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)
101 login_results = self._download_webpage(
102 self._PASSWORD_CHALLENGE_URL, None,
103 note='Logging in', errnote='unable to log in', fatal=False,
104 data=urlencode_postdata(login_form))
105 if login_results is False:
108 error_msg = self._html_search_regex(
109 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
110 login_results, 'error message', default=None)
112 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
114 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
115 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
118 # TODO add SMS and phone call support - these require making a request and then prompting the user
120 if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
121 tfa_code = self._get_tfa_info('2-step verification code')
124 self._downloader.report_warning(
125 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
126 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
129 tfa_code = remove_start(tfa_code, 'G-')
131 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
133 tfa_form_strs.update({
138 tfa_data = urlencode_postdata(tfa_form_strs)
140 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
141 tfa_results = self._download_webpage(
143 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
145 if tfa_results is False:
148 if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
149 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
151 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
152 self._downloader.report_warning('unable to log in - did the page structure change?')
154 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
155 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
158 if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
159 self._downloader.report_warning('unable to log in: bad username or password')
163 def _real_initialize(self):
164 if self._downloader is None:
167 if not self._login():
171 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
172 # Extract entries from page with "Load more" button
173 def _entries(self, page, playlist_id):
174 more_widget_html = content_html = page
175 for page_num in itertools.count(1):
176 for entry in self._process_page(content_html):
179 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
183 more = self._download_json(
184 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
185 'Downloading page #%s' % page_num,
186 transform_source=uppercase_escape)
187 content_html = more['content_html']
188 if not content_html.strip():
189 # Some webpages show a "Load more" button but they don't
192 more_widget_html = more['load_more_widget_html']
195 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
196 def _process_page(self, content):
197 for video_id, video_title in self.extract_videos_from_page(content):
198 yield self.url_result(video_id, 'Youtube', video_id, video_title)
200 def extract_videos_from_page(self, page):
203 for mobj in re.finditer(self._VIDEO_RE, page):
204 # The link with index 0 is not the first video of the playlist (not sure if still actual)
205 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
207 video_id = mobj.group('id')
208 video_title = unescapeHTML(mobj.group('title'))
210 video_title = video_title.strip()
212 idx = ids_in_page.index(video_id)
213 if video_title and not titles_in_page[idx]:
214 titles_in_page[idx] = video_title
216 ids_in_page.append(video_id)
217 titles_in_page.append(video_title)
218 return zip(ids_in_page, titles_in_page)
221 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
222 def _process_page(self, content):
223 for playlist_id in orderedSet(re.findall(
224 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
226 yield self.url_result(
227 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
229 def _real_extract(self, url):
230 playlist_id = self._match_id(url)
231 webpage = self._download_webpage(url, playlist_id)
232 title = self._og_search_title(webpage, fatal=False)
233 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
236 class YoutubeIE(YoutubeBaseInfoExtractor):
237 IE_DESC = 'YouTube.com'
238 _VALID_URL = r"""(?x)^
240 (?:https?://|//) # http(s):// or protocol-independent URL
241 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
242 (?:www\.)?deturl\.com/www\.youtube\.com/|
243 (?:www\.)?pwnyoutube\.com/|
244 (?:www\.)?yourepeat\.com/|
245 tube\.majestyc\.net/|
246 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
247 (?:.*?\#/)? # handle anchor (#/) redirect urls
248 (?: # the various things that can precede the ID:
249 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
250 |(?: # or the v= param in all its forms
251 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
252 (?:\?|\#!?) # the params delimiter ? or # or #!
253 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
258 youtu\.be| # just youtu.be/xxxx
259 vid\.plus| # or vid.plus/xxxx
260 zwearz\.com/watch| # or zwearz.com/watch/xxxx
262 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
264 )? # all until now is optional -> you can pass the naked ID
265 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
266 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
267 (?(1).+)? # if we found the ID, everything can follow
269 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
271 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
272 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
273 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
274 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
275 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
276 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
277 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
278 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
279 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
280 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
281 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
282 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
283 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
284 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
285 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
286 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
287 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
288 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
292 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
293 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
294 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
295 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
296 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
297 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
298 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
300 # Apple HTTP Live Streaming
301 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
302 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
303 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
304 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
305 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
306 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
307 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
308 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
311 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
312 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
313 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
314 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
315 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
316 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
317 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
318 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
319 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
320 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
321 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
325 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
326 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
327 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
328 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
331 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
332 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
333 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
334 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
335 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
336 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
337 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
338 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
339 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
340 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
341 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
342 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
343 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
344 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
345 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
346 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
347 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
349 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
350 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
351 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
355 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
356 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
358 # Dash webm audio with opus inside
359 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
360 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
361 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
364 '_rtmp': {'protocol': 'rtmp'},
366 _SUBTITLE_FORMATS = ('ttml', 'vtt')
371 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
375 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
376 'uploader': 'Philipp Hagemeister',
377 'uploader_id': 'phihag',
378 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
379 'upload_date': '20121002',
380 'license': 'Standard YouTube License',
381 '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 .',
382 'categories': ['Science & Technology'],
383 'tags': ['youtube-dl'],
385 'dislike_count': int,
391 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
392 'note': 'Test generic use_cipher_signature video (#897)',
396 'upload_date': '20120506',
397 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
398 'alt_title': 'I Love It (feat. Charli XCX)',
399 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
400 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
401 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
402 'iconic ep', 'iconic', 'love', 'it'],
403 'uploader': 'Icona Pop',
404 'uploader_id': 'IconaPop',
405 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
406 'license': 'Standard YouTube License',
407 'creator': 'Icona Pop',
411 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
412 'note': 'Test VEVO video with age protection (#956)',
416 'upload_date': '20130703',
417 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
418 'alt_title': 'Tunnel Vision',
419 'description': 'md5:64249768eec3bc4276236606ea996373',
420 'uploader': 'justintimberlakeVEVO',
421 'uploader_id': 'justintimberlakeVEVO',
422 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
423 'license': 'Standard YouTube License',
424 'creator': 'Justin Timberlake',
429 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
430 'note': 'Embed-only video (#1746)',
434 'upload_date': '20120608',
435 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
436 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
437 'uploader': 'SET India',
438 'uploader_id': 'setindia',
439 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
440 'license': 'Standard YouTube License',
445 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
446 'note': 'Use the first video ID in the URL',
450 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
451 'uploader': 'Philipp Hagemeister',
452 'uploader_id': 'phihag',
453 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
454 'upload_date': '20121002',
455 'license': 'Standard YouTube License',
456 '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 .',
457 'categories': ['Science & Technology'],
458 'tags': ['youtube-dl'],
460 'dislike_count': int,
463 'skip_download': True,
467 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
468 'note': '256k DASH audio (format 141) via DASH manifest',
472 'upload_date': '20121002',
473 'uploader_id': '8KVIDEO',
474 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
476 'uploader': '8KVIDEO',
477 'license': 'Standard YouTube License',
478 'title': 'UHDTV TEST 8K VIDEO.mp4'
481 'youtube_include_dash_manifest': True,
484 'skip': 'format 141 not served anymore',
486 # DASH manifest with encrypted signature
488 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
492 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
493 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
494 'uploader': 'AfrojackVEVO',
495 'uploader_id': 'AfrojackVEVO',
496 'upload_date': '20131011',
497 'license': 'Standard YouTube License',
500 'youtube_include_dash_manifest': True,
501 'format': '141/bestaudio[ext=m4a]',
504 # JS player signature function name containing $
506 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
510 'title': 'Taylor Swift - Shake It Off',
511 'alt_title': 'Shake It Off',
512 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
513 'uploader': 'TaylorSwiftVEVO',
514 'uploader_id': 'TaylorSwiftVEVO',
515 'upload_date': '20140818',
516 'license': 'Standard YouTube License',
517 'creator': 'Taylor Swift',
520 'youtube_include_dash_manifest': True,
521 'format': '141/bestaudio[ext=m4a]',
526 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
530 'upload_date': '20100909',
531 'uploader': 'The Amazing Atheist',
532 'uploader_id': 'TheAmazingAtheist',
533 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
534 'license': 'Standard YouTube License',
535 'title': 'Burning Everyone\'s Koran',
536 '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',
539 # Normal age-gate video (No vevo, embed allowed)
541 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
545 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
546 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
547 'uploader': 'The Witcher',
548 'uploader_id': 'WitcherGame',
549 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
550 'upload_date': '20140605',
551 'license': 'Standard YouTube License',
555 # Age-gate video with encrypted signature
557 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
561 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
562 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
563 'uploader': 'LloydVEVO',
564 'uploader_id': 'LloydVEVO',
565 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
566 'upload_date': '20110629',
567 'license': 'Standard YouTube License',
571 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
573 'url': '__2ABJjxzNo',
577 'upload_date': '20100430',
578 'uploader_id': 'deadmau5',
579 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
580 'creator': 'deadmau5',
581 'description': 'md5:12c56784b8032162bb936a5f76d55360',
582 'uploader': 'deadmau5',
583 'license': 'Standard YouTube License',
584 'title': 'Deadmau5 - Some Chords (HD)',
585 'alt_title': 'Some Chords',
587 'expected_warnings': [
588 'DASH manifest missing',
591 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
593 'url': 'lqQg6PlCWgI',
597 'upload_date': '20150827',
598 'uploader_id': 'olympic',
599 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
600 'license': 'Standard YouTube License',
601 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
602 'uploader': 'Olympic',
603 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
606 'skip_download': 'requires avconv',
611 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
615 'stretched_ratio': 16 / 9.,
616 'upload_date': '20110310',
617 'uploader_id': 'AllenMeow',
618 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
619 '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',
621 'license': 'Standard YouTube License',
622 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
625 # url_encoded_fmt_stream_map is empty string
627 'url': 'qEJwOuvDf7I',
631 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
633 'upload_date': '20150404',
634 'uploader_id': 'spbelect',
635 'uploader': 'Наблюдатели Петербурга',
638 'skip_download': 'requires avconv',
640 'skip': 'This live event has ended.',
642 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
644 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
648 'title': 'md5:7b81415841e02ecd4313668cde88737a',
649 'description': 'md5:116377fd2963b81ec4ce64b542173306',
650 'upload_date': '20150625',
651 'uploader_id': 'dorappi2000',
652 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
653 'uploader': 'dorappi2000',
654 'license': 'Standard YouTube License',
655 'formats': 'mincount:32',
658 # DASH manifest with segment_list
660 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
661 'md5': '8ce563a1d667b599d21064e982ab9e31',
665 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
666 'uploader': 'Airtek',
667 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
668 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
669 'license': 'Standard YouTube License',
670 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
673 'youtube_include_dash_manifest': True,
674 'format': '135', # bestvideo
676 'skip': 'This live event has ended.',
679 # Multifeed videos (multiple cameras), URL is for Main Camera
680 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
683 'title': 'teamPGP: Rocket League Noob Stream',
684 'description': 'md5:dc7872fb300e143831327f1bae3af010',
690 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
691 'description': 'md5:dc7872fb300e143831327f1bae3af010',
692 'upload_date': '20150721',
693 'uploader': 'Beer Games Beer',
694 'uploader_id': 'beergamesbeer',
695 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
696 'license': 'Standard YouTube License',
702 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
703 'description': 'md5:dc7872fb300e143831327f1bae3af010',
704 'upload_date': '20150721',
705 'uploader': 'Beer Games Beer',
706 'uploader_id': 'beergamesbeer',
707 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
708 'license': 'Standard YouTube License',
714 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
715 'description': 'md5:dc7872fb300e143831327f1bae3af010',
716 'upload_date': '20150721',
717 'uploader': 'Beer Games Beer',
718 'uploader_id': 'beergamesbeer',
719 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
720 'license': 'Standard YouTube License',
726 'title': 'teamPGP: Rocket League Noob Stream (zim)',
727 'description': 'md5:dc7872fb300e143831327f1bae3af010',
728 'upload_date': '20150721',
729 'uploader': 'Beer Games Beer',
730 'uploader_id': 'beergamesbeer',
731 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
732 'license': 'Standard YouTube License',
736 'skip_download': True,
740 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
741 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
744 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
747 'skip': 'Not multifeed anymore',
750 'url': 'http://vid.plus/FlRa-iH7PGw',
751 'only_matching': True,
754 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
755 'only_matching': True,
758 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
759 # Also tests cut-off URL expansion in video description (see
760 # https://github.com/rg3/youtube-dl/issues/1892,
761 # https://github.com/rg3/youtube-dl/issues/8164)
762 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
766 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
767 'alt_title': 'Dark Walk',
768 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
769 'upload_date': '20151119',
770 'uploader_id': 'IronSoulElf',
771 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
772 'uploader': 'IronSoulElf',
773 'license': 'Standard YouTube License',
774 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
777 'skip_download': True,
781 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
782 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
783 'only_matching': True,
786 # Video with yt:stretch=17:0
787 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
791 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
792 'description': 'md5:ee18a25c350637c8faff806845bddee9',
793 'upload_date': '20151107',
794 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
795 'uploader': 'CH GAMER DROID',
798 'skip_download': True,
800 'skip': 'This video does not exist.',
803 # Video licensed under Creative Commons
804 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
808 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
809 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
810 'upload_date': '20150127',
811 'uploader_id': 'BerkmanCenter',
812 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
813 'uploader': 'BerkmanCenter',
814 'license': 'Creative Commons Attribution license (reuse allowed)',
817 'skip_download': True,
821 # Channel-like uploader_url
822 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
826 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
827 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
828 'upload_date': '20151119',
829 'uploader': 'Bernie 2016',
830 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
831 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
832 'license': 'Creative Commons Attribution license (reuse allowed)',
835 'skip_download': True,
839 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
840 'only_matching': True,
843 # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
844 'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
845 'only_matching': True,
849 def __init__(self, *args, **kwargs):
850 super(YoutubeIE, self).__init__(*args, **kwargs)
851 self._player_cache = {}
853 def report_video_info_webpage_download(self, video_id):
854 """Report attempt to download video info webpage."""
855 self.to_screen('%s: Downloading video info webpage' % video_id)
857 def report_information_extraction(self, video_id):
858 """Report attempt to extract video information."""
859 self.to_screen('%s: Extracting video information' % video_id)
861 def report_unavailable_format(self, video_id, format):
862 """Report extracted video URL."""
863 self.to_screen('%s: Format %s not available' % (video_id, format))
865 def report_rtmp_download(self):
866 """Indicate the download will use the RTMP protocol."""
867 self.to_screen('RTMP download detected')
869 def _signature_cache_id(self, example_sig):
870 """ Return a string representation of a signature """
871 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
873 def _extract_signature_function(self, video_id, player_url, example_sig):
875 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
878 raise ExtractorError('Cannot identify player %r' % player_url)
879 player_type = id_m.group('ext')
880 player_id = id_m.group('id')
882 # Read from filesystem cache
883 func_id = '%s_%s_%s' % (
884 player_type, player_id, self._signature_cache_id(example_sig))
885 assert os.path.basename(func_id) == func_id
887 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
888 if cache_spec is not None:
889 return lambda s: ''.join(s[i] for i in cache_spec)
892 'Downloading player %s' % player_url
893 if self._downloader.params.get('verbose') else
894 'Downloading %s player %s' % (player_type, player_id)
896 if player_type == 'js':
897 code = self._download_webpage(
898 player_url, video_id,
900 errnote='Download of %s failed' % player_url)
901 res = self._parse_sig_js(code)
902 elif player_type == 'swf':
903 urlh = self._request_webpage(
904 player_url, video_id,
906 errnote='Download of %s failed' % player_url)
908 res = self._parse_sig_swf(code)
910 assert False, 'Invalid player type %r' % player_type
912 test_string = ''.join(map(compat_chr, range(len(example_sig))))
913 cache_res = res(test_string)
914 cache_spec = [ord(c) for c in cache_res]
916 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
919 def _print_sig_code(self, func, example_sig):
920 def gen_sig_code(idxs):
921 def _genslice(start, end, step):
922 starts = '' if start == 0 else str(start)
923 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
924 steps = '' if step == 1 else (':%d' % step)
925 return 's[%s%s%s]' % (starts, ends, steps)
928 # Quelch pyflakes warnings - start will be set when step is set
929 start = '(Never used)'
930 for i, prev in zip(idxs[1:], idxs[:-1]):
934 yield _genslice(start, prev, step)
937 if i - prev in [-1, 1]:
946 yield _genslice(start, i, step)
948 test_string = ''.join(map(compat_chr, range(len(example_sig))))
949 cache_res = func(test_string)
950 cache_spec = [ord(c) for c in cache_res]
951 expr_code = ' + '.join(gen_sig_code(cache_spec))
952 signature_id_tuple = '(%s)' % (
953 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
954 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
955 ' return %s\n') % (signature_id_tuple, expr_code)
956 self.to_screen('Extracted signature function:\n' + code)
958 def _parse_sig_js(self, jscode):
959 funcname = self._search_regex(
960 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
961 'Initial JS player signature function name')
963 jsi = JSInterpreter(jscode)
964 initial_function = jsi.extract_function(funcname)
965 return lambda s: initial_function([s])
967 def _parse_sig_swf(self, file_contents):
968 swfi = SWFInterpreter(file_contents)
969 TARGET_CLASSNAME = 'SignatureDecipher'
970 searched_class = swfi.extract_class(TARGET_CLASSNAME)
971 initial_function = swfi.extract_function(searched_class, 'decipher')
972 return lambda s: initial_function([s])
974 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
975 """Turn the encrypted s field into a working signature"""
977 if player_url is None:
978 raise ExtractorError('Cannot decrypt signature without player_url')
980 if player_url.startswith('//'):
981 player_url = 'https:' + player_url
983 player_id = (player_url, self._signature_cache_id(s))
984 if player_id not in self._player_cache:
985 func = self._extract_signature_function(
986 video_id, player_url, s
988 self._player_cache[player_id] = func
989 func = self._player_cache[player_id]
990 if self._downloader.params.get('youtube_print_sig_code'):
991 self._print_sig_code(func, s)
993 except Exception as e:
994 tb = traceback.format_exc()
995 raise ExtractorError(
996 'Signature extraction failed: ' + tb, cause=e)
998 def _get_subtitles(self, video_id, webpage):
1000 subs_doc = self._download_xml(
1001 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1002 video_id, note=False)
1003 except ExtractorError as err:
1004 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1008 for track in subs_doc.findall('track'):
1009 lang = track.attrib['lang_code']
1010 if lang in sub_lang_list:
1013 for ext in self._SUBTITLE_FORMATS:
1014 params = compat_urllib_parse_urlencode({
1018 'name': track.attrib['name'].encode('utf-8'),
1020 sub_formats.append({
1021 'url': 'https://www.youtube.com/api/timedtext?' + params,
1024 sub_lang_list[lang] = sub_formats
1025 if not sub_lang_list:
1026 self._downloader.report_warning('video doesn\'t have subtitles')
1028 return sub_lang_list
1030 def _get_ytplayer_config(self, video_id, webpage):
1032 # User data may contain arbitrary character sequences that may affect
1033 # JSON extraction with regex, e.g. when '};' is contained the second
1034 # regex won't capture the whole JSON. Yet working around by trying more
1035 # concrete regex first keeping in mind proper quoted string handling
1036 # to be implemented in future that will replace this workaround (see
1037 # https://github.com/rg3/youtube-dl/issues/7468,
1038 # https://github.com/rg3/youtube-dl/pull/7599)
1039 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1040 r';ytplayer\.config\s*=\s*({.+?});',
1042 config = self._search_regex(
1043 patterns, webpage, 'ytplayer.config', default=None)
1045 return self._parse_json(
1046 uppercase_escape(config), video_id, fatal=False)
1048 def _get_automatic_captions(self, video_id, webpage):
1049 """We need the webpage for getting the captions url, pass it as an
1050 argument to speed up the process."""
1051 self.to_screen('%s: Looking for automatic captions' % video_id)
1052 player_config = self._get_ytplayer_config(video_id, webpage)
1053 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1054 if not player_config:
1055 self._downloader.report_warning(err_msg)
1058 args = player_config['args']
1059 caption_url = args.get('ttsurl')
1061 timestamp = args['timestamp']
1062 # We get the available subtitles
1063 list_params = compat_urllib_parse_urlencode({
1068 list_url = caption_url + '&' + list_params
1069 caption_list = self._download_xml(list_url, video_id)
1070 original_lang_node = caption_list.find('track')
1071 if original_lang_node is None:
1072 self._downloader.report_warning('Video doesn\'t have automatic captions')
1074 original_lang = original_lang_node.attrib['lang_code']
1075 caption_kind = original_lang_node.attrib.get('kind', '')
1078 for lang_node in caption_list.findall('target'):
1079 sub_lang = lang_node.attrib['lang_code']
1081 for ext in self._SUBTITLE_FORMATS:
1082 params = compat_urllib_parse_urlencode({
1083 'lang': original_lang,
1087 'kind': caption_kind,
1089 sub_formats.append({
1090 'url': caption_url + '&' + params,
1093 sub_lang_list[sub_lang] = sub_formats
1094 return sub_lang_list
1096 # Some videos don't provide ttsurl but rather caption_tracks and
1097 # caption_translation_languages (e.g. 20LmZk1hakA)
1098 caption_tracks = args['caption_tracks']
1099 caption_translation_languages = args['caption_translation_languages']
1100 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1101 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1102 caption_qs = compat_parse_qs(parsed_caption_url.query)
1105 for lang in caption_translation_languages.split(','):
1106 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1107 sub_lang = lang_qs.get('lc', [None])[0]
1111 for ext in self._SUBTITLE_FORMATS:
1113 'tlang': [sub_lang],
1116 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1117 query=compat_urllib_parse_urlencode(caption_qs, True)))
1118 sub_formats.append({
1122 sub_lang_list[sub_lang] = sub_formats
1123 return sub_lang_list
1124 # An extractor error can be raise by the download process if there are
1125 # no automatic captions but there are subtitles
1126 except (KeyError, ExtractorError):
1127 self._downloader.report_warning(err_msg)
1130 def _mark_watched(self, video_id, video_info):
1131 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1132 if not playback_url:
1134 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1135 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1137 # cpn generation algorithm is reverse engineered from base.js.
1138 # In fact it works even with dummy cpn.
1139 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1140 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1146 playback_url = compat_urlparse.urlunparse(
1147 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1149 self._download_webpage(
1150 playback_url, video_id, 'Marking watched',
1151 'Unable to mark watched', fatal=False)
1154 def extract_id(cls, url):
1155 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1157 raise ExtractorError('Invalid URL: %s' % url)
1158 video_id = mobj.group(2)
1161 def _extract_from_m3u8(self, manifest_url, video_id):
1164 def _get_urls(_manifest):
1165 lines = _manifest.split('\n')
1166 urls = filter(lambda l: l and not l.startswith('#'),
1169 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1170 formats_urls = _get_urls(manifest)
1171 for format_url in formats_urls:
1172 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1173 url_map[itag] = format_url
1176 def _extract_annotations(self, video_id):
1177 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1178 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1180 def _real_extract(self, url):
1181 url, smuggled_data = unsmuggle_url(url, {})
1184 'http' if self._downloader.params.get('prefer_insecure', False)
1189 parsed_url = compat_urllib_parse_urlparse(url)
1190 for component in [parsed_url.fragment, parsed_url.query]:
1191 query = compat_parse_qs(component)
1192 if start_time is None and 't' in query:
1193 start_time = parse_duration(query['t'][0])
1194 if start_time is None and 'start' in query:
1195 start_time = parse_duration(query['start'][0])
1196 if end_time is None and 'end' in query:
1197 end_time = parse_duration(query['end'][0])
1199 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1200 mobj = re.search(self._NEXT_URL_RE, url)
1202 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1203 video_id = self.extract_id(url)
1206 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1207 video_webpage = self._download_webpage(url, video_id)
1209 # Attempt to extract SWF player URL
1210 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1211 if mobj is not None:
1212 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1218 def add_dash_mpd(video_info):
1219 dash_mpd = video_info.get('dashmpd')
1220 if dash_mpd and dash_mpd[0] not in dash_mpds:
1221 dash_mpds.append(dash_mpd[0])
1224 embed_webpage = None
1226 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1228 # We simulate the access to the video from www.youtube.com/v/{video_id}
1229 # this can be viewed without login into Youtube
1230 url = proto + '://www.youtube.com/embed/%s' % video_id
1231 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1232 data = compat_urllib_parse_urlencode({
1233 'video_id': video_id,
1234 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1235 'sts': self._search_regex(
1236 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1238 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1239 video_info_webpage = self._download_webpage(
1240 video_info_url, video_id,
1241 note='Refetching age-gated info webpage',
1242 errnote='unable to download video info webpage')
1243 video_info = compat_parse_qs(video_info_webpage)
1244 add_dash_mpd(video_info)
1248 # Try looking directly into the video webpage
1249 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1251 args = ytplayer_config['args']
1252 if args.get('url_encoded_fmt_stream_map'):
1253 # Convert to the same format returned by compat_parse_qs
1254 video_info = dict((k, [v]) for k, v in args.items())
1255 add_dash_mpd(video_info)
1256 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1258 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1259 # We also try looking in get_video_info since it may contain different dashmpd
1260 # URL that points to a DASH manifest with possibly different itag set (some itags
1261 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1262 # manifest pointed by get_video_info's dashmpd).
1263 # The general idea is to take a union of itags of both DASH manifests (for example
1264 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1265 self.report_video_info_webpage_download(video_id)
1266 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1268 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1269 % (proto, video_id, el_type))
1270 video_info_webpage = self._download_webpage(
1272 video_id, note=False,
1273 errnote='unable to download video info webpage')
1274 get_video_info = compat_parse_qs(video_info_webpage)
1275 if get_video_info.get('use_cipher_signature') != ['True']:
1276 add_dash_mpd(get_video_info)
1278 video_info = get_video_info
1279 if 'token' in get_video_info:
1280 # Different get_video_info requests may report different results, e.g.
1281 # some may report video unavailability, but some may serve it without
1282 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1283 # the original webpage as well as el=info and el=embedded get_video_info
1284 # requests report video unavailability due to geo restriction while
1285 # el=detailpage succeeds and returns valid data). This is probably
1286 # due to YouTube measures against IP ranges of hosting providers.
1287 # Working around by preferring the first succeeded video_info containing
1288 # the token if no such video_info yet was found.
1289 if 'token' not in video_info:
1290 video_info = get_video_info
1292 if 'token' not in video_info:
1293 if 'reason' in video_info:
1294 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1295 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1297 raise ExtractorError('YouTube said: This video is available in %s only' % (
1298 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1300 raise ExtractorError(
1301 'YouTube said: %s' % video_info['reason'][0],
1302 expected=True, video_id=video_id)
1304 raise ExtractorError(
1305 '"token" parameter not in video info for unknown reason',
1309 if 'title' in video_info:
1310 video_title = video_info['title'][0]
1312 self._downloader.report_warning('Unable to extract video title')
1316 video_description = get_element_by_id("eow-description", video_webpage)
1317 if video_description:
1318 video_description = re.sub(r'''(?x)
1320 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1321 (?:title|href)="([^"]+)"\s+
1322 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1326 ''', r'\1', video_description)
1327 video_description = clean_html(video_description)
1329 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1331 video_description = unescapeHTML(fd_mobj.group(1))
1333 video_description = ''
1335 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1336 if not self._downloader.params.get('noplaylist'):
1339 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1340 for feed in multifeed_metadata_list.split(','):
1341 # Unquote should take place before split on comma (,) since textual
1342 # fields may contain comma as well (see
1343 # https://github.com/rg3/youtube-dl/issues/8536)
1344 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1346 '_type': 'url_transparent',
1347 'ie_key': 'Youtube',
1349 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1350 {'force_singlefeed': True}),
1351 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1353 feed_ids.append(feed_data['id'][0])
1355 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1356 % (', '.join(feed_ids), video_id))
1357 return self.playlist_result(entries, video_id, video_title, video_description)
1358 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1360 if 'view_count' in video_info:
1361 view_count = int(video_info['view_count'][0])
1365 # Check for "rental" videos
1366 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1367 raise ExtractorError('"rental" videos not supported')
1369 # Start extracting information
1370 self.report_information_extraction(video_id)
1373 if 'author' not in video_info:
1374 raise ExtractorError('Unable to extract uploader name')
1375 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1378 video_uploader_id = None
1379 video_uploader_url = None
1381 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1383 if mobj is not None:
1384 video_uploader_id = mobj.group('uploader_id')
1385 video_uploader_url = mobj.group('uploader_url')
1387 self._downloader.report_warning('unable to extract uploader nickname')
1390 # We try first to get a high quality image:
1391 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1392 video_webpage, re.DOTALL)
1393 if m_thumb is not None:
1394 video_thumbnail = m_thumb.group(1)
1395 elif 'thumbnail_url' not in video_info:
1396 self._downloader.report_warning('unable to extract video thumbnail')
1397 video_thumbnail = None
1398 else: # don't panic if we can't find it
1399 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1402 upload_date = self._html_search_meta(
1403 'datePublished', video_webpage, 'upload date', default=None)
1405 upload_date = self._search_regex(
1406 [r'(?s)id="eow-date.*?>(.*?)</span>',
1407 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1408 video_webpage, 'upload date', default=None)
1410 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1411 upload_date = unified_strdate(upload_date)
1413 video_license = self._html_search_regex(
1414 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1415 video_webpage, 'license', default=None)
1417 m_music = re.search(
1418 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1421 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1422 video_creator = clean_html(m_music.group('creator'))
1424 video_alt_title = video_creator = None
1426 m_cat_container = self._search_regex(
1427 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1428 video_webpage, 'categories', default=None)
1430 category = self._html_search_regex(
1431 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1433 video_categories = None if category is None else [category]
1435 video_categories = None
1438 unescapeHTML(m.group('content'))
1439 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1441 def _extract_count(count_name):
1442 return str_to_int(self._search_regex(
1443 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1444 % re.escape(count_name),
1445 video_webpage, count_name, default=None))
1447 like_count = _extract_count('like')
1448 dislike_count = _extract_count('dislike')
1451 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1452 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1454 if 'length_seconds' not in video_info:
1455 self._downloader.report_warning('unable to extract video duration')
1456 video_duration = None
1458 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1461 video_annotations = None
1462 if self._downloader.params.get('writeannotations', False):
1463 video_annotations = self._extract_annotations(video_id)
1465 def _map_to_format_list(urlmap):
1467 for itag, video_real_url in urlmap.items():
1470 'url': video_real_url,
1471 'player_url': player_url,
1473 if itag in self._formats:
1474 dct.update(self._formats[itag])
1478 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1479 self.report_rtmp_download()
1481 'format_id': '_rtmp',
1483 'url': video_info['conn'][0],
1484 'player_url': player_url,
1486 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1487 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1488 if 'rtmpe%3Dyes' in encoded_url_map:
1489 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1491 fmt_list = video_info.get('fmt_list', [''])[0]
1493 for fmt in fmt_list.split(','):
1494 spec = fmt.split('/')
1496 width_height = spec[1].split('x')
1497 if len(width_height) == 2:
1498 formats_spec[spec[0]] = {
1499 'resolution': spec[1],
1500 'width': int_or_none(width_height[0]),
1501 'height': int_or_none(width_height[1]),
1504 for url_data_str in encoded_url_map.split(','):
1505 url_data = compat_parse_qs(url_data_str)
1506 if 'itag' not in url_data or 'url' not in url_data:
1508 format_id = url_data['itag'][0]
1509 url = url_data['url'][0]
1511 if 'sig' in url_data:
1512 url += '&signature=' + url_data['sig'][0]
1513 elif 's' in url_data:
1514 encrypted_sig = url_data['s'][0]
1515 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1517 jsplayer_url_json = self._search_regex(
1519 embed_webpage if age_gate else video_webpage,
1520 'JS player URL (1)', default=None)
1521 if not jsplayer_url_json and not age_gate:
1522 # We need the embed website after all
1523 if embed_webpage is None:
1524 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1525 embed_webpage = self._download_webpage(
1526 embed_url, video_id, 'Downloading embed webpage')
1527 jsplayer_url_json = self._search_regex(
1528 ASSETS_RE, embed_webpage, 'JS player URL')
1530 player_url = json.loads(jsplayer_url_json)
1531 if player_url is None:
1532 player_url_json = self._search_regex(
1533 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1534 video_webpage, 'age gate player URL')
1535 player_url = json.loads(player_url_json)
1537 if self._downloader.params.get('verbose'):
1538 if player_url is None:
1539 player_version = 'unknown'
1540 player_desc = 'unknown'
1542 if player_url.endswith('swf'):
1543 player_version = self._search_regex(
1544 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1545 'flash player', fatal=False)
1546 player_desc = 'flash player %s' % player_version
1548 player_version = self._search_regex(
1549 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1551 'html5 player', fatal=False)
1552 player_desc = 'html5 player %s' % player_version
1554 parts_sizes = self._signature_cache_id(encrypted_sig)
1555 self.to_screen('{%s} signature length %s, %s' %
1556 (format_id, parts_sizes, player_desc))
1558 signature = self._decrypt_signature(
1559 encrypted_sig, video_id, player_url, age_gate)
1560 url += '&signature=' + signature
1561 if 'ratebypass' not in url:
1562 url += '&ratebypass=yes'
1565 'format_id': format_id,
1567 'player_url': player_url,
1569 if format_id in self._formats:
1570 dct.update(self._formats[format_id])
1571 if format_id in formats_spec:
1572 dct.update(formats_spec[format_id])
1574 # Some itags are not included in DASH manifest thus corresponding formats will
1575 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1576 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1577 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1578 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1581 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1582 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1585 'fps': int_or_none(url_data.get('fps', [None])[0]),
1586 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1588 for key, value in more_fields.items():
1591 type_ = url_data.get('type', [None])[0]
1593 type_split = type_.split(';')
1594 kind_ext = type_split[0].split('/')
1595 if len(kind_ext) == 2:
1597 dct['ext'] = mimetype2ext(type_split[0])
1598 if kind in ('audio', 'video'):
1600 for mobj in re.finditer(
1601 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1602 if mobj.group('key') == 'codecs':
1603 codecs = mobj.group('val')
1606 codecs = codecs.split(',')
1607 if len(codecs) == 2:
1608 acodec, vcodec = codecs[1], codecs[0]
1610 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1616 elif video_info.get('hlsvp'):
1617 manifest_url = video_info['hlsvp'][0]
1618 url_map = self._extract_from_m3u8(manifest_url, video_id)
1619 formats = _map_to_format_list(url_map)
1620 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1621 for a_format in formats:
1622 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1624 unavailable_message = self._html_search_regex(
1625 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1626 video_webpage, 'unavailable message', default=None)
1627 if unavailable_message:
1628 raise ExtractorError(unavailable_message, expected=True)
1629 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1631 # Look for the DASH manifest
1632 if self._downloader.params.get('youtube_include_dash_manifest', True):
1633 dash_mpd_fatal = True
1634 for mpd_url in dash_mpds:
1637 def decrypt_sig(mobj):
1639 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1640 return '/signature/%s' % dec_s
1642 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1644 for df in self._extract_mpd_formats(
1645 mpd_url, video_id, fatal=dash_mpd_fatal,
1646 formats_dict=self._formats):
1647 # Do not overwrite DASH format found in some previous DASH manifest
1648 if df['format_id'] not in dash_formats:
1649 dash_formats[df['format_id']] = df
1650 # Additional DASH manifests may end up in HTTP Error 403 therefore
1651 # allow them to fail without bug report message if we already have
1652 # some DASH manifest succeeded. This is temporary workaround to reduce
1653 # burst of bug reports until we figure out the reason and whether it
1654 # can be fixed at all.
1655 dash_mpd_fatal = False
1656 except (ExtractorError, KeyError) as e:
1657 self.report_warning(
1658 'Skipping DASH manifest: %r' % e, video_id)
1660 # Remove the formats we found through non-DASH, they
1661 # contain less info and it can be wrong, because we use
1662 # fixed values (for example the resolution). See
1663 # https://github.com/rg3/youtube-dl/issues/5774 for an
1665 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1666 formats.extend(dash_formats.values())
1668 # Check for malformed aspect ratio
1669 stretched_m = re.search(
1670 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1673 w = float(stretched_m.group('w'))
1674 h = float(stretched_m.group('h'))
1675 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1676 # We will only process correct ratios.
1680 if f.get('vcodec') != 'none':
1681 f['stretched_ratio'] = ratio
1683 self._sort_formats(formats)
1685 self.mark_watched(video_id, video_info)
1689 'uploader': video_uploader,
1690 'uploader_id': video_uploader_id,
1691 'uploader_url': video_uploader_url,
1692 'upload_date': upload_date,
1693 'license': video_license,
1694 'creator': video_creator,
1695 'title': video_title,
1696 'alt_title': video_alt_title,
1697 'thumbnail': video_thumbnail,
1698 'description': video_description,
1699 'categories': video_categories,
1701 'subtitles': video_subtitles,
1702 'automatic_captions': automatic_captions,
1703 'duration': video_duration,
1704 'age_limit': 18 if age_gate else 0,
1705 'annotations': video_annotations,
1706 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1707 'view_count': view_count,
1708 'like_count': like_count,
1709 'dislike_count': dislike_count,
1710 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1713 'start_time': start_time,
1714 'end_time': end_time,
1718 class YoutubeSharedVideoIE(InfoExtractor):
1719 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?.*\bci=(?P<id>[0-9A-Za-z_-]{11})'
1720 IE_NAME = 'youtube:shared'
1723 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1725 'id': 'uPDB5I9wfp8',
1727 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1728 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1729 'upload_date': '20160219',
1730 'uploader': 'Pocoyo - Português (BR)',
1731 'uploader_id': 'PocoyoBrazil',
1733 'add_ie': ['Youtube'],
1735 # There are already too many Youtube downloads
1736 'skip_download': True,
1740 def _real_extract(self, url):
1741 video_id = self._match_id(url)
1743 webpage = self._download_webpage(url, video_id)
1745 real_video_id = self._html_search_meta(
1746 'videoId', webpage, 'YouTube video id', fatal=True)
1748 return self.url_result(real_video_id, YoutubeIE.ie_key())
1751 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1752 IE_DESC = 'YouTube.com playlists'
1753 _VALID_URL = r"""(?x)(?:
1758 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1759 \? (?:.*?[&;])*? (?:p|a|list)=
1763 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1764 # Top tracks, they can also include dots
1769 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1771 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1772 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1773 IE_NAME = 'youtube:playlist'
1775 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1777 'title': 'ytdl test PL',
1778 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1780 'playlist_count': 3,
1782 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1784 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1785 'title': 'YDL_Empty_List',
1787 'playlist_count': 0,
1789 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1790 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1792 'title': '29C3: Not my department',
1793 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1795 'playlist_count': 95,
1797 'note': 'issue #673',
1798 'url': 'PLBB231211A4F62143',
1800 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1801 'id': 'PLBB231211A4F62143',
1803 'playlist_mincount': 26,
1805 'note': 'Large playlist',
1806 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1808 'title': 'Uploads from Cauchemar',
1809 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1811 'playlist_mincount': 799,
1813 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1815 'title': 'YDL_safe_search',
1816 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1818 'playlist_count': 2,
1821 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1822 'playlist_count': 4,
1825 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1828 'note': 'Embedded SWF player',
1829 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1830 'playlist_count': 4,
1833 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1836 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1837 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1839 'title': 'Uploads from Interstellar Movie',
1840 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1842 'playlist_mincout': 21,
1845 def _real_initialize(self):
1848 def _extract_mix(self, playlist_id):
1849 # The mixes are generated from a single video
1850 # the id of the playlist is just 'RD' + video_id
1852 last_id = playlist_id[-11:]
1853 for n in itertools.count(1):
1854 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1855 webpage = self._download_webpage(
1856 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1857 new_ids = orderedSet(re.findall(
1858 r'''(?xs)data-video-username=".*?".*?
1859 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1861 # Fetch new pages until all the videos are repeated, it seems that
1862 # there are always 51 unique videos.
1863 new_ids = [_id for _id in new_ids if _id not in ids]
1869 url_results = self._ids_to_results(ids)
1871 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1873 search_title('playlist-title') or
1874 search_title('title long-title') or
1875 search_title('title'))
1876 title = clean_html(title_span)
1878 return self.playlist_result(url_results, playlist_id, title)
1880 def _extract_playlist(self, playlist_id):
1881 url = self._TEMPLATE_URL % playlist_id
1882 page = self._download_webpage(url, playlist_id)
1884 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1885 match = match.strip()
1886 # Check if the playlist exists or is private
1887 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1888 raise ExtractorError(
1889 'The playlist doesn\'t exist or is private, use --username or '
1890 '--netrc to access it.',
1892 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1893 raise ExtractorError(
1894 'Invalid parameters. Maybe URL is incorrect.',
1896 elif re.match(r'[^<]*Choose your language[^<]*', match):
1899 self.report_warning('Youtube gives an alert message: ' + match)
1901 playlist_title = self._html_search_regex(
1902 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1905 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1907 def _check_download_just_video(self, url, playlist_id):
1908 # Check if it's a video-specific URL
1909 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1910 if 'v' in query_dict:
1911 video_id = query_dict['v'][0]
1912 if self._downloader.params.get('noplaylist'):
1913 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1914 return self.url_result(video_id, 'Youtube', video_id=video_id)
1916 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1918 def _real_extract(self, url):
1919 # Extract playlist id
1920 mobj = re.match(self._VALID_URL, url)
1922 raise ExtractorError('Invalid URL: %s' % url)
1923 playlist_id = mobj.group(1) or mobj.group(2)
1925 video = self._check_download_just_video(url, playlist_id)
1929 if playlist_id.startswith(('RD', 'UL', 'PU')):
1930 # Mixes require a custom extraction process
1931 return self._extract_mix(playlist_id)
1933 return self._extract_playlist(playlist_id)
1936 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1937 IE_DESC = 'YouTube.com channels'
1938 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1939 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1940 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1941 IE_NAME = 'youtube:channel'
1943 'note': 'paginated channel',
1944 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1945 'playlist_mincount': 91,
1947 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1948 'title': 'Uploads from lex will',
1951 'note': 'Age restricted channel',
1952 # from https://www.youtube.com/user/DeusExOfficial
1953 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1954 'playlist_mincount': 64,
1956 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1957 'title': 'Uploads from Deus Ex',
1962 def suitable(cls, url):
1963 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1964 else super(YoutubeChannelIE, cls).suitable(url))
1966 def _build_template_url(self, url, channel_id):
1967 return self._TEMPLATE_URL % channel_id
1969 def _real_extract(self, url):
1970 channel_id = self._match_id(url)
1972 url = self._build_template_url(url, channel_id)
1974 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1975 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1976 # otherwise fallback on channel by page extraction
1977 channel_page = self._download_webpage(
1978 url + '?view=57', channel_id,
1979 'Downloading channel page', fatal=False)
1980 if channel_page is False:
1981 channel_playlist_id = False
1983 channel_playlist_id = self._html_search_meta(
1984 'channelId', channel_page, 'channel id', default=None)
1985 if not channel_playlist_id:
1986 channel_url = self._html_search_meta(
1987 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
1988 channel_page, 'channel url', default=None)
1990 channel_playlist_id = self._search_regex(
1991 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
1992 channel_url, 'channel id', default=None)
1993 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1994 playlist_id = 'UU' + channel_playlist_id[2:]
1995 return self.url_result(
1996 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1998 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1999 autogenerated = re.search(r'''(?x)
2001 channel-header-autogenerated-label|
2002 yt-channel-title-autogenerated
2003 )[^"]*"''', channel_page) is not None
2006 # The videos are contained in a single page
2007 # the ajax pages can't be used, they are empty
2010 video_id, 'Youtube', video_id=video_id,
2011 video_title=video_title)
2012 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2013 return self.playlist_result(entries, channel_id)
2016 next(self._entries(channel_page, channel_id))
2017 except StopIteration:
2018 alert_message = self._html_search_regex(
2019 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2020 channel_page, 'alert', default=None, group='alert')
2022 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2024 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2027 class YoutubeUserIE(YoutubeChannelIE):
2028 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2029 _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_-]+)'
2030 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2031 IE_NAME = 'youtube:user'
2034 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2035 'playlist_mincount': 320,
2037 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2038 'title': 'Uploads from The Linux Foundation',
2041 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2042 # but not https://www.youtube.com/user/12minuteathlete/videos
2043 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2044 'playlist_mincount': 249,
2046 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2047 'title': 'Uploads from 12 Minute Athlete',
2050 'url': 'ytuser:phihag',
2051 'only_matching': True,
2053 'url': 'https://www.youtube.com/c/gametrailers',
2054 'only_matching': True,
2056 'url': 'https://www.youtube.com/gametrailers',
2057 'only_matching': True,
2059 # This channel is not available.
2060 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2061 'only_matching': True,
2065 def suitable(cls, url):
2066 # Don't return True if the url can be extracted with other youtube
2067 # extractor, the regex would is too permissive and it would match.
2068 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2069 if any(ie.suitable(url) for ie in other_yt_ies):
2072 return super(YoutubeUserIE, cls).suitable(url)
2074 def _build_template_url(self, url, channel_id):
2075 mobj = re.match(self._VALID_URL, url)
2076 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2079 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2080 IE_DESC = 'YouTube.com live streams'
2081 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2082 IE_NAME = 'youtube:live'
2085 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2087 'id': 'a48o2S1cPoo',
2089 'title': 'The Young Turks - Live Main Show',
2090 'uploader': 'The Young Turks',
2091 'uploader_id': 'TheYoungTurks',
2092 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2093 'upload_date': '20150715',
2094 'license': 'Standard YouTube License',
2095 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2096 'categories': ['News & Politics'],
2097 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2099 'dislike_count': int,
2102 'skip_download': True,
2105 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2106 'only_matching': True,
2109 def _real_extract(self, url):
2110 mobj = re.match(self._VALID_URL, url)
2111 channel_id = mobj.group('id')
2112 base_url = mobj.group('base_url')
2113 webpage = self._download_webpage(url, channel_id, fatal=False)
2115 page_type = self._og_search_property(
2116 'type', webpage, 'page type', default=None)
2117 video_id = self._html_search_meta(
2118 'videoId', webpage, 'video id', default=None)
2119 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2120 return self.url_result(video_id, YoutubeIE.ie_key())
2121 return self.url_result(base_url)
2124 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2125 IE_DESC = 'YouTube.com user/channel playlists'
2126 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2127 IE_NAME = 'youtube:playlists'
2130 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2131 'playlist_mincount': 4,
2133 'id': 'ThirstForScience',
2134 'title': 'Thirst for Science',
2137 # with "Load more" button
2138 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2139 'playlist_mincount': 70,
2142 'title': 'Игорь Клейнер',
2145 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2146 'playlist_mincount': 17,
2148 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2149 'title': 'Chem Player',
2154 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2155 IE_DESC = 'YouTube.com searches'
2156 # there doesn't appear to be a real limit, for example if you search for
2157 # 'python' you get more than 8.000.000 results
2158 _MAX_RESULTS = float('inf')
2159 IE_NAME = 'youtube:search'
2160 _SEARCH_KEY = 'ytsearch'
2161 _EXTRA_QUERY_ARGS = {}
2164 def _get_n_results(self, query, n):
2165 """Get a specified number of results for a query"""
2170 for pagenum in itertools.count(1):
2172 'search_query': query.encode('utf-8'),
2176 url_query.update(self._EXTRA_QUERY_ARGS)
2177 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2178 data = self._download_json(
2179 result_url, video_id='query "%s"' % query,
2180 note='Downloading page %s' % pagenum,
2181 errnote='Unable to download API page')
2182 html_content = data[1]['body']['content']
2184 if 'class="search-message' in html_content:
2185 raise ExtractorError(
2186 '[youtube] No video results', expected=True)
2188 new_videos = self._ids_to_results(orderedSet(re.findall(
2189 r'href="/watch\?v=(.{11})', html_content)))
2190 videos += new_videos
2191 if not new_videos or len(videos) > limit:
2196 return self.playlist_result(videos, query)
2199 class YoutubeSearchDateIE(YoutubeSearchIE):
2200 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2201 _SEARCH_KEY = 'ytsearchdate'
2202 IE_DESC = 'YouTube.com searches, newest videos first'
2203 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2206 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2207 IE_DESC = 'YouTube.com search URLs'
2208 IE_NAME = 'youtube:search_url'
2209 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2210 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2212 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2213 'playlist_mincount': 5,
2215 'title': 'youtube-dl test video',
2218 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2219 'only_matching': True,
2222 def _real_extract(self, url):
2223 mobj = re.match(self._VALID_URL, url)
2224 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2225 webpage = self._download_webpage(url, query)
2226 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2229 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2230 IE_DESC = 'YouTube.com (multi-season) shows'
2231 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2232 IE_NAME = 'youtube:show'
2234 'url': 'https://www.youtube.com/show/airdisasters',
2235 'playlist_mincount': 5,
2237 'id': 'airdisasters',
2238 'title': 'Air Disasters',
2242 def _real_extract(self, url):
2243 playlist_id = self._match_id(url)
2244 return super(YoutubeShowIE, self)._real_extract(
2245 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2248 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2250 Base class for feed extractors
2251 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2253 _LOGIN_REQUIRED = True
2257 return 'youtube:%s' % self._FEED_NAME
2259 def _real_initialize(self):
2262 def _real_extract(self, url):
2263 page = self._download_webpage(
2264 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2266 # The extraction process is the same as for playlists, but the regex
2267 # for the video ids doesn't contain an index
2269 more_widget_html = content_html = page
2270 for page_num in itertools.count(1):
2271 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2273 # 'recommended' feed has infinite 'load more' and each new portion spins
2274 # the same videos in (sometimes) slightly different order, so we'll check
2275 # for unicity and break when portion has no new videos
2276 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2282 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2286 more = self._download_json(
2287 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2288 'Downloading page #%s' % page_num,
2289 transform_source=uppercase_escape)
2290 content_html = more['content_html']
2291 more_widget_html = more['load_more_widget_html']
2293 return self.playlist_result(
2294 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2297 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2298 IE_NAME = 'youtube:watchlater'
2299 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2300 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2303 'url': 'https://www.youtube.com/playlist?list=WL',
2304 'only_matching': True,
2306 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2307 'only_matching': True,
2310 def _real_extract(self, url):
2311 video = self._check_download_just_video(url, 'WL')
2314 return self._extract_playlist('WL')
2317 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2318 IE_NAME = 'youtube:favorites'
2319 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2320 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2321 _LOGIN_REQUIRED = True
2323 def _real_extract(self, url):
2324 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2325 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2326 return self.url_result(playlist_id, 'YoutubePlaylist')
2329 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2330 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2331 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2332 _FEED_NAME = 'recommended'
2333 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2336 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2337 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2338 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2339 _FEED_NAME = 'subscriptions'
2340 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2343 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2344 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2345 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2346 _FEED_NAME = 'history'
2347 _PLAYLIST_TITLE = 'Youtube History'
2350 class YoutubeTruncatedURLIE(InfoExtractor):
2351 IE_NAME = 'youtube:truncated_url'
2352 IE_DESC = False # Do not list
2353 _VALID_URL = r'''(?x)
2355 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2358 annotation_id=annotation_[^&]+|
2364 attribution_link\?a=[^&]+
2370 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2371 'only_matching': True,
2373 'url': 'http://www.youtube.com/watch?',
2374 'only_matching': True,
2376 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2377 'only_matching': True,
2379 'url': 'https://www.youtube.com/watch?feature=foo',
2380 'only_matching': True,
2382 'url': 'https://www.youtube.com/watch?hl=en-GB',
2383 'only_matching': True,
2385 'url': 'https://www.youtube.com/watch?t=2372',
2386 'only_matching': True,
2389 def _real_extract(self, url):
2390 raise ExtractorError(
2391 'Did you forget to quote the URL? Remember that & is a meta '
2392 'character in most shells, so you want to put the URL in quotes, '
2394 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2395 ' or simply youtube-dl BaW_jenozKc .',
2399 class YoutubeTruncatedIDIE(InfoExtractor):
2400 IE_NAME = 'youtube:truncated_id'
2401 IE_DESC = False # Do not list
2402 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2405 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2406 'only_matching': True,
2409 def _real_extract(self, url):
2410 video_id = self._match_id(url)
2411 raise ExtractorError(
2412 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),