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 (
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlparse,
33 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 _NETRC_MACHINE = 'youtube'
57 # If True it will raise an error if no login info is provided
58 _LOGIN_REQUIRED = False
60 def _set_language(self):
62 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
63 # YouTube sets the expire time to about two months
64 expire_time=time.time() + 2 * 30 * 24 * 3600)
66 def _ids_to_results(self, ids):
68 self.url_result(vid_id, 'Youtube', video_id=vid_id)
73 Attempt to log in to YouTube.
74 True is returned if successful or skipped.
75 False is returned if login failed.
77 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
79 (username, password) = self._get_login_info()
80 # No authentication to be performed
82 if self._LOGIN_REQUIRED:
83 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
86 login_page = self._download_webpage(
87 self._LOGIN_URL, None,
88 note='Downloading login page',
89 errnote='unable to fetch login page', fatal=False)
90 if login_page is False:
93 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
94 login_page, 'Login GALX parameter')
98 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
103 'PersistentCookie': 'yes',
105 'bgresponse': 'js_disabled',
106 'checkConnection': '',
107 'checkedDomains': 'youtube',
114 'service': 'youtube',
119 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
121 req = sanitized_Request(self._LOGIN_URL, login_data)
122 login_results = self._download_webpage(
124 note='Logging in', errnote='unable to log in', fatal=False)
125 if login_results is False:
128 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
129 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
132 # TODO add SMS and phone call support - these require making a request and then prompting the user
134 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
135 tfa_code = self._get_tfa_info('2-step verification code')
138 self._downloader.report_warning(
139 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
140 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
143 tfa_code = remove_start(tfa_code, 'G-')
145 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
147 tfa_form_strs.update({
152 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
154 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
155 tfa_results = self._download_webpage(
157 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
159 if tfa_results is False:
162 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
163 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
165 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
166 self._downloader.report_warning('unable to log in - did the page structure change?')
168 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
169 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
172 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
173 self._downloader.report_warning('unable to log in: bad username or password')
177 def _real_initialize(self):
178 if self._downloader is None:
181 if not self._login():
185 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
186 # Extract entries from page with "Load more" button
187 def _entries(self, page, playlist_id):
188 more_widget_html = content_html = page
189 for page_num in itertools.count(1):
190 for entry in self._process_page(content_html):
193 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
197 more = self._download_json(
198 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
199 'Downloading page #%s' % page_num,
200 transform_source=uppercase_escape)
201 content_html = more['content_html']
202 if not content_html.strip():
203 # Some webpages show a "Load more" button but they don't
206 more_widget_html = more['load_more_widget_html']
209 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
210 def _process_page(self, content):
211 for video_id, video_title in self.extract_videos_from_page(content):
212 yield self.url_result(video_id, 'Youtube', video_id, video_title)
214 def extract_videos_from_page(self, page):
217 for mobj in re.finditer(self._VIDEO_RE, page):
218 # The link with index 0 is not the first video of the playlist (not sure if still actual)
219 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
221 video_id = mobj.group('id')
222 video_title = unescapeHTML(mobj.group('title'))
224 video_title = video_title.strip()
226 idx = ids_in_page.index(video_id)
227 if video_title and not titles_in_page[idx]:
228 titles_in_page[idx] = video_title
230 ids_in_page.append(video_id)
231 titles_in_page.append(video_title)
232 return zip(ids_in_page, titles_in_page)
235 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
236 def _process_page(self, content):
237 for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
238 yield self.url_result(
239 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
241 def _real_extract(self, url):
242 playlist_id = self._match_id(url)
243 webpage = self._download_webpage(url, playlist_id)
244 title = self._og_search_title(webpage, fatal=False)
245 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
248 class YoutubeIE(YoutubeBaseInfoExtractor):
249 IE_DESC = 'YouTube.com'
250 _VALID_URL = r"""(?x)^
252 (?:https?://|//) # http(s):// or protocol-independent URL
253 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
254 (?:www\.)?deturl\.com/www\.youtube\.com/|
255 (?:www\.)?pwnyoutube\.com/|
256 (?:www\.)?yourepeat\.com/|
257 tube\.majestyc\.net/|
258 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
259 (?:.*?\#/)? # handle anchor (#/) redirect urls
260 (?: # the various things that can precede the ID:
261 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
262 |(?: # or the v= param in all its forms
263 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
264 (?:\?|\#!?) # the params delimiter ? or # or #!
265 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
270 youtu\.be| # just youtu.be/xxxx
271 vid\.plus # or vid.plus/xxxx
273 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
275 )? # all until now is optional -> you can pass the naked ID
276 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
277 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
278 (?(1).+)? # if we found the ID, everything can follow
280 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
282 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
283 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
284 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
285 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
286 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
287 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
288 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
290 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
291 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
292 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
293 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
294 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
295 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
296 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
297 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
298 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
303 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
304 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
305 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
306 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
307 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
308 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
309 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
311 # Apple HTTP Live Streaming
312 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
313 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
314 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
315 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
316 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
317 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
318 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
321 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
322 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
323 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
325 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
326 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
327 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
328 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
329 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
330 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
331 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
334 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
335 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
336 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
339 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
344 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
345 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
346 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
350 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
354 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
355 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
356 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
357 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
358 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
359 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
363 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
364 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
366 # Dash webm audio with opus inside
367 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
368 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
369 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
372 '_rtmp': {'protocol': 'rtmp'},
374 _SUBTITLE_FORMATS = ('ttml', 'vtt')
379 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
383 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
384 'uploader': 'Philipp Hagemeister',
385 'uploader_id': 'phihag',
386 'upload_date': '20121002',
387 'license': 'Standard YouTube License',
388 '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 .',
389 'categories': ['Science & Technology'],
390 'tags': ['youtube-dl'],
392 'dislike_count': int,
398 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
399 'note': 'Test generic use_cipher_signature video (#897)',
403 'upload_date': '20120506',
404 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
405 'alt_title': 'I Love It (feat. Charli XCX)',
406 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
407 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
408 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
409 'iconic ep', 'iconic', 'love', 'it'],
410 'uploader': 'Icona Pop',
411 'uploader_id': 'IconaPop',
412 'license': 'Standard YouTube License',
413 'creator': 'Icona Pop',
417 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
418 'note': 'Test VEVO video with age protection (#956)',
422 'upload_date': '20130703',
423 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
424 'alt_title': 'Tunnel Vision',
425 'description': 'md5:64249768eec3bc4276236606ea996373',
426 'uploader': 'justintimberlakeVEVO',
427 'uploader_id': 'justintimberlakeVEVO',
428 'license': 'Standard YouTube License',
429 'creator': 'Justin Timberlake',
434 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
435 'note': 'Embed-only video (#1746)',
439 'upload_date': '20120608',
440 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
441 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
442 'uploader': 'SET India',
443 'uploader_id': 'setindia',
444 'license': 'Standard YouTube License',
449 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
450 'note': 'Use the first video ID in the URL',
454 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
455 'uploader': 'Philipp Hagemeister',
456 'uploader_id': 'phihag',
457 'upload_date': '20121002',
458 'license': 'Standard YouTube License',
459 '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 .',
460 'categories': ['Science & Technology'],
461 'tags': ['youtube-dl'],
463 'dislike_count': int,
466 'skip_download': True,
470 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
471 'note': '256k DASH audio (format 141) via DASH manifest',
475 'upload_date': '20121002',
476 'uploader_id': '8KVIDEO',
478 'uploader': '8KVIDEO',
479 'license': 'Standard YouTube License',
480 'title': 'UHDTV TEST 8K VIDEO.mp4'
483 'youtube_include_dash_manifest': True,
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,
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,
527 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
531 'upload_date': '20100909',
532 'uploader': 'The Amazing Atheist',
533 'uploader_id': '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 'upload_date': '20140605',
550 'license': 'Standard YouTube License',
554 # Age-gate video with encrypted signature
556 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
560 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
561 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
562 'uploader': 'LloydVEVO',
563 'uploader_id': 'LloydVEVO',
564 'upload_date': '20110629',
565 'license': 'Standard YouTube License',
569 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
571 'url': '__2ABJjxzNo',
575 'upload_date': '20100430',
576 'uploader_id': 'deadmau5',
577 'creator': 'deadmau5',
578 'description': 'md5:12c56784b8032162bb936a5f76d55360',
579 'uploader': 'deadmau5',
580 'license': 'Standard YouTube License',
581 'title': 'Deadmau5 - Some Chords (HD)',
582 'alt_title': 'Some Chords',
584 'expected_warnings': [
585 'DASH manifest missing',
588 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
590 'url': 'lqQg6PlCWgI',
594 'upload_date': '20150827',
595 'uploader_id': 'olympic',
596 'license': 'Standard YouTube License',
597 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
598 'uploader': 'Olympics',
599 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
602 'skip_download': 'requires avconv',
607 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
611 'stretched_ratio': 16 / 9.,
612 'upload_date': '20110310',
613 'uploader_id': 'AllenMeow',
614 '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',
616 'license': 'Standard YouTube License',
617 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
620 # url_encoded_fmt_stream_map is empty string
622 'url': 'qEJwOuvDf7I',
626 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
628 'upload_date': '20150404',
629 'uploader_id': 'spbelect',
630 'uploader': 'Наблюдатели Петербурга',
633 'skip_download': 'requires avconv',
635 'skip': 'This live event has ended.',
637 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
639 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
643 'title': 'md5:7b81415841e02ecd4313668cde88737a',
644 'description': 'md5:116377fd2963b81ec4ce64b542173306',
645 'upload_date': '20150625',
646 'uploader_id': 'dorappi2000',
647 'uploader': 'dorappi2000',
648 'license': 'Standard YouTube License',
649 'formats': 'mincount:33',
652 # DASH manifest with segment_list
654 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
655 'md5': '8ce563a1d667b599d21064e982ab9e31',
659 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
660 'uploader': 'Airtek',
661 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
662 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
663 'license': 'Standard YouTube License',
664 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
667 'youtube_include_dash_manifest': True,
668 'format': '135', # bestvideo
672 # Multifeed videos (multiple cameras), URL is for Main Camera
673 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
676 'title': 'teamPGP: Rocket League Noob Stream',
677 'description': 'md5:dc7872fb300e143831327f1bae3af010',
683 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
684 'description': 'md5:dc7872fb300e143831327f1bae3af010',
685 'upload_date': '20150721',
686 'uploader': 'Beer Games Beer',
687 'uploader_id': 'beergamesbeer',
688 'license': 'Standard YouTube License',
694 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
695 'description': 'md5:dc7872fb300e143831327f1bae3af010',
696 'upload_date': '20150721',
697 'uploader': 'Beer Games Beer',
698 'uploader_id': 'beergamesbeer',
699 'license': 'Standard YouTube License',
705 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
706 'description': 'md5:dc7872fb300e143831327f1bae3af010',
707 'upload_date': '20150721',
708 'uploader': 'Beer Games Beer',
709 'uploader_id': 'beergamesbeer',
710 'license': 'Standard YouTube License',
716 'title': 'teamPGP: Rocket League Noob Stream (zim)',
717 'description': 'md5:dc7872fb300e143831327f1bae3af010',
718 'upload_date': '20150721',
719 'uploader': 'Beer Games Beer',
720 'uploader_id': 'beergamesbeer',
721 'license': 'Standard YouTube License',
725 'skip_download': True,
729 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
730 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
733 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
738 'url': 'http://vid.plus/FlRa-iH7PGw',
739 'only_matching': True,
742 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
743 # Also tests cut-off URL expansion in video description (see
744 # https://github.com/rg3/youtube-dl/issues/1892,
745 # https://github.com/rg3/youtube-dl/issues/8164)
746 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
750 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
751 'alt_title': 'Dark Walk',
752 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
753 'upload_date': '20151119',
754 'uploader_id': 'IronSoulElf',
755 'uploader': 'IronSoulElf',
756 'license': 'Standard YouTube License',
757 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
760 'skip_download': True,
764 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
765 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
766 'only_matching': True,
769 # Video with yt:stretch=17:0
770 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
774 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
775 'description': 'md5:ee18a25c350637c8faff806845bddee9',
776 'upload_date': '20151107',
777 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
778 'uploader': 'CH GAMER DROID',
781 'skip_download': True,
785 # Video licensed under Creative Commons
786 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
790 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
791 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
792 'upload_date': '20150127',
793 'uploader_id': 'BerkmanCenter',
794 'uploader': 'BerkmanCenter',
795 'license': 'Creative Commons Attribution license (reuse allowed)',
798 'skip_download': True,
802 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
803 'only_matching': True,
807 def __init__(self, *args, **kwargs):
808 super(YoutubeIE, self).__init__(*args, **kwargs)
809 self._player_cache = {}
811 def report_video_info_webpage_download(self, video_id):
812 """Report attempt to download video info webpage."""
813 self.to_screen('%s: Downloading video info webpage' % video_id)
815 def report_information_extraction(self, video_id):
816 """Report attempt to extract video information."""
817 self.to_screen('%s: Extracting video information' % video_id)
819 def report_unavailable_format(self, video_id, format):
820 """Report extracted video URL."""
821 self.to_screen('%s: Format %s not available' % (video_id, format))
823 def report_rtmp_download(self):
824 """Indicate the download will use the RTMP protocol."""
825 self.to_screen('RTMP download detected')
827 def _signature_cache_id(self, example_sig):
828 """ Return a string representation of a signature """
829 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
831 def _extract_signature_function(self, video_id, player_url, example_sig):
833 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
836 raise ExtractorError('Cannot identify player %r' % player_url)
837 player_type = id_m.group('ext')
838 player_id = id_m.group('id')
840 # Read from filesystem cache
841 func_id = '%s_%s_%s' % (
842 player_type, player_id, self._signature_cache_id(example_sig))
843 assert os.path.basename(func_id) == func_id
845 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
846 if cache_spec is not None:
847 return lambda s: ''.join(s[i] for i in cache_spec)
850 'Downloading player %s' % player_url
851 if self._downloader.params.get('verbose') else
852 'Downloading %s player %s' % (player_type, player_id)
854 if player_type == 'js':
855 code = self._download_webpage(
856 player_url, video_id,
858 errnote='Download of %s failed' % player_url)
859 res = self._parse_sig_js(code)
860 elif player_type == 'swf':
861 urlh = self._request_webpage(
862 player_url, video_id,
864 errnote='Download of %s failed' % player_url)
866 res = self._parse_sig_swf(code)
868 assert False, 'Invalid player type %r' % player_type
870 test_string = ''.join(map(compat_chr, range(len(example_sig))))
871 cache_res = res(test_string)
872 cache_spec = [ord(c) for c in cache_res]
874 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
877 def _print_sig_code(self, func, example_sig):
878 def gen_sig_code(idxs):
879 def _genslice(start, end, step):
880 starts = '' if start == 0 else str(start)
881 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
882 steps = '' if step == 1 else (':%d' % step)
883 return 's[%s%s%s]' % (starts, ends, steps)
886 # Quelch pyflakes warnings - start will be set when step is set
887 start = '(Never used)'
888 for i, prev in zip(idxs[1:], idxs[:-1]):
892 yield _genslice(start, prev, step)
895 if i - prev in [-1, 1]:
904 yield _genslice(start, i, step)
906 test_string = ''.join(map(compat_chr, range(len(example_sig))))
907 cache_res = func(test_string)
908 cache_spec = [ord(c) for c in cache_res]
909 expr_code = ' + '.join(gen_sig_code(cache_spec))
910 signature_id_tuple = '(%s)' % (
911 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
912 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
913 ' return %s\n') % (signature_id_tuple, expr_code)
914 self.to_screen('Extracted signature function:\n' + code)
916 def _parse_sig_js(self, jscode):
917 funcname = self._search_regex(
918 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
919 'Initial JS player signature function name')
921 jsi = JSInterpreter(jscode)
922 initial_function = jsi.extract_function(funcname)
923 return lambda s: initial_function([s])
925 def _parse_sig_swf(self, file_contents):
926 swfi = SWFInterpreter(file_contents)
927 TARGET_CLASSNAME = 'SignatureDecipher'
928 searched_class = swfi.extract_class(TARGET_CLASSNAME)
929 initial_function = swfi.extract_function(searched_class, 'decipher')
930 return lambda s: initial_function([s])
932 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
933 """Turn the encrypted s field into a working signature"""
935 if player_url is None:
936 raise ExtractorError('Cannot decrypt signature without player_url')
938 if player_url.startswith('//'):
939 player_url = 'https:' + player_url
941 player_id = (player_url, self._signature_cache_id(s))
942 if player_id not in self._player_cache:
943 func = self._extract_signature_function(
944 video_id, player_url, s
946 self._player_cache[player_id] = func
947 func = self._player_cache[player_id]
948 if self._downloader.params.get('youtube_print_sig_code'):
949 self._print_sig_code(func, s)
951 except Exception as e:
952 tb = traceback.format_exc()
953 raise ExtractorError(
954 'Signature extraction failed: ' + tb, cause=e)
956 def _get_subtitles(self, video_id, webpage):
958 subs_doc = self._download_xml(
959 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
960 video_id, note=False)
961 except ExtractorError as err:
962 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
966 for track in subs_doc.findall('track'):
967 lang = track.attrib['lang_code']
968 if lang in sub_lang_list:
971 for ext in self._SUBTITLE_FORMATS:
972 params = compat_urllib_parse.urlencode({
976 'name': track.attrib['name'].encode('utf-8'),
979 'url': 'https://www.youtube.com/api/timedtext?' + params,
982 sub_lang_list[lang] = sub_formats
983 if not sub_lang_list:
984 self._downloader.report_warning('video doesn\'t have subtitles')
988 def _get_ytplayer_config(self, video_id, webpage):
990 # User data may contain arbitrary character sequences that may affect
991 # JSON extraction with regex, e.g. when '};' is contained the second
992 # regex won't capture the whole JSON. Yet working around by trying more
993 # concrete regex first keeping in mind proper quoted string handling
994 # to be implemented in future that will replace this workaround (see
995 # https://github.com/rg3/youtube-dl/issues/7468,
996 # https://github.com/rg3/youtube-dl/pull/7599)
997 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
998 r';ytplayer\.config\s*=\s*({.+?});',
1000 config = self._search_regex(
1001 patterns, webpage, 'ytplayer.config', default=None)
1003 return self._parse_json(
1004 uppercase_escape(config), video_id, fatal=False)
1006 def _get_automatic_captions(self, video_id, webpage):
1007 """We need the webpage for getting the captions url, pass it as an
1008 argument to speed up the process."""
1009 self.to_screen('%s: Looking for automatic captions' % video_id)
1010 player_config = self._get_ytplayer_config(video_id, webpage)
1011 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1012 if not player_config:
1013 self._downloader.report_warning(err_msg)
1016 args = player_config['args']
1017 caption_url = args.get('ttsurl')
1019 timestamp = args['timestamp']
1020 # We get the available subtitles
1021 list_params = compat_urllib_parse.urlencode({
1026 list_url = caption_url + '&' + list_params
1027 caption_list = self._download_xml(list_url, video_id)
1028 original_lang_node = caption_list.find('track')
1029 if original_lang_node is None:
1030 self._downloader.report_warning('Video doesn\'t have automatic captions')
1032 original_lang = original_lang_node.attrib['lang_code']
1033 caption_kind = original_lang_node.attrib.get('kind', '')
1036 for lang_node in caption_list.findall('target'):
1037 sub_lang = lang_node.attrib['lang_code']
1039 for ext in self._SUBTITLE_FORMATS:
1040 params = compat_urllib_parse.urlencode({
1041 'lang': original_lang,
1045 'kind': caption_kind,
1047 sub_formats.append({
1048 'url': caption_url + '&' + params,
1051 sub_lang_list[sub_lang] = sub_formats
1052 return sub_lang_list
1054 # Some videos don't provide ttsurl but rather caption_tracks and
1055 # caption_translation_languages (e.g. 20LmZk1hakA)
1056 caption_tracks = args['caption_tracks']
1057 caption_translation_languages = args['caption_translation_languages']
1058 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1059 parsed_caption_url = compat_urlparse.urlparse(caption_url)
1060 caption_qs = compat_parse_qs(parsed_caption_url.query)
1063 for lang in caption_translation_languages.split(','):
1064 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1065 sub_lang = lang_qs.get('lc', [None])[0]
1069 for ext in self._SUBTITLE_FORMATS:
1071 'tlang': [sub_lang],
1074 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1075 query=compat_urllib_parse.urlencode(caption_qs, True)))
1076 sub_formats.append({
1080 sub_lang_list[sub_lang] = sub_formats
1081 return sub_lang_list
1082 # An extractor error can be raise by the download process if there are
1083 # no automatic captions but there are subtitles
1084 except (KeyError, ExtractorError):
1085 self._downloader.report_warning(err_msg)
1088 def _mark_watched(self, video_id, video_info):
1089 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1090 if not playback_url:
1092 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1093 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1095 # cpn generation algorithm is reverse engineered from base.js.
1096 # In fact it works even with dummy cpn.
1097 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1098 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1104 playback_url = compat_urlparse.urlunparse(
1105 parsed_playback_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
1107 self._download_webpage(
1108 playback_url, video_id, 'Marking watched',
1109 'Unable to mark watched', fatal=False)
1112 def extract_id(cls, url):
1113 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1115 raise ExtractorError('Invalid URL: %s' % url)
1116 video_id = mobj.group(2)
1119 def _extract_from_m3u8(self, manifest_url, video_id):
1122 def _get_urls(_manifest):
1123 lines = _manifest.split('\n')
1124 urls = filter(lambda l: l and not l.startswith('#'),
1127 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1128 formats_urls = _get_urls(manifest)
1129 for format_url in formats_urls:
1130 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1131 url_map[itag] = format_url
1134 def _extract_annotations(self, video_id):
1135 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1136 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1138 def _real_extract(self, url):
1139 url, smuggled_data = unsmuggle_url(url, {})
1142 'http' if self._downloader.params.get('prefer_insecure', False)
1147 parsed_url = compat_urllib_parse_urlparse(url)
1148 for component in [parsed_url.fragment, parsed_url.query]:
1149 query = compat_parse_qs(component)
1150 if start_time is None and 't' in query:
1151 start_time = parse_duration(query['t'][0])
1152 if start_time is None and 'start' in query:
1153 start_time = parse_duration(query['start'][0])
1154 if end_time is None and 'end' in query:
1155 end_time = parse_duration(query['end'][0])
1157 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1158 mobj = re.search(self._NEXT_URL_RE, url)
1160 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1161 video_id = self.extract_id(url)
1164 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1165 video_webpage = self._download_webpage(url, video_id)
1167 # Attempt to extract SWF player URL
1168 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1169 if mobj is not None:
1170 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1176 def add_dash_mpd(video_info):
1177 dash_mpd = video_info.get('dashmpd')
1178 if dash_mpd and dash_mpd[0] not in dash_mpds:
1179 dash_mpds.append(dash_mpd[0])
1182 embed_webpage = None
1184 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1186 # We simulate the access to the video from www.youtube.com/v/{video_id}
1187 # this can be viewed without login into Youtube
1188 url = proto + '://www.youtube.com/embed/%s' % video_id
1189 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1190 data = compat_urllib_parse.urlencode({
1191 'video_id': video_id,
1192 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1193 'sts': self._search_regex(
1194 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1196 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1197 video_info_webpage = self._download_webpage(
1198 video_info_url, video_id,
1199 note='Refetching age-gated info webpage',
1200 errnote='unable to download video info webpage')
1201 video_info = compat_parse_qs(video_info_webpage)
1202 add_dash_mpd(video_info)
1206 # Try looking directly into the video webpage
1207 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1209 args = ytplayer_config['args']
1210 if args.get('url_encoded_fmt_stream_map'):
1211 # Convert to the same format returned by compat_parse_qs
1212 video_info = dict((k, [v]) for k, v in args.items())
1213 add_dash_mpd(video_info)
1214 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1216 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1217 # We also try looking in get_video_info since it may contain different dashmpd
1218 # URL that points to a DASH manifest with possibly different itag set (some itags
1219 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1220 # manifest pointed by get_video_info's dashmpd).
1221 # The general idea is to take a union of itags of both DASH manifests (for example
1222 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1223 self.report_video_info_webpage_download(video_id)
1224 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1226 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1227 % (proto, video_id, el_type))
1228 video_info_webpage = self._download_webpage(
1230 video_id, note=False,
1231 errnote='unable to download video info webpage')
1232 get_video_info = compat_parse_qs(video_info_webpage)
1233 if get_video_info.get('use_cipher_signature') != ['True']:
1234 add_dash_mpd(get_video_info)
1236 video_info = get_video_info
1237 if 'token' in get_video_info:
1238 # Different get_video_info requests may report different results, e.g.
1239 # some may report video unavailability, but some may serve it without
1240 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1241 # the original webpage as well as el=info and el=embedded get_video_info
1242 # requests report video unavailability due to geo restriction while
1243 # el=detailpage succeeds and returns valid data). This is probably
1244 # due to YouTube measures against IP ranges of hosting providers.
1245 # Working around by preferring the first succeeded video_info containing
1246 # the token if no such video_info yet was found.
1247 if 'token' not in video_info:
1248 video_info = get_video_info
1250 if 'token' not in video_info:
1251 if 'reason' in video_info:
1252 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1253 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1255 raise ExtractorError('YouTube said: This video is available in %s only' % (
1256 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1258 raise ExtractorError(
1259 'YouTube said: %s' % video_info['reason'][0],
1260 expected=True, video_id=video_id)
1262 raise ExtractorError(
1263 '"token" parameter not in video info for unknown reason',
1267 if 'title' in video_info:
1268 video_title = video_info['title'][0]
1270 self._downloader.report_warning('Unable to extract video title')
1274 video_description = get_element_by_id("eow-description", video_webpage)
1275 if video_description:
1276 video_description = re.sub(r'''(?x)
1278 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1279 (?:title|href)="([^"]+)"\s+
1280 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1281 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1284 ''', r'\1', video_description)
1285 video_description = clean_html(video_description)
1287 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1289 video_description = unescapeHTML(fd_mobj.group(1))
1291 video_description = ''
1293 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1294 if not self._downloader.params.get('noplaylist'):
1297 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1298 for feed in multifeed_metadata_list.split(','):
1299 # Unquote should take place before split on comma (,) since textual
1300 # fields may contain comma as well (see
1301 # https://github.com/rg3/youtube-dl/issues/8536)
1302 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1304 '_type': 'url_transparent',
1305 'ie_key': 'Youtube',
1307 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1308 {'force_singlefeed': True}),
1309 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1311 feed_ids.append(feed_data['id'][0])
1313 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1314 % (', '.join(feed_ids), video_id))
1315 return self.playlist_result(entries, video_id, video_title, video_description)
1316 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1318 if 'view_count' in video_info:
1319 view_count = int(video_info['view_count'][0])
1323 # Check for "rental" videos
1324 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1325 raise ExtractorError('"rental" videos not supported')
1327 # Start extracting information
1328 self.report_information_extraction(video_id)
1331 if 'author' not in video_info:
1332 raise ExtractorError('Unable to extract uploader name')
1333 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1336 video_uploader_id = None
1337 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1338 if mobj is not None:
1339 video_uploader_id = mobj.group(1)
1341 self._downloader.report_warning('unable to extract uploader nickname')
1344 # We try first to get a high quality image:
1345 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1346 video_webpage, re.DOTALL)
1347 if m_thumb is not None:
1348 video_thumbnail = m_thumb.group(1)
1349 elif 'thumbnail_url' not in video_info:
1350 self._downloader.report_warning('unable to extract video thumbnail')
1351 video_thumbnail = None
1352 else: # don't panic if we can't find it
1353 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1356 upload_date = self._html_search_meta(
1357 'datePublished', video_webpage, 'upload date', default=None)
1359 upload_date = self._search_regex(
1360 [r'(?s)id="eow-date.*?>(.*?)</span>',
1361 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1362 video_webpage, 'upload date', default=None)
1364 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1365 upload_date = unified_strdate(upload_date)
1367 video_license = self._html_search_regex(
1368 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1369 video_webpage, 'license', default=None)
1371 m_music = re.search(
1372 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1375 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1376 video_creator = clean_html(m_music.group('creator'))
1378 video_alt_title = video_creator = None
1380 m_cat_container = self._search_regex(
1381 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1382 video_webpage, 'categories', default=None)
1384 category = self._html_search_regex(
1385 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1387 video_categories = None if category is None else [category]
1389 video_categories = None
1392 unescapeHTML(m.group('content'))
1393 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1395 def _extract_count(count_name):
1396 return str_to_int(self._search_regex(
1397 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1398 % re.escape(count_name),
1399 video_webpage, count_name, default=None))
1401 like_count = _extract_count('like')
1402 dislike_count = _extract_count('dislike')
1405 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1406 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1408 if 'length_seconds' not in video_info:
1409 self._downloader.report_warning('unable to extract video duration')
1410 video_duration = None
1412 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1415 video_annotations = None
1416 if self._downloader.params.get('writeannotations', False):
1417 video_annotations = self._extract_annotations(video_id)
1419 def _map_to_format_list(urlmap):
1421 for itag, video_real_url in urlmap.items():
1424 'url': video_real_url,
1425 'player_url': player_url,
1427 if itag in self._formats:
1428 dct.update(self._formats[itag])
1432 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1433 self.report_rtmp_download()
1435 'format_id': '_rtmp',
1437 'url': video_info['conn'][0],
1438 'player_url': player_url,
1440 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1441 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1442 if 'rtmpe%3Dyes' in encoded_url_map:
1443 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1445 fmt_list = video_info.get('fmt_list', [''])[0]
1447 for fmt in fmt_list.split(','):
1448 spec = fmt.split('/')
1450 width_height = spec[1].split('x')
1451 if len(width_height) == 2:
1452 formats_spec[spec[0]] = {
1453 'resolution': spec[1],
1454 'width': int_or_none(width_height[0]),
1455 'height': int_or_none(width_height[1]),
1458 for url_data_str in encoded_url_map.split(','):
1459 url_data = compat_parse_qs(url_data_str)
1460 if 'itag' not in url_data or 'url' not in url_data:
1462 format_id = url_data['itag'][0]
1463 url = url_data['url'][0]
1465 if 'sig' in url_data:
1466 url += '&signature=' + url_data['sig'][0]
1467 elif 's' in url_data:
1468 encrypted_sig = url_data['s'][0]
1469 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1471 jsplayer_url_json = self._search_regex(
1473 embed_webpage if age_gate else video_webpage,
1474 'JS player URL (1)', default=None)
1475 if not jsplayer_url_json and not age_gate:
1476 # We need the embed website after all
1477 if embed_webpage is None:
1478 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1479 embed_webpage = self._download_webpage(
1480 embed_url, video_id, 'Downloading embed webpage')
1481 jsplayer_url_json = self._search_regex(
1482 ASSETS_RE, embed_webpage, 'JS player URL')
1484 player_url = json.loads(jsplayer_url_json)
1485 if player_url is None:
1486 player_url_json = self._search_regex(
1487 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1488 video_webpage, 'age gate player URL')
1489 player_url = json.loads(player_url_json)
1491 if self._downloader.params.get('verbose'):
1492 if player_url is None:
1493 player_version = 'unknown'
1494 player_desc = 'unknown'
1496 if player_url.endswith('swf'):
1497 player_version = self._search_regex(
1498 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1499 'flash player', fatal=False)
1500 player_desc = 'flash player %s' % player_version
1502 player_version = self._search_regex(
1503 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1505 'html5 player', fatal=False)
1506 player_desc = 'html5 player %s' % player_version
1508 parts_sizes = self._signature_cache_id(encrypted_sig)
1509 self.to_screen('{%s} signature length %s, %s' %
1510 (format_id, parts_sizes, player_desc))
1512 signature = self._decrypt_signature(
1513 encrypted_sig, video_id, player_url, age_gate)
1514 url += '&signature=' + signature
1515 if 'ratebypass' not in url:
1516 url += '&ratebypass=yes'
1519 'format_id': format_id,
1521 'player_url': player_url,
1523 if format_id in self._formats:
1524 dct.update(self._formats[format_id])
1525 if format_id in formats_spec:
1526 dct.update(formats_spec[format_id])
1528 # Some itags are not included in DASH manifest thus corresponding formats will
1529 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1530 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1531 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1532 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1535 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1536 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1539 'fps': int_or_none(url_data.get('fps', [None])[0]),
1540 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1542 for key, value in more_fields.items():
1545 type_ = url_data.get('type', [None])[0]
1547 type_split = type_.split(';')
1548 kind_ext = type_split[0].split('/')
1549 if len(kind_ext) == 2:
1551 dct['ext'] = mimetype2ext(type_split[0])
1552 if kind in ('audio', 'video'):
1554 for mobj in re.finditer(
1555 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1556 if mobj.group('key') == 'codecs':
1557 codecs = mobj.group('val')
1560 codecs = codecs.split(',')
1561 if len(codecs) == 2:
1562 acodec, vcodec = codecs[1], codecs[0]
1564 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1570 elif video_info.get('hlsvp'):
1571 manifest_url = video_info['hlsvp'][0]
1572 url_map = self._extract_from_m3u8(manifest_url, video_id)
1573 formats = _map_to_format_list(url_map)
1574 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1575 for a_format in formats:
1576 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1578 unavailable_message = self._html_search_regex(
1579 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1580 video_webpage, 'unavailable message', default=None)
1581 if unavailable_message:
1582 raise ExtractorError(unavailable_message, expected=True)
1583 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1585 # Look for the DASH manifest
1586 if self._downloader.params.get('youtube_include_dash_manifest', True):
1587 dash_mpd_fatal = True
1588 for mpd_url in dash_mpds:
1591 def decrypt_sig(mobj):
1593 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1594 return '/signature/%s' % dec_s
1596 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1598 for df in self._extract_mpd_formats(
1599 mpd_url, video_id, fatal=dash_mpd_fatal,
1600 formats_dict=self._formats):
1601 # Do not overwrite DASH format found in some previous DASH manifest
1602 if df['format_id'] not in dash_formats:
1603 dash_formats[df['format_id']] = df
1604 # Additional DASH manifests may end up in HTTP Error 403 therefore
1605 # allow them to fail without bug report message if we already have
1606 # some DASH manifest succeeded. This is temporary workaround to reduce
1607 # burst of bug reports until we figure out the reason and whether it
1608 # can be fixed at all.
1609 dash_mpd_fatal = False
1610 except (ExtractorError, KeyError) as e:
1611 self.report_warning(
1612 'Skipping DASH manifest: %r' % e, video_id)
1614 # Remove the formats we found through non-DASH, they
1615 # contain less info and it can be wrong, because we use
1616 # fixed values (for example the resolution). See
1617 # https://github.com/rg3/youtube-dl/issues/5774 for an
1619 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1620 formats.extend(dash_formats.values())
1622 # Check for malformed aspect ratio
1623 stretched_m = re.search(
1624 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1627 w = float(stretched_m.group('w'))
1628 h = float(stretched_m.group('h'))
1629 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1630 # We will only process correct ratios.
1634 if f.get('vcodec') != 'none':
1635 f['stretched_ratio'] = ratio
1637 self._sort_formats(formats)
1639 self.mark_watched(video_id, video_info)
1643 'uploader': video_uploader,
1644 'uploader_id': video_uploader_id,
1645 'upload_date': upload_date,
1646 'license': video_license,
1647 'creator': video_creator,
1648 'title': video_title,
1649 'alt_title': video_alt_title,
1650 'thumbnail': video_thumbnail,
1651 'description': video_description,
1652 'categories': video_categories,
1654 'subtitles': video_subtitles,
1655 'automatic_captions': automatic_captions,
1656 'duration': video_duration,
1657 'age_limit': 18 if age_gate else 0,
1658 'annotations': video_annotations,
1659 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1660 'view_count': view_count,
1661 'like_count': like_count,
1662 'dislike_count': dislike_count,
1663 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1666 'start_time': start_time,
1667 'end_time': end_time,
1671 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1672 IE_DESC = 'YouTube.com playlists'
1673 _VALID_URL = r"""(?x)(?:
1678 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1679 \? (?:.*?[&;])*? (?:p|a|list)=
1683 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1684 # Top tracks, they can also include dots
1689 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1691 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1692 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1693 IE_NAME = 'youtube:playlist'
1695 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1697 'title': 'ytdl test PL',
1698 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1700 'playlist_count': 3,
1702 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1704 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1705 'title': 'YDL_Empty_List',
1707 'playlist_count': 0,
1709 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1710 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1712 'title': '29C3: Not my department',
1713 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1715 'playlist_count': 95,
1717 'note': 'issue #673',
1718 'url': 'PLBB231211A4F62143',
1720 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1721 'id': 'PLBB231211A4F62143',
1723 'playlist_mincount': 26,
1725 'note': 'Large playlist',
1726 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1728 'title': 'Uploads from Cauchemar',
1729 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1731 'playlist_mincount': 799,
1733 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1735 'title': 'YDL_safe_search',
1736 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1738 'playlist_count': 2,
1741 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1742 'playlist_count': 4,
1745 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1748 'note': 'Embedded SWF player',
1749 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1750 'playlist_count': 4,
1753 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1756 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1757 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1759 'title': 'Uploads from Interstellar Movie',
1760 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1762 'playlist_mincout': 21,
1765 def _real_initialize(self):
1768 def _extract_mix(self, playlist_id):
1769 # The mixes are generated from a single video
1770 # the id of the playlist is just 'RD' + video_id
1771 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1772 webpage = self._download_webpage(
1773 url, playlist_id, 'Downloading Youtube mix')
1774 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1776 search_title('playlist-title') or
1777 search_title('title long-title') or
1778 search_title('title'))
1779 title = clean_html(title_span)
1780 ids = orderedSet(re.findall(
1781 r'''(?xs)data-video-username=".*?".*?
1782 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1784 url_results = self._ids_to_results(ids)
1786 return self.playlist_result(url_results, playlist_id, title)
1788 def _extract_playlist(self, playlist_id):
1789 url = self._TEMPLATE_URL % playlist_id
1790 page = self._download_webpage(url, playlist_id)
1792 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1793 match = match.strip()
1794 # Check if the playlist exists or is private
1795 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1796 raise ExtractorError(
1797 'The playlist doesn\'t exist or is private, use --username or '
1798 '--netrc to access it.',
1800 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1801 raise ExtractorError(
1802 'Invalid parameters. Maybe URL is incorrect.',
1804 elif re.match(r'[^<]*Choose your language[^<]*', match):
1807 self.report_warning('Youtube gives an alert message: ' + match)
1809 playlist_title = self._html_search_regex(
1810 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1813 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1815 def _check_download_just_video(self, url, playlist_id):
1816 # Check if it's a video-specific URL
1817 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1818 if 'v' in query_dict:
1819 video_id = query_dict['v'][0]
1820 if self._downloader.params.get('noplaylist'):
1821 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1822 return self.url_result(video_id, 'Youtube', video_id=video_id)
1824 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1826 def _real_extract(self, url):
1827 # Extract playlist id
1828 mobj = re.match(self._VALID_URL, url)
1830 raise ExtractorError('Invalid URL: %s' % url)
1831 playlist_id = mobj.group(1) or mobj.group(2)
1833 video = self._check_download_just_video(url, playlist_id)
1837 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1838 # Mixes require a custom extraction process
1839 return self._extract_mix(playlist_id)
1841 return self._extract_playlist(playlist_id)
1844 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1845 IE_DESC = 'YouTube.com channels'
1846 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1847 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1848 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1849 IE_NAME = 'youtube:channel'
1851 'note': 'paginated channel',
1852 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1853 'playlist_mincount': 91,
1855 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1856 'title': 'Uploads from lex will',
1859 'note': 'Age restricted channel',
1860 # from https://www.youtube.com/user/DeusExOfficial
1861 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1862 'playlist_mincount': 64,
1864 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1865 'title': 'Uploads from Deus Ex',
1870 def suitable(cls, url):
1871 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1873 def _real_extract(self, url):
1874 channel_id = self._match_id(url)
1876 url = self._TEMPLATE_URL % channel_id
1878 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1879 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1880 # otherwise fallback on channel by page extraction
1881 channel_page = self._download_webpage(
1882 url + '?view=57', channel_id,
1883 'Downloading channel page', fatal=False)
1884 if channel_page is False:
1885 channel_playlist_id = False
1887 channel_playlist_id = self._html_search_meta(
1888 'channelId', channel_page, 'channel id', default=None)
1889 if not channel_playlist_id:
1890 channel_playlist_id = self._search_regex(
1891 r'data-(?:channel-external-|yt)id="([^"]+)"',
1892 channel_page, 'channel id', default=None)
1893 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1894 playlist_id = 'UU' + channel_playlist_id[2:]
1895 return self.url_result(
1896 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1898 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1899 autogenerated = re.search(r'''(?x)
1901 channel-header-autogenerated-label|
1902 yt-channel-title-autogenerated
1903 )[^"]*"''', channel_page) is not None
1906 # The videos are contained in a single page
1907 # the ajax pages can't be used, they are empty
1910 video_id, 'Youtube', video_id=video_id,
1911 video_title=video_title)
1912 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1913 return self.playlist_result(entries, channel_id)
1915 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1918 class YoutubeUserIE(YoutubeChannelIE):
1919 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1920 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1921 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1922 IE_NAME = 'youtube:user'
1925 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1926 'playlist_mincount': 320,
1928 'title': 'TheLinuxFoundation',
1931 'url': 'ytuser:phihag',
1932 'only_matching': True,
1936 def suitable(cls, url):
1937 # Don't return True if the url can be extracted with other youtube
1938 # extractor, the regex would is too permissive and it would match.
1939 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1940 if any(ie.suitable(url) for ie in other_ies):
1943 return super(YoutubeUserIE, cls).suitable(url)
1946 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1947 IE_DESC = 'YouTube.com user/channel playlists'
1948 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1949 IE_NAME = 'youtube:playlists'
1952 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1953 'playlist_mincount': 4,
1955 'id': 'ThirstForScience',
1956 'title': 'Thirst for Science',
1959 # with "Load more" button
1960 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1961 'playlist_mincount': 70,
1964 'title': 'Игорь Клейнер',
1967 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1968 'playlist_mincount': 17,
1970 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1971 'title': 'Chem Player',
1976 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1977 IE_DESC = 'YouTube.com searches'
1978 # there doesn't appear to be a real limit, for example if you search for
1979 # 'python' you get more than 8.000.000 results
1980 _MAX_RESULTS = float('inf')
1981 IE_NAME = 'youtube:search'
1982 _SEARCH_KEY = 'ytsearch'
1983 _EXTRA_QUERY_ARGS = {}
1986 def _get_n_results(self, query, n):
1987 """Get a specified number of results for a query"""
1992 for pagenum in itertools.count(1):
1994 'search_query': query.encode('utf-8'),
1998 url_query.update(self._EXTRA_QUERY_ARGS)
1999 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
2000 data = self._download_json(
2001 result_url, video_id='query "%s"' % query,
2002 note='Downloading page %s' % pagenum,
2003 errnote='Unable to download API page')
2004 html_content = data[1]['body']['content']
2006 if 'class="search-message' in html_content:
2007 raise ExtractorError(
2008 '[youtube] No video results', expected=True)
2010 new_videos = self._ids_to_results(orderedSet(re.findall(
2011 r'href="/watch\?v=(.{11})', html_content)))
2012 videos += new_videos
2013 if not new_videos or len(videos) > limit:
2018 return self.playlist_result(videos, query)
2021 class YoutubeSearchDateIE(YoutubeSearchIE):
2022 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2023 _SEARCH_KEY = 'ytsearchdate'
2024 IE_DESC = 'YouTube.com searches, newest videos first'
2025 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2028 class YoutubeSearchURLIE(InfoExtractor):
2029 IE_DESC = 'YouTube.com search URLs'
2030 IE_NAME = 'youtube:search_url'
2031 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2033 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2034 'playlist_mincount': 5,
2036 'title': 'youtube-dl test video',
2039 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2040 'only_matching': True,
2043 def _real_extract(self, url):
2044 mobj = re.match(self._VALID_URL, url)
2045 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2047 webpage = self._download_webpage(url, query)
2048 result_code = self._search_regex(
2049 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2051 part_codes = re.findall(
2052 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2054 for part_code in part_codes:
2055 part_title = self._html_search_regex(
2056 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2057 part_url_snippet = self._html_search_regex(
2058 r'(?s)href="([^"]+)"', part_code, 'item URL')
2059 part_url = compat_urlparse.urljoin(
2060 'https://www.youtube.com/', part_url_snippet)
2064 'title': part_title,
2068 '_type': 'playlist',
2074 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2075 IE_DESC = 'YouTube.com (multi-season) shows'
2076 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2077 IE_NAME = 'youtube:show'
2079 'url': 'https://www.youtube.com/show/airdisasters',
2080 'playlist_mincount': 5,
2082 'id': 'airdisasters',
2083 'title': 'Air Disasters',
2087 def _real_extract(self, url):
2088 playlist_id = self._match_id(url)
2089 return super(YoutubeShowIE, self)._real_extract(
2090 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2093 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2095 Base class for feed extractors
2096 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2098 _LOGIN_REQUIRED = True
2102 return 'youtube:%s' % self._FEED_NAME
2104 def _real_initialize(self):
2107 def _real_extract(self, url):
2108 page = self._download_webpage(
2109 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2111 # The extraction process is the same as for playlists, but the regex
2112 # for the video ids doesn't contain an index
2114 more_widget_html = content_html = page
2115 for page_num in itertools.count(1):
2116 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2118 # 'recommended' feed has infinite 'load more' and each new portion spins
2119 # the same videos in (sometimes) slightly different order, so we'll check
2120 # for unicity and break when portion has no new videos
2121 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2127 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2131 more = self._download_json(
2132 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2133 'Downloading page #%s' % page_num,
2134 transform_source=uppercase_escape)
2135 content_html = more['content_html']
2136 more_widget_html = more['load_more_widget_html']
2138 return self.playlist_result(
2139 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2142 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2143 IE_NAME = 'youtube:watchlater'
2144 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2145 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2148 'url': 'https://www.youtube.com/playlist?list=WL',
2149 'only_matching': True,
2151 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2152 'only_matching': True,
2155 def _real_extract(self, url):
2156 video = self._check_download_just_video(url, 'WL')
2159 return self._extract_playlist('WL')
2162 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2163 IE_NAME = 'youtube:favorites'
2164 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2165 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2166 _LOGIN_REQUIRED = True
2168 def _real_extract(self, url):
2169 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2170 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2171 return self.url_result(playlist_id, 'YoutubePlaylist')
2174 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2175 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2176 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2177 _FEED_NAME = 'recommended'
2178 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2181 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2182 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2183 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2184 _FEED_NAME = 'subscriptions'
2185 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2188 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2189 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2190 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2191 _FEED_NAME = 'history'
2192 _PLAYLIST_TITLE = 'Youtube History'
2195 class YoutubeTruncatedURLIE(InfoExtractor):
2196 IE_NAME = 'youtube:truncated_url'
2197 IE_DESC = False # Do not list
2198 _VALID_URL = r'''(?x)
2200 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2203 annotation_id=annotation_[^&]+|
2209 attribution_link\?a=[^&]+
2215 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2216 'only_matching': True,
2218 'url': 'http://www.youtube.com/watch?',
2219 'only_matching': True,
2221 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2222 'only_matching': True,
2224 'url': 'https://www.youtube.com/watch?feature=foo',
2225 'only_matching': True,
2227 'url': 'https://www.youtube.com/watch?hl=en-GB',
2228 'only_matching': True,
2230 'url': 'https://www.youtube.com/watch?t=2372',
2231 'only_matching': True,
2234 def _real_extract(self, url):
2235 raise ExtractorError(
2236 'Did you forget to quote the URL? Remember that & is a meta '
2237 'character in most shells, so you want to put the URL in quotes, '
2239 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2240 ' or simply youtube-dl BaW_jenozKc .',
2244 class YoutubeTruncatedIDIE(InfoExtractor):
2245 IE_NAME = 'youtube:truncated_id'
2246 IE_DESC = False # Do not list
2247 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2250 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2251 'only_matching': True,
2254 def _real_extract(self, url):
2255 video_id = self._match_id(url)
2256 raise ExtractorError(
2257 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),