
1056 self._downloader.report_warning(err_msg)
1059 args = player_config['args']
1060 caption_url = args.get('ttsurl')
1062 timestamp = args['timestamp']
1063 # We get the available subtitles
1064 list_params = compat_urllib_parse_urlencode({
1069 list_url = caption_url + '&' + list_params
1070 caption_list = self._download_xml(list_url, video_id)
1071 original_lang_node = caption_list.find('track')
1072 if original_lang_node is None:
1073 self._downloader.report_warning('Video doesn\'t have automatic captions')
1075 original_lang = original_lang_node.attrib['lang_code']
1076 caption_kind = original_lang_node.attrib.get('kind', '')
1079 for lang_node in caption_list.findall('target'):
1080 sub_lang = lang_node.attrib['lang_code']
1082 for ext in self._SUBTITLE_FORMATS:
1083 params = compat_urllib_parse_urlencode({
1084 'lang': original_lang,
1088 'kind': caption_kind,
1090 sub_formats.append({
1091 'url': caption_url + '&' + params,
1094 sub_lang_list[sub_lang] = sub_formats
1095 return sub_lang_list
1097 # Some videos don't provide ttsurl but rather caption_tracks and
1098 # caption_translation_languages (e.g. 20LmZk1hakA)
1099 caption_tracks = args['caption_tracks']
1100 caption_translation_languages = args['caption_translation_languages']
1101 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1102 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1103 caption_qs = compat_parse_qs(parsed_caption_url.query)
1106 for lang in caption_translation_languages.split(','):
1107 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1108 sub_lang = lang_qs.get('lc', [None])[0]
1112 for ext in self._SUBTITLE_FORMATS:
1114 'tlang': [sub_lang],
1117 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1118 query=compat_urllib_parse_urlencode(caption_qs, True)))
1119 sub_formats.append({
1123 sub_lang_list[sub_lang] = sub_formats
1124 return sub_lang_list
1125 # An extractor error can be raise by the download process if there are
1126 # no automatic captions but there are subtitles
1127 except (KeyError, ExtractorError):
1128 self._downloader.report_warning(err_msg)
1131 def _mark_watched(self, video_id, video_info):
1132 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1133 if not playback_url:
1135 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1136 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1138 # cpn generation algorithm is reverse engineered from base.js.
1139 # In fact it works even with dummy cpn.
1140 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1141 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1147 playback_url = compat_urlparse.urlunparse(
1148 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1150 self._download_webpage(
1151 playback_url, video_id, 'Marking watched',
1152 'Unable to mark watched', fatal=False)
1155 def extract_id(cls, url):
1156 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1158 raise ExtractorError('Invalid URL: %s' % url)
1159 video_id = mobj.group(2)
1162 def _extract_from_m3u8(self, manifest_url, video_id):
1165 def _get_urls(_manifest):
1166 lines = _manifest.split('\n')
1167 urls = filter(lambda l: l and not l.startswith('#'),
1170 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1171 formats_urls = _get_urls(manifest)
1172 for format_url in formats_urls:
1173 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1174 url_map[itag] = format_url
1177 def _extract_annotations(self, video_id):
1178 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1179 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1181 def _real_extract(self, url):
1182 url, smuggled_data = unsmuggle_url(url, {})
1185 'http' if self._downloader.params.get('prefer_insecure', False)
1190 parsed_url = compat_urllib_parse_urlparse(url)
1191 for component in [parsed_url.fragment, parsed_url.query]:
1192 query = compat_parse_qs(component)
1193 if start_time is None and 't' in query:
1194 start_time = parse_duration(query['t'][0])
1195 if start_time is None and 'start' in query:
1196 start_time = parse_duration(query['start'][0])
1197 if end_time is None and 'end' in query:
1198 end_time = parse_duration(query['end'][0])
1200 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1201 mobj = re.search(self._NEXT_URL_RE, url)
1203 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1204 video_id = self.extract_id(url)
1207 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1208 video_webpage = self._download_webpage(url, video_id)
1210 # Attempt to extract SWF player URL
1211 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1212 if mobj is not None:
1213 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1219 def add_dash_mpd(video_info):
1220 dash_mpd = video_info.get('dashmpd')
1221 if dash_mpd and dash_mpd[0] not in dash_mpds:
1222 dash_mpds.append(dash_mpd[0])
1225 embed_webpage = None
1227 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1229 # We simulate the access to the video from www.youtube.com/v/{video_id}
1230 # this can be viewed without login into Youtube
1231 url = proto + '://www.youtube.com/embed/%s' % video_id
1232 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1233 data = compat_urllib_parse_urlencode({
1234 'video_id': video_id,
1235 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1236 'sts': self._search_regex(
1237 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1239 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1240 video_info_webpage = self._download_webpage(
1241 video_info_url, video_id,
1242 note='Refetching age-gated info webpage',
1243 errnote='unable to download video info webpage')
1244 video_info = compat_parse_qs(video_info_webpage)
1245 add_dash_mpd(video_info)
1249 # Try looking directly into the video webpage
1250 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1252 args = ytplayer_config['args']
1253 if args.get('url_encoded_fmt_stream_map'):
1254 # Convert to the same format returned by compat_parse_qs
1255 video_info = dict((k, [v]) for k, v in args.items())
1256 add_dash_mpd(video_info)
1257 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1259 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1260 # We also try looking in get_video_info since it may contain different dashmpd
1261 # URL that points to a DASH manifest with possibly different itag set (some itags
1262 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1263 # manifest pointed by get_video_info's dashmpd).
1264 # The general idea is to take a union of itags of both DASH manifests (for example
1265 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1266 self.report_video_info_webpage_download(video_id)
1267 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1269 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1270 % (proto, video_id, el_type))
1271 video_info_webpage = self._download_webpage(
1273 video_id, note=False,
1274 errnote='unable to download video info webpage')
1275 get_video_info = compat_parse_qs(video_info_webpage)
1276 if get_video_info.get('use_cipher_signature') != ['True']:
1277 add_dash_mpd(get_video_info)
1279 video_info = get_video_info
1280 if 'token' in get_video_info:
1281 # Different get_video_info requests may report different results, e.g.
1282 # some may report video unavailability, but some may serve it without
1283 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1284 # the original webpage as well as el=info and el=embedded get_video_info
1285 # requests report video unavailability due to geo restriction while
1286 # el=detailpage succeeds and returns valid data). This is probably
1287 # due to YouTube measures against IP ranges of hosting providers.
1288 # Working around by preferring the first succeeded video_info containing
1289 # the token if no such video_info yet was found.
1290 if 'token' not in video_info:
1291 video_info = get_video_info
1293 if 'token' not in video_info:
1294 if 'reason' in video_info:
1295 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1296 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1298 raise ExtractorError('YouTube said: This video is available in %s only' % (
1299 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1301 raise ExtractorError(
1302 'YouTube said: %s' % video_info['reason'][0],
1303 expected=True, video_id=video_id)
1305 raise ExtractorError(
1306 '"token" parameter not in video info for unknown reason',
1310 if 'title' in video_info:
1311 video_title = video_info['title'][0]
1313 self._downloader.report_warning('Unable to extract video title')
1317 video_description = get_element_by_id("eow-description", video_webpage)
1318 if video_description:
1319 video_description = re.sub(r'''(?x)
1321 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1322 (?:title|href)="([^"]+)"\s+
1323 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1327 ''', r'\1', video_description)
1328 video_description = clean_html(video_description)
1330 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1332 video_description = unescapeHTML(fd_mobj.group(1))
1334 video_description = ''
1336 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1337 if not self._downloader.params.get('noplaylist'):
1340 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1341 for feed in multifeed_metadata_list.split(','):
1342 # Unquote should take place before split on comma (,) since textual
1343 # fields may contain comma as well (see
1344 # https://github.com/rg3/youtube-dl/issues/8536)
1345 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1347 '_type': 'url_transparent',
1348 'ie_key': 'Youtube',
1350 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1351 {'force_singlefeed': True}),
1352 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1354 feed_ids.append(feed_data['id'][0])
1356 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1357 % (', '.join(feed_ids), video_id))
1358 return self.playlist_result(entries, video_id, video_title, video_description)
1359 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1361 if 'view_count' in video_info:
1362 view_count = int(video_info['view_count'][0])
1366 # Check for "rental" videos
1367 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1368 raise ExtractorError('"rental" videos not supported')
1370 # Start extracting information
1371 self.report_information_extraction(video_id)
1374 if 'author' not in video_info:
1375 raise ExtractorError('Unable to extract uploader name')
1376 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1379 video_uploader_id = None
1380 video_uploader_url = None
1382 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1384 if mobj is not None:
1385 video_uploader_id = mobj.group('uploader_id')
1386 video_uploader_url = mobj.group('uploader_url')
1388 self._downloader.report_warning('unable to extract uploader nickname')
1391 # We try first to get a high quality image:
1392 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1393 video_webpage, re.DOTALL)
1394 if m_thumb is not None:
1395 video_thumbnail = m_thumb.group(1)
1396 elif 'thumbnail_url' not in video_info:
1397 self._downloader.report_warning('unable to extract video thumbnail')
1398 video_thumbnail = None
1399 else: # don't panic if we can't find it
1400 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1403 upload_date = self._html_search_meta(
1404 'datePublished', video_webpage, 'upload date', default=None)
1406 upload_date = self._search_regex(
1407 [r'(?s)id="eow-date.*?>(.*?)</span>',
1408 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1409 video_webpage, 'upload date', default=None)
1411 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1412 upload_date = unified_strdate(upload_date)
1414 video_license = self._html_search_regex(
1415 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1416 video_webpage, 'license', default=None)
1418 m_music = re.search(
1419 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1422 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1423 video_creator = clean_html(m_music.group('creator'))
1425 video_alt_title = video_creator = None
1427 m_cat_container = self._search_regex(
1428 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1429 video_webpage, 'categories', default=None)
1431 category = self._html_search_regex(
1432 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1434 video_categories = None if category is None else [category]
1436 video_categories = None
1439 unescapeHTML(m.group('content'))
1440 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1442 def _extract_count(count_name):
1443 return str_to_int(self._search_regex(
1444 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1445 % re.escape(count_name),
1446 video_webpage, count_name, default=None))
1448 like_count = _extract_count('like')
1449 dislike_count = _extract_count('dislike')
1452 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1453 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1455 if 'length_seconds' not in video_info:
1456 self._downloader.report_warning('unable to extract video duration')
1457 video_duration = None
1459 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1462 video_annotations = None
1463 if self._downloader.params.get('writeannotations', False):
1464 video_annotations = self._extract_annotations(video_id)
1466 def _map_to_format_list(urlmap):
1468 for itag, video_real_url in urlmap.items():
1471 'url': video_real_url,
1472 'player_url': player_url,
1474 if itag in self._formats:
1475 dct.update(self._formats[itag])
1479 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1480 self.report_rtmp_download()
1482 'format_id': '_rtmp',
1484 'url': video_info['conn'][0],
1485 'player_url': player_url,
1487 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1488 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1489 if 'rtmpe%3Dyes' in encoded_url_map:
1490 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1492 fmt_list = video_info.get('fmt_list', [''])[0]
1494 for fmt in fmt_list.split(','):
1495 spec = fmt.split('/')
1497 width_height = spec[1].split('x')
1498 if len(width_height) == 2:
1499 formats_spec[spec[0]] = {
1500 'resolution': spec[1],
1501 'width': int_or_none(width_height[0]),
1502 'height': int_or_none(width_height[1]),
1505 for url_data_str in encoded_url_map.split(','):
1506 url_data = compat_parse_qs(url_data_str)
1507 if 'itag' not in url_data or 'url' not in url_data:
1509 format_id = url_data['itag'][0]
1510 url = url_data['url'][0]
1512 if 'sig' in url_data:
1513 url += '&signature=' + url_data['sig'][0]
1514 elif 's' in url_data:
1515 encrypted_sig = url_data['s'][0]
1516 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1518 jsplayer_url_json = self._search_regex(
1520 embed_webpage if age_gate else video_webpage,
1521 'JS player URL (1)', default=None)
1522 if not jsplayer_url_json and not age_gate:
1523 # We need the embed website after all
1524 if embed_webpage is None:
1525 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1526 embed_webpage = self._download_webpage(
1527 embed_url, video_id, 'Downloading embed webpage')
1528 jsplayer_url_json = self._search_regex(
1529 ASSETS_RE, embed_webpage, 'JS player URL')
1531 player_url = json.loads(jsplayer_url_json)
1532 if player_url is None:
1533 player_url_json = self._search_regex(
1534 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1535 video_webpage, 'age gate player URL')
1536 player_url = json.loads(player_url_json)
1538 if self._downloader.params.get('verbose'):
1539 if player_url is None:
1540 player_version = 'unknown'
1541 player_desc = 'unknown'
1543 if player_url.endswith('swf'):
1544 player_version = self._search_regex(
1545 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1546 'flash player', fatal=False)
1547 player_desc = 'flash player %s' % player_version
1549 player_version = self._search_regex(
1550 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1552 'html5 player', fatal=False)
1553 player_desc = 'html5 player %s' % player_version
1555 parts_sizes = self._signature_cache_id(encrypted_sig)
1556 self.to_screen('{%s} signature length %s, %s' %
1557 (format_id, parts_sizes, player_desc))
1559 signature = self._decrypt_signature(
1560 encrypted_sig, video_id, player_url, age_gate)
1561 url += '&signature=' + signature
1562 if 'ratebypass' not in url:
1563 url += '&ratebypass=yes'
1566 'format_id': format_id,
1568 'player_url': player_url,
1570 if format_id in self._formats:
1571 dct.update(self._formats[format_id])
1572 if format_id in formats_spec:
1573 dct.update(formats_spec[format_id])
1575 # Some itags are not included in DASH manifest thus corresponding formats will
1576 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1577 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1578 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1579 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1582 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1583 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1586 'fps': int_or_none(url_data.get('fps', [None])[0]),
1587 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1589 for key, value in more_fields.items():
1592 type_ = url_data.get('type', [None])[0]
1594 type_split = type_.split(';')
1595 kind_ext = type_split[0].split('/')
1596 if len(kind_ext) == 2:
1598 dct['ext'] = mimetype2ext(type_split[0])
1599 if kind in ('audio', 'video'):
1601 for mobj in re.finditer(
1602 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1603 if mobj.group('key') == 'codecs':
1604 codecs = mobj.group('val')
1607 codecs = codecs.split(',')
1608 if len(codecs) == 2:
1609 acodec, vcodec = codecs[1], codecs[0]
1611 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1617 elif video_info.get('hlsvp'):
1618 manifest_url = video_info['hlsvp'][0]
1619 url_map = self._extract_from_m3u8(manifest_url, video_id)
1620 formats = _map_to_format_list(url_map)
1621 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1622 for a_format in formats:
1623 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1625 unavailable_message = self._html_search_regex(
1626 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1627 video_webpage, 'unavailable message', default=None)
1628 if unavailable_message:
1629 raise ExtractorError(unavailable_message, expected=True)
1630 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1632 # Look for the DASH manifest
1633 if self._downloader.params.get('youtube_include_dash_manifest', True):
1634 dash_mpd_fatal = True
1635 for mpd_url in dash_mpds:
1638 def decrypt_sig(mobj):
1640 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1641 return '/signature/%s' % dec_s
1643 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1645 for df in self._extract_mpd_formats(
1646 mpd_url, video_id, fatal=dash_mpd_fatal,
1647 formats_dict=self._formats):
1648 # Do not overwrite DASH format found in some previous DASH manifest
1649 if df['format_id'] not in dash_formats:
1650 dash_formats[df['format_id']] = df
1651 # Additional DASH manifests may end up in HTTP Error 403 therefore
1652 # allow them to fail without bug report message if we already have
1653 # some DASH manifest succeeded. This is temporary workaround to reduce
1654 # burst of bug reports until we figure out the reason and whether it
1655 # can be fixed at all.
1656 dash_mpd_fatal = False
1657 except (ExtractorError, KeyError) as e:
1658 self.report_warning(
1659 'Skipping DASH manifest: %r' % e, video_id)
1661 # Remove the formats we found through non-DASH, they
1662 # contain less info and it can be wrong, because we use
1663 # fixed values (for example the resolution). See
1664 # https://github.com/rg3/youtube-dl/issues/5774 for an
1666 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1667 formats.extend(dash_formats.values())
1669 # Check for malformed aspect ratio
1670 stretched_m = re.search(
1671 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1674 w = float(stretched_m.group('w'))
1675 h = float(stretched_m.group('h'))
1676 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1677 # We will only process correct ratios.
1681 if f.get('vcodec') != 'none':
1682 f['stretched_ratio'] = ratio
1684 self._sort_formats(formats)
1686 self.mark_watched(video_id, video_info)
1690 'uploader': video_uploader,
1691 'uploader_id': video_uploader_id,
1692 'uploader_url': video_uploader_url,
1693 'upload_date': upload_date,
1694 'license': video_license,
1695 'creator': video_creator,
1696 'title': video_title,
1697 'alt_title': video_alt_title,
1698 'thumbnail': video_thumbnail,
1699 'description': video_description,
1700 'categories': video_categories,
1702 'subtitles': video_subtitles,
1703 'automatic_captions': automatic_captions,
1704 'duration': video_duration,
1705 'age_limit': 18 if age_gate else 0,
1706 'annotations': video_annotations,
1707 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1708 'view_count': view_count,
1709 'like_count': like_count,
1710 'dislike_count': dislike_count,
1711 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1714 'start_time': start_time,
1715 'end_time': end_time,
1719 class YoutubeSharedVideoIE(InfoExtractor):
1720 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1721 IE_NAME = 'youtube:shared'
1724 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1726 'id': 'uPDB5I9wfp8',
1728 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1729 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1730 'upload_date': '20160219',
1731 'uploader': 'Pocoyo - Português (BR)',
1732 'uploader_id': 'PocoyoBrazil',
1734 'add_ie': ['Youtube'],
1736 # There are already too many Youtube downloads
1737 'skip_download': True,
1741 def _real_extract(self, url):
1742 video_id = self._match_id(url)
1744 webpage = self._download_webpage(url, video_id)
1746 real_video_id = self._html_search_meta(
1747 'videoId', webpage, 'YouTube video id', fatal=True)
1749 return self.url_result(real_video_id, YoutubeIE.ie_key())
1752 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1753 IE_DESC = 'YouTube.com playlists'
1754 _VALID_URL = r"""(?x)(?:
1759 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1760 \? (?:.*?[&;])*? (?:p|a|list)=
1764 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1765 # Top tracks, they can also include dots
1770 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1772 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1773 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1774 IE_NAME = 'youtube:playlist'
1776 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1778 'title': 'ytdl test PL',
1779 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1781 'playlist_count': 3,
1783 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1785 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1786 'title': 'YDL_Empty_List',
1788 'playlist_count': 0,
1790 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1791 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1793 'title': '29C3: Not my department',
1794 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1796 'playlist_count': 95,
1798 'note': 'issue #673',
1799 'url': 'PLBB231211A4F62143',
1801 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1802 'id': 'PLBB231211A4F62143',
1804 'playlist_mincount': 26,
1806 'note': 'Large playlist',
1807 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1809 'title': 'Uploads from Cauchemar',
1810 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1812 'playlist_mincount': 799,
1814 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1816 'title': 'YDL_safe_search',
1817 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1819 'playlist_count': 2,
1822 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1823 'playlist_count': 4,
1826 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1829 'note': 'Embedded SWF player',
1830 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1831 'playlist_count': 4,
1834 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1837 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1838 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1840 'title': 'Uploads from Interstellar Movie',
1841 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1843 'playlist_mincout': 21,
1846 def _real_initialize(self):
1849 def _extract_mix(self, playlist_id):
1850 # The mixes are generated from a single video
1851 # the id of the playlist is just 'RD' + video_id
1853 last_id = playlist_id[-11:]
1854 for n in itertools.count(1):
1855 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1856 webpage = self._download_webpage(
1857 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1858 new_ids = orderedSet(re.findall(
1859 r'''(?xs)data-video-username=".*?".*?
1860 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1862 # Fetch new pages until all the videos are repeated, it seems that
1863 # there are always 51 unique videos.
1864 new_ids = [_id for _id in new_ids if _id not in ids]
1870 url_results = self._ids_to_results(ids)
1872 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1874 search_title('playlist-title') or
1875 search_title('title long-title') or
1876 search_title('title'))
1877 title = clean_html(title_span)
1879 return self.playlist_result(url_results, playlist_id, title)
1881 def _extract_playlist(self, playlist_id):
1882 url = self._TEMPLATE_URL % playlist_id
1883 page = self._download_webpage(url, playlist_id)
1885 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1886 match = match.strip()
1887 # Check if the playlist exists or is private
1888 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1889 raise ExtractorError(
1890 'The playlist doesn\'t exist or is private, use --username or '
1891 '--netrc to access it.',
1893 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1894 raise ExtractorError(
1895 'Invalid parameters. Maybe URL is incorrect.',
1897 elif re.match(r'[^<]*Choose your language[^<]*', match):
1900 self.report_warning('Youtube gives an alert message: ' + match)
1902 playlist_title = self._html_search_regex(
1903 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1906 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1908 def _check_download_just_video(self, url, playlist_id):
1909 # Check if it's a video-specific URL
1910 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1911 if 'v' in query_dict:
1912 video_id = query_dict['v'][0]
1913 if self._downloader.params.get('noplaylist'):
1914 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1915 return self.url_result(video_id, 'Youtube', video_id=video_id)
1917 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1919 def _real_extract(self, url):
1920 # Extract playlist id
1921 mobj = re.match(self._VALID_URL, url)
1923 raise ExtractorError('Invalid URL: %s' % url)
1924 playlist_id = mobj.group(1) or mobj.group(2)
1926 video = self._check_download_just_video(url, playlist_id)
1930 if playlist_id.startswith(('RD', 'UL', 'PU')):
1931 # Mixes require a custom extraction process
1932 return self._extract_mix(playlist_id)
1934 return self._extract_playlist(playlist_id)
1937 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1938 IE_DESC = 'YouTube.com channels'
1939 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1940 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1941 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1942 IE_NAME = 'youtube:channel'
1944 'note': 'paginated channel',
1945 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1946 'playlist_mincount': 91,
1948 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1949 'title': 'Uploads from lex will',
1952 'note': 'Age restricted channel',
1953 # from https://www.youtube.com/user/DeusExOfficial
1954 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1955 'playlist_mincount': 64,
1957 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1958 'title': 'Uploads from Deus Ex',
1963 def suitable(cls, url):
1964 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1965 else super(YoutubeChannelIE, cls).suitable(url))
1967 def _build_template_url(self, url, channel_id):
1968 return self._TEMPLATE_URL % channel_id
1970 def _real_extract(self, url):
1971 channel_id = self._match_id(url)
1973 url = self._build_template_url(url, channel_id)
1975 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1976 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1977 # otherwise fallback on channel by page extraction
1978 channel_page = self._download_webpage(
1979 url + '?view=57', channel_id,
1980 'Downloading channel page', fatal=False)
1981 if channel_page is False:
1982 channel_playlist_id = False
1984 channel_playlist_id = self._html_search_meta(
1985 'channelId', channel_page, 'channel id', default=None)
1986 if not channel_playlist_id:
1987 channel_url = self._html_search_meta(
1988 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
1989 channel_page, 'channel url', default=None)
1991 channel_playlist_id = self._search_regex(
1992 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
1993 channel_url, 'channel id', default=None)
1994 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1995 playlist_id = 'UU' + channel_playlist_id[2:]
1996 return self.url_result(
1997 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1999 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2000 autogenerated = re.search(r'''(?x)
2002 channel-header-autogenerated-label|
2003 yt-channel-title-autogenerated
2004 )[^"]*"''', channel_page) is not None
2007 # The videos are contained in a single page
2008 # the ajax pages can't be used, they are empty
2011 video_id, 'Youtube', video_id=video_id,
2012 video_title=video_title)
2013 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2014 return self.playlist_result(entries, channel_id)
2017 next(self._entries(channel_page, channel_id))
2018 except StopIteration:
2019 alert_message = self._html_search_regex(
2020 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2021 channel_page, 'alert', default=None, group='alert')
2023 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2025 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2028 class YoutubeUserIE(YoutubeChannelIE):
2029 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2030 _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_-]+)'
2031 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2032 IE_NAME = 'youtube:user'
2035 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2036 'playlist_mincount': 320,
2038 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2039 'title': 'Uploads from The Linux Foundation',
2042 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2043 # but not https://www.youtube.com/user/12minuteathlete/videos
2044 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2045 'playlist_mincount': 249,
2047 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2048 'title': 'Uploads from 12 Minute Athlete',
2051 'url': 'ytuser:phihag',
2052 'only_matching': True,
2054 'url': 'https://www.youtube.com/c/gametrailers',
2055 'only_matching': True,
2057 'url': 'https://www.youtube.com/gametrailers',
2058 'only_matching': True,
2060 # This channel is not available.
2061 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2062 'only_matching': True,
2066 def suitable(cls, url):
2067 # Don't return True if the url can be extracted with other youtube
2068 # extractor, the regex would is too permissive and it would match.
2069 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2070 if any(ie.suitable(url) for ie in other_yt_ies):
2073 return super(YoutubeUserIE, cls).suitable(url)
2075 def _build_template_url(self, url, channel_id):
2076 mobj = re.match(self._VALID_URL, url)
2077 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2080 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2081 IE_DESC = 'YouTube.com live streams'
2082 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2083 IE_NAME = 'youtube:live'
2086 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2088 'id': 'a48o2S1cPoo',
2090 'title': 'The Young Turks - Live Main Show',
2091 'uploader': 'The Young Turks',
2092 'uploader_id': 'TheYoungTurks',
2093 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2094 'upload_date': '20150715',
2095 'license': 'Standard YouTube License',
2096 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2097 'categories': ['News & Politics'],
2098 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2100 'dislike_count': int,
2103 'skip_download': True,
2106 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2107 'only_matching': True,
2110 def _real_extract(self, url):
2111 mobj = re.match(self._VALID_URL, url)
2112 channel_id = mobj.group('id')
2113 base_url = mobj.group('base_url')
2114 webpage = self._download_webpage(url, channel_id, fatal=False)
2116 page_type = self._og_search_property(
2117 'type', webpage, 'page type', default=None)
2118 video_id = self._html_search_meta(
2119 'videoId', webpage, 'video id', default=None)
2120 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2121 return self.url_result(video_id, YoutubeIE.ie_key())
2122 return self.url_result(base_url)
2125 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2126 IE_DESC = 'YouTube.com user/channel playlists'
2127 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2128 IE_NAME = 'youtube:playlists'
2131 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2132 'playlist_mincount': 4,
2134 'id': 'ThirstForScience',
2135 'title': 'Thirst for Science',
2138 # with "Load more" button
2139 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2140 'playlist_mincount': 70,
2143 'title': 'Игорь Клейнер',
2146 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2147 'playlist_mincount': 17,
2149 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2150 'title': 'Chem Player',
2155 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2156 IE_DESC = 'YouTube.com searches'
2157 # there doesn't appear to be a real limit, for example if you search for
2158 # 'python' you get more than 8.000.000 results
2159 _MAX_RESULTS = float('inf')
2160 IE_NAME = 'youtube:search'
2161 _SEARCH_KEY = 'ytsearch'
2162 _EXTRA_QUERY_ARGS = {}
2165 def _get_n_results(self, query, n):
2166 """Get a specified number of results for a query"""
2171 for pagenum in itertools.count(1):
2173 'search_query': query.encode('utf-8'),
2177 url_query.update(self._EXTRA_QUERY_ARGS)
2178 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2179 data = self._download_json(
2180 result_url, video_id='query "%s"' % query,
2181 note='Downloading page %s' % pagenum,
2182 errnote='Unable to download API page')
2183 html_content = data[1]['body']['content']
2185 if 'class="search-message' in html_content:
2186 raise ExtractorError(
2187 '[youtube] No video results', expected=True)
2189 new_videos = self._ids_to_results(orderedSet(re.findall(
2190 r'href="/watch\?v=(.{11})', html_content)))
2191 videos += new_videos
2192 if not new_videos or len(videos) > limit:
2197 return self.playlist_result(videos, query)
2200 class YoutubeSearchDateIE(YoutubeSearchIE):
2201 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2202 _SEARCH_KEY = 'ytsearchdate'
2203 IE_DESC = 'YouTube.com searches, newest videos first'
2204 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2207 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2208 IE_DESC = 'YouTube.com search URLs'
2209 IE_NAME = 'youtube:search_url'
2210 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2211 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2213 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2214 'playlist_mincount': 5,
2216 'title': 'youtube-dl test video',
2219 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2220 'only_matching': True,
2223 def _real_extract(self, url):
2224 mobj = re.match(self._VALID_URL, url)
2225 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2226 webpage = self._download_webpage(url, query)
2227 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2230 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2231 IE_DESC = 'YouTube.com (multi-season) shows'
2232 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2233 IE_NAME = 'youtube:show'
2235 'url': 'https://www.youtube.com/show/airdisasters',
2236 'playlist_mincount': 5,
2238 'id': 'airdisasters',
2239 'title': 'Air Disasters',
2243 def _real_extract(self, url):
2244 playlist_id = self._match_id(url)
2245 return super(YoutubeShowIE, self)._real_extract(
2246 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2249 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2251 Base class for feed extractors
2252 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2254 _LOGIN_REQUIRED = True
2258 return 'youtube:%s' % self._FEED_NAME
2260 def _real_initialize(self):
2263 def _real_extract(self, url):
2264 page = self._download_webpage(
2265 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2267 # The extraction process is the same as for playlists, but the regex
2268 # for the video ids doesn't contain an index
2270 more_widget_html = content_html = page
2271 for page_num in itertools.count(1):
2272 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2274 # 'recommended' feed has infinite 'load more' and each new portion spins
2275 # the same videos in (sometimes) slightly different order, so we'll check
2276 # for unicity and break when portion has no new videos
2277 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2283 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2287 more = self._download_json(
2288 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2289 'Downloading page #%s' % page_num,
2290 transform_source=uppercase_escape)
2291 content_html = more['content_html']
2292 more_widget_html = more['load_more_widget_html']
2294 return self.playlist_result(
2295 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2298 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2299 IE_NAME = 'youtube:watchlater'
2300 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2301 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2304 'url': 'https://www.youtube.com/playlist?list=WL',
2305 'only_matching': True,
2307 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2308 'only_matching': True,
2311 def _real_extract(self, url):
2312 video = self._check_download_just_video(url, 'WL')
2315 return self._extract_playlist('WL')
2318 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2319 IE_NAME = 'youtube:favorites'
2320 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2321 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2322 _LOGIN_REQUIRED = True
2324 def _real_extract(self, url):
2325 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2326 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2327 return self.url_result(playlist_id, 'YoutubePlaylist')
2330 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2331 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2332 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2333 _FEED_NAME = 'recommended'
2334 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2337 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2338 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2339 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2340 _FEED_NAME = 'subscriptions'
2341 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2344 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2345 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2346 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2347 _FEED_NAME = 'history'
2348 _PLAYLIST_TITLE = 'Youtube History'
2351 class YoutubeTruncatedURLIE(InfoExtractor):
2352 IE_NAME = 'youtube:truncated_url'
2353 IE_DESC = False # Do not list
2354 _VALID_URL = r'''(?x)
2356 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2359 annotation_id=annotation_[^&]+|
2365 attribution_link\?a=[^&]+
2371 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2372 'only_matching': True,
2374 'url': 'http://www.youtube.com/watch?',
2375 'only_matching': True,
2377 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2378 'only_matching': True,
2380 'url': 'https://www.youtube.com/watch?feature=foo',
2381 'only_matching': True,
2383 'url': 'https://www.youtube.com/watch?hl=en-GB',
2384 'only_matching': True,
2386 'url': 'https://www.youtube.com/watch?t=2372',
2387 'only_matching': True,
2390 def _real_extract(self, url):
2391 raise ExtractorError(
2392 'Did you forget to quote the URL? Remember that & is a meta '
2393 'character in most shells, so you want to put the URL in quotes, '
2395 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2396 ' or simply youtube-dl BaW_jenozKc .',
2400 class YoutubeTruncatedIDIE(InfoExtractor):
2401 IE_NAME = 'youtube:truncated_id'
2402 IE_DESC = False # Do not list
2403 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2406 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2407 'only_matching': True,
2410 def _real_extract(self, url):
2411 video_id = self._match_id(url)
2412 raise ExtractorError(
2413 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),