3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
23 compat_urllib_request,
32 get_element_by_attribute,
48 class YoutubeBaseInfoExtractor(InfoExtractor):
49 """Provide base functions for Youtube extractors"""
50 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
51 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
52 _NETRC_MACHINE = 'youtube'
53 # If True it will raise an error if no login info is provided
54 _LOGIN_REQUIRED = False
56 def _set_language(self):
58 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
59 # YouTube sets the expire time to about two months
60 expire_time=time.time() + 2 * 30 * 24 * 3600)
62 def _ids_to_results(self, ids):
64 self.url_result(vid_id, 'Youtube', video_id=vid_id)
69 Attempt to log in to YouTube.
70 True is returned if successful or skipped.
71 False is returned if login failed.
73 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
75 (username, password) = self._get_login_info()
76 # No authentication to be performed
78 if self._LOGIN_REQUIRED:
79 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
82 login_page = self._download_webpage(
83 self._LOGIN_URL, None,
84 note='Downloading login page',
85 errnote='unable to fetch login page', fatal=False)
86 if login_page is False:
89 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
90 login_page, 'Login GALX parameter')
94 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
99 'PersistentCookie': 'yes',
101 'bgresponse': 'js_disabled',
102 'checkConnection': '',
103 'checkedDomains': 'youtube',
110 'service': 'youtube',
115 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
117 req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
118 login_results = self._download_webpage(
120 note='Logging in', errnote='unable to log in', fatal=False)
121 if login_results is False:
124 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
125 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
128 # TODO add SMS and phone call support - these require making a request and then prompting the user
130 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
131 tfa_code = self._get_tfa_info('2-step verification code')
134 self._downloader.report_warning(
135 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
136 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
139 tfa_code = remove_start(tfa_code, 'G-')
141 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
143 tfa_form_strs.update({
148 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
150 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
151 tfa_results = self._download_webpage(
153 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
155 if tfa_results is False:
158 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
159 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
161 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
162 self._downloader.report_warning('unable to log in - did the page structure change?')
164 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
165 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
168 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
169 self._downloader.report_warning('unable to log in: bad username or password')
173 def _real_initialize(self):
174 if self._downloader is None:
177 if not self._login():
181 class YoutubePlaylistBaseInfoExtractor(InfoExtractor):
182 # Extract the video ids from the playlist pages
183 def _entries(self, page, playlist_id):
184 more_widget_html = content_html = page
185 for page_num in itertools.count(1):
186 for video_id, video_title in self.extract_videos_from_page(content_html):
187 yield self.url_result(
188 video_id, 'Youtube', video_id=video_id,
189 video_title=video_title)
191 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
195 more = self._download_json(
196 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
197 'Downloading page #%s' % page_num,
198 transform_source=uppercase_escape)
199 content_html = more['content_html']
200 if not content_html.strip():
201 # Some webpages show a "Load more" button but they don't
204 more_widget_html = more['load_more_widget_html']
206 def extract_videos_from_page(self, page):
209 for mobj in re.finditer(self._VIDEO_RE, page):
210 # The link with index 0 is not the first video of the playlist (not sure if still actual)
211 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
213 video_id = mobj.group('id')
214 video_title = unescapeHTML(mobj.group('title'))
216 video_title = video_title.strip()
218 idx = ids_in_page.index(video_id)
219 if video_title and not titles_in_page[idx]:
220 titles_in_page[idx] = video_title
222 ids_in_page.append(video_id)
223 titles_in_page.append(video_title)
224 return zip(ids_in_page, titles_in_page)
227 class YoutubePlaylistsBaseInfoExtractor(InfoExtractor):
228 def _real_extract(self, url):
229 playlist_id = self._match_id(url)
230 webpage = self._download_webpage(url, playlist_id)
232 self.url_result(compat_urlparse.urljoin(url, playlist), 'YoutubePlaylist')
233 for playlist in re.findall(r'href="(/playlist\?list=.+?)"', webpage)]
234 title = self._og_search_title(webpage, fatal=False)
235 return self.playlist_result(entries, playlist_id, title)
238 class YoutubeIE(YoutubeBaseInfoExtractor):
239 IE_DESC = 'YouTube.com'
240 _VALID_URL = r"""(?x)^
242 (?:https?://|//) # http(s):// or protocol-independent URL
243 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
244 (?:www\.)?deturl\.com/www\.youtube\.com/|
245 (?:www\.)?pwnyoutube\.com/|
246 (?:www\.)?yourepeat\.com/|
247 tube\.majestyc\.net/|
248 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
249 (?:.*?\#/)? # handle anchor (#/) redirect urls
250 (?: # the various things that can precede the ID:
251 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
252 |(?: # or the v= param in all its forms
253 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
254 (?:\?|\#!?) # the params delimiter ? or # or #!
255 (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx)
260 youtu\.be| # just youtu.be/xxxx
261 vid\.plus # or vid.plus/xxxx
263 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
265 )? # all until now is optional -> you can pass the naked ID
266 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
267 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
268 (?(1).+)? # if we found the ID, everything can follow
270 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
272 '5': {'ext': 'flv', 'width': 400, 'height': 240},
273 '6': {'ext': 'flv', 'width': 450, 'height': 270},
274 '13': {'ext': '3gp'},
275 '17': {'ext': '3gp', 'width': 176, 'height': 144},
276 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
277 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
278 '34': {'ext': 'flv', 'width': 640, 'height': 360},
279 '35': {'ext': 'flv', 'width': 854, 'height': 480},
280 '36': {'ext': '3gp', 'width': 320, 'height': 240},
281 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
282 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
283 '43': {'ext': 'webm', 'width': 640, 'height': 360},
284 '44': {'ext': 'webm', 'width': 854, 'height': 480},
285 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
286 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
287 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
288 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
292 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
293 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
294 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
295 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
296 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
297 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
298 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
300 # Apple HTTP Live Streaming
301 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
302 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
303 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
304 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
305 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
306 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
307 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
310 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
311 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
312 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
313 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
314 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
315 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
316 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
317 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
318 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
319 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
320 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
323 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
324 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
325 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
328 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
329 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
330 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
331 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
332 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
333 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
334 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'},
335 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
336 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
337 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
338 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
339 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
340 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
341 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
342 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
343 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
344 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
345 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
346 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
347 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
348 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
351 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
352 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
354 # Dash webm audio with opus inside
355 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
356 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
357 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
360 '_rtmp': {'protocol': 'rtmp'},
366 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
370 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
371 'uploader': 'Philipp Hagemeister',
372 'uploader_id': 'phihag',
373 'upload_date': '20121002',
374 '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 .',
375 'categories': ['Science & Technology'],
376 'tags': ['youtube-dl'],
378 'dislike_count': int,
384 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
385 'note': 'Test generic use_cipher_signature video (#897)',
389 'upload_date': '20120506',
390 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
391 'description': 'md5:782e8651347686cba06e58f71ab51773',
392 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
393 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
394 'iconic ep', 'iconic', 'love', 'it'],
395 'uploader': 'Icona Pop',
396 'uploader_id': 'IconaPop',
400 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
401 'note': 'Test VEVO video with age protection (#956)',
405 'upload_date': '20130703',
406 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
407 'description': 'md5:64249768eec3bc4276236606ea996373',
408 'uploader': 'justintimberlakeVEVO',
409 'uploader_id': 'justintimberlakeVEVO',
414 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
415 'note': 'Embed-only video (#1746)',
419 'upload_date': '20120608',
420 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
421 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
422 'uploader': 'SET India',
423 'uploader_id': 'setindia'
427 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
428 'note': 'Use the first video ID in the URL',
432 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
433 'uploader': 'Philipp Hagemeister',
434 'uploader_id': 'phihag',
435 'upload_date': '20121002',
436 '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 .',
437 'categories': ['Science & Technology'],
438 'tags': ['youtube-dl'],
440 'dislike_count': int,
443 'skip_download': True,
447 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
448 'note': '256k DASH audio (format 141) via DASH manifest',
452 'upload_date': '20121002',
453 'uploader_id': '8KVIDEO',
455 'uploader': '8KVIDEO',
456 'title': 'UHDTV TEST 8K VIDEO.mp4'
459 'youtube_include_dash_manifest': True,
463 # DASH manifest with encrypted signature
465 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
469 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
470 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
471 'uploader': 'AfrojackVEVO',
472 'uploader_id': 'AfrojackVEVO',
473 'upload_date': '20131011',
476 'youtube_include_dash_manifest': True,
480 # JS player signature function name containing $
482 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
486 'title': 'Taylor Swift - Shake It Off',
487 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
488 'uploader': 'TaylorSwiftVEVO',
489 'uploader_id': 'TaylorSwiftVEVO',
490 'upload_date': '20140818',
493 'youtube_include_dash_manifest': True,
499 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
503 'upload_date': '20100909',
504 'uploader': 'The Amazing Atheist',
505 'uploader_id': 'TheAmazingAtheist',
506 'title': 'Burning Everyone\'s Koran',
507 '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',
510 # Normal age-gate video (No vevo, embed allowed)
512 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
516 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
517 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
518 'uploader': 'The Witcher',
519 'uploader_id': 'WitcherGame',
520 'upload_date': '20140605',
524 # Age-gate video with encrypted signature
526 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
530 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
531 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
532 'uploader': 'LloydVEVO',
533 'uploader_id': 'LloydVEVO',
534 'upload_date': '20110629',
538 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
540 'url': '__2ABJjxzNo',
544 'upload_date': '20100430',
545 'uploader_id': 'deadmau5',
546 'description': 'md5:12c56784b8032162bb936a5f76d55360',
547 'uploader': 'deadmau5',
548 'title': 'Deadmau5 - Some Chords (HD)',
550 'expected_warnings': [
551 'DASH manifest missing',
554 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
556 'url': 'lqQg6PlCWgI',
560 'upload_date': '20120724',
561 'uploader_id': 'olympic',
562 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
563 'uploader': 'Olympics',
564 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
567 'skip_download': 'requires avconv',
572 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
576 'stretched_ratio': 16 / 9.,
577 'upload_date': '20110310',
578 'uploader_id': 'AllenMeow',
579 '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',
581 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
584 # url_encoded_fmt_stream_map is empty string
586 'url': 'qEJwOuvDf7I',
590 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
592 'upload_date': '20150404',
593 'uploader_id': 'spbelect',
594 'uploader': 'Наблюдатели Петербурга',
597 'skip_download': 'requires avconv',
600 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
602 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
606 'title': 'md5:7b81415841e02ecd4313668cde88737a',
607 'description': 'md5:116377fd2963b81ec4ce64b542173306',
608 'upload_date': '20150625',
609 'uploader_id': 'dorappi2000',
610 'uploader': 'dorappi2000',
611 'formats': 'mincount:33',
614 # DASH manifest with segment_list
616 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
617 'md5': '8ce563a1d667b599d21064e982ab9e31',
621 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
622 'uploader': 'Airtek',
623 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
624 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
625 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
628 'youtube_include_dash_manifest': True,
629 'format': '135', # bestvideo
633 # Multifeed videos (multiple cameras), URL is for Main Camera
634 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
637 'title': 'teamPGP: Rocket League Noob Stream',
638 'description': 'md5:dc7872fb300e143831327f1bae3af010',
644 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
645 'description': 'md5:dc7872fb300e143831327f1bae3af010',
646 'upload_date': '20150721',
647 'uploader': 'Beer Games Beer',
648 'uploader_id': 'beergamesbeer',
654 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
655 'description': 'md5:dc7872fb300e143831327f1bae3af010',
656 'upload_date': '20150721',
657 'uploader': 'Beer Games Beer',
658 'uploader_id': 'beergamesbeer',
664 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
665 'description': 'md5:dc7872fb300e143831327f1bae3af010',
666 'upload_date': '20150721',
667 'uploader': 'Beer Games Beer',
668 'uploader_id': 'beergamesbeer',
674 'title': 'teamPGP: Rocket League Noob Stream (zim)',
675 'description': 'md5:dc7872fb300e143831327f1bae3af010',
676 'upload_date': '20150721',
677 'uploader': 'Beer Games Beer',
678 'uploader_id': 'beergamesbeer',
682 'skip_download': True,
686 'url': 'http://vid.plus/FlRa-iH7PGw',
687 'only_matching': True,
691 def __init__(self, *args, **kwargs):
692 super(YoutubeIE, self).__init__(*args, **kwargs)
693 self._player_cache = {}
695 def report_video_info_webpage_download(self, video_id):
696 """Report attempt to download video info webpage."""
697 self.to_screen('%s: Downloading video info webpage' % video_id)
699 def report_information_extraction(self, video_id):
700 """Report attempt to extract video information."""
701 self.to_screen('%s: Extracting video information' % video_id)
703 def report_unavailable_format(self, video_id, format):
704 """Report extracted video URL."""
705 self.to_screen('%s: Format %s not available' % (video_id, format))
707 def report_rtmp_download(self):
708 """Indicate the download will use the RTMP protocol."""
709 self.to_screen('RTMP download detected')
711 def _signature_cache_id(self, example_sig):
712 """ Return a string representation of a signature """
713 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
715 def _extract_signature_function(self, video_id, player_url, example_sig):
717 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
720 raise ExtractorError('Cannot identify player %r' % player_url)
721 player_type = id_m.group('ext')
722 player_id = id_m.group('id')
724 # Read from filesystem cache
725 func_id = '%s_%s_%s' % (
726 player_type, player_id, self._signature_cache_id(example_sig))
727 assert os.path.basename(func_id) == func_id
729 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
730 if cache_spec is not None:
731 return lambda s: ''.join(s[i] for i in cache_spec)
734 'Downloading player %s' % player_url
735 if self._downloader.params.get('verbose') else
736 'Downloading %s player %s' % (player_type, player_id)
738 if player_type == 'js':
739 code = self._download_webpage(
740 player_url, video_id,
742 errnote='Download of %s failed' % player_url)
743 res = self._parse_sig_js(code)
744 elif player_type == 'swf':
745 urlh = self._request_webpage(
746 player_url, video_id,
748 errnote='Download of %s failed' % player_url)
750 res = self._parse_sig_swf(code)
752 assert False, 'Invalid player type %r' % player_type
754 test_string = ''.join(map(compat_chr, range(len(example_sig))))
755 cache_res = res(test_string)
756 cache_spec = [ord(c) for c in cache_res]
758 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
761 def _print_sig_code(self, func, example_sig):
762 def gen_sig_code(idxs):
763 def _genslice(start, end, step):
764 starts = '' if start == 0 else str(start)
765 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
766 steps = '' if step == 1 else (':%d' % step)
767 return 's[%s%s%s]' % (starts, ends, steps)
770 # Quelch pyflakes warnings - start will be set when step is set
771 start = '(Never used)'
772 for i, prev in zip(idxs[1:], idxs[:-1]):
776 yield _genslice(start, prev, step)
779 if i - prev in [-1, 1]:
788 yield _genslice(start, i, step)
790 test_string = ''.join(map(compat_chr, range(len(example_sig))))
791 cache_res = func(test_string)
792 cache_spec = [ord(c) for c in cache_res]
793 expr_code = ' + '.join(gen_sig_code(cache_spec))
794 signature_id_tuple = '(%s)' % (
795 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
796 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
797 ' return %s\n') % (signature_id_tuple, expr_code)
798 self.to_screen('Extracted signature function:\n' + code)
800 def _parse_sig_js(self, jscode):
801 funcname = self._search_regex(
802 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
803 'Initial JS player signature function name')
805 jsi = JSInterpreter(jscode)
806 initial_function = jsi.extract_function(funcname)
807 return lambda s: initial_function([s])
809 def _parse_sig_swf(self, file_contents):
810 swfi = SWFInterpreter(file_contents)
811 TARGET_CLASSNAME = 'SignatureDecipher'
812 searched_class = swfi.extract_class(TARGET_CLASSNAME)
813 initial_function = swfi.extract_function(searched_class, 'decipher')
814 return lambda s: initial_function([s])
816 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
817 """Turn the encrypted s field into a working signature"""
819 if player_url is None:
820 raise ExtractorError('Cannot decrypt signature without player_url')
822 if player_url.startswith('//'):
823 player_url = 'https:' + player_url
825 player_id = (player_url, self._signature_cache_id(s))
826 if player_id not in self._player_cache:
827 func = self._extract_signature_function(
828 video_id, player_url, s
830 self._player_cache[player_id] = func
831 func = self._player_cache[player_id]
832 if self._downloader.params.get('youtube_print_sig_code'):
833 self._print_sig_code(func, s)
835 except Exception as e:
836 tb = traceback.format_exc()
837 raise ExtractorError(
838 'Signature extraction failed: ' + tb, cause=e)
840 def _get_subtitles(self, video_id, webpage):
842 subs_doc = self._download_xml(
843 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
844 video_id, note=False)
845 except ExtractorError as err:
846 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
850 for track in subs_doc.findall('track'):
851 lang = track.attrib['lang_code']
852 if lang in sub_lang_list:
855 for ext in ['sbv', 'vtt', 'srt']:
856 params = compat_urllib_parse.urlencode({
860 'name': track.attrib['name'].encode('utf-8'),
863 'url': 'https://www.youtube.com/api/timedtext?' + params,
866 sub_lang_list[lang] = sub_formats
867 if not sub_lang_list:
868 self._downloader.report_warning('video doesn\'t have subtitles')
872 def _get_automatic_captions(self, video_id, webpage):
873 """We need the webpage for getting the captions url, pass it as an
874 argument to speed up the process."""
875 self.to_screen('%s: Looking for automatic captions' % video_id)
876 mobj = re.search(r';ytplayer.config = ({.*?});', webpage)
877 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
879 self._downloader.report_warning(err_msg)
881 player_config = json.loads(mobj.group(1))
883 args = player_config['args']
884 caption_url = args['ttsurl']
885 timestamp = args['timestamp']
886 # We get the available subtitles
887 list_params = compat_urllib_parse.urlencode({
892 list_url = caption_url + '&' + list_params
893 caption_list = self._download_xml(list_url, video_id)
894 original_lang_node = caption_list.find('track')
895 if original_lang_node is None:
896 self._downloader.report_warning('Video doesn\'t have automatic captions')
898 original_lang = original_lang_node.attrib['lang_code']
899 caption_kind = original_lang_node.attrib.get('kind', '')
902 for lang_node in caption_list.findall('target'):
903 sub_lang = lang_node.attrib['lang_code']
905 for ext in ['sbv', 'vtt', 'srt']:
906 params = compat_urllib_parse.urlencode({
907 'lang': original_lang,
911 'kind': caption_kind,
914 'url': caption_url + '&' + params,
917 sub_lang_list[sub_lang] = sub_formats
919 # An extractor error can be raise by the download process if there are
920 # no automatic captions but there are subtitles
921 except (KeyError, ExtractorError):
922 self._downloader.report_warning(err_msg)
926 def extract_id(cls, url):
927 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
929 raise ExtractorError('Invalid URL: %s' % url)
930 video_id = mobj.group(2)
933 def _extract_from_m3u8(self, manifest_url, video_id):
936 def _get_urls(_manifest):
937 lines = _manifest.split('\n')
938 urls = filter(lambda l: l and not l.startswith('#'),
941 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
942 formats_urls = _get_urls(manifest)
943 for format_url in formats_urls:
944 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
945 url_map[itag] = format_url
948 def _extract_annotations(self, video_id):
949 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
950 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
952 def _parse_dash_manifest(
953 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
954 def decrypt_sig(mobj):
956 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
957 return '/signature/%s' % dec_s
958 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
959 dash_doc = self._download_xml(
960 dash_manifest_url, video_id,
961 note='Downloading DASH manifest',
962 errnote='Could not download DASH manifest',
965 if dash_doc is False:
969 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
970 mime_type = a.attrib.get('mimeType')
971 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
972 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
975 if mime_type == 'text/vtt':
976 # TODO implement WebVTT downloading
978 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
979 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
980 format_id = r.attrib['id']
981 video_url = url_el.text
982 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
984 'format_id': format_id,
986 'width': int_or_none(r.attrib.get('width')),
987 'height': int_or_none(r.attrib.get('height')),
988 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
989 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
990 'filesize': filesize,
991 'fps': int_or_none(r.attrib.get('frameRate')),
993 if segment_list is not None:
995 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
996 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
997 'protocol': 'http_dash_segments',
1000 existing_format = next(
1001 fo for fo in formats
1002 if fo['format_id'] == format_id)
1003 except StopIteration:
1004 full_info = self._formats.get(format_id, {}).copy()
1006 codecs = r.attrib.get('codecs')
1008 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
1009 full_info['vcodec'] = codecs
1010 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
1011 full_info['acodec'] = codecs
1012 formats.append(full_info)
1014 existing_format.update(f)
1016 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
1019 def _real_extract(self, url):
1020 url, smuggled_data = unsmuggle_url(url, {})
1023 'http' if self._downloader.params.get('prefer_insecure', False)
1028 parsed_url = compat_urllib_parse_urlparse(url)
1029 for component in [parsed_url.fragment, parsed_url.query]:
1030 query = compat_parse_qs(component)
1031 if start_time is None and 't' in query:
1032 start_time = parse_duration(query['t'][0])
1033 if start_time is None and 'start' in query:
1034 start_time = parse_duration(query['start'][0])
1035 if end_time is None and 'end' in query:
1036 end_time = parse_duration(query['end'][0])
1038 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1039 mobj = re.search(self._NEXT_URL_RE, url)
1041 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1042 video_id = self.extract_id(url)
1045 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1046 video_webpage = self._download_webpage(url, video_id)
1048 # Attempt to extract SWF player URL
1049 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1050 if mobj is not None:
1051 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1057 def add_dash_mpd(video_info):
1058 dash_mpd = video_info.get('dashmpd')
1059 if dash_mpd and dash_mpd[0] not in dash_mpds:
1060 dash_mpds.append(dash_mpd[0])
1063 embed_webpage = None
1065 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1067 # We simulate the access to the video from www.youtube.com/v/{video_id}
1068 # this can be viewed without login into Youtube
1069 url = proto + '://www.youtube.com/embed/%s' % video_id
1070 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1071 data = compat_urllib_parse.urlencode({
1072 'video_id': video_id,
1073 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1074 'sts': self._search_regex(
1075 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1077 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1078 video_info_webpage = self._download_webpage(
1079 video_info_url, video_id,
1080 note='Refetching age-gated info webpage',
1081 errnote='unable to download video info webpage')
1082 video_info = compat_parse_qs(video_info_webpage)
1083 add_dash_mpd(video_info)
1087 # Try looking directly into the video webpage
1088 mobj = re.search(r';ytplayer\.config\s*=\s*({.*?});', video_webpage)
1090 json_code = uppercase_escape(mobj.group(1))
1091 ytplayer_config = json.loads(json_code)
1092 args = ytplayer_config['args']
1093 if args.get('url_encoded_fmt_stream_map'):
1094 # Convert to the same format returned by compat_parse_qs
1095 video_info = dict((k, [v]) for k, v in args.items())
1096 add_dash_mpd(video_info)
1097 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1099 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1100 # We also try looking in get_video_info since it may contain different dashmpd
1101 # URL that points to a DASH manifest with possibly different itag set (some itags
1102 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1103 # manifest pointed by get_video_info's dashmpd).
1104 # The general idea is to take a union of itags of both DASH manifests (for example
1105 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1106 self.report_video_info_webpage_download(video_id)
1107 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1109 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1110 % (proto, video_id, el_type))
1111 video_info_webpage = self._download_webpage(
1113 video_id, note=False,
1114 errnote='unable to download video info webpage')
1115 get_video_info = compat_parse_qs(video_info_webpage)
1116 if get_video_info.get('use_cipher_signature') != ['True']:
1117 add_dash_mpd(get_video_info)
1119 video_info = get_video_info
1120 if 'token' in get_video_info:
1121 # Different get_video_info requests may report different results, e.g.
1122 # some may report video unavailability, but some may serve it without
1123 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1124 # the original webpage as well as el=info and el=embedded get_video_info
1125 # requests report video unavailability due to geo restriction while
1126 # el=detailpage succeeds and returns valid data). This is probably
1127 # due to YouTube measures against IP ranges of hosting providers.
1128 # Working around by preferring the first succeeded video_info containing
1129 # the token if no such video_info yet was found.
1130 if 'token' not in video_info:
1131 video_info = get_video_info
1133 if 'token' not in video_info:
1134 if 'reason' in video_info:
1135 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1136 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1138 raise ExtractorError('YouTube said: This video is available in %s only' % (
1139 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1141 raise ExtractorError(
1142 'YouTube said: %s' % video_info['reason'][0],
1143 expected=True, video_id=video_id)
1145 raise ExtractorError(
1146 '"token" parameter not in video info for unknown reason',
1150 if 'title' in video_info:
1151 video_title = video_info['title'][0]
1153 self._downloader.report_warning('Unable to extract video title')
1157 video_description = get_element_by_id("eow-description", video_webpage)
1158 if video_description:
1159 video_description = re.sub(r'''(?x)
1161 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1163 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1164 class="yt-uix-redirect-link"\s*>
1167 ''', r'\1', video_description)
1168 video_description = clean_html(video_description)
1170 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1172 video_description = unescapeHTML(fd_mobj.group(1))
1174 video_description = ''
1176 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1177 if not self._downloader.params.get('noplaylist'):
1180 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1181 for feed in multifeed_metadata_list.split(','):
1182 feed_data = compat_parse_qs(feed)
1184 '_type': 'url_transparent',
1185 'ie_key': 'Youtube',
1187 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1188 {'force_singlefeed': True}),
1189 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1191 feed_ids.append(feed_data['id'][0])
1193 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1194 % (', '.join(feed_ids), video_id))
1195 return self.playlist_result(entries, video_id, video_title, video_description)
1196 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1198 if 'view_count' in video_info:
1199 view_count = int(video_info['view_count'][0])
1203 # Check for "rental" videos
1204 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1205 raise ExtractorError('"rental" videos not supported')
1207 # Start extracting information
1208 self.report_information_extraction(video_id)
1211 if 'author' not in video_info:
1212 raise ExtractorError('Unable to extract uploader name')
1213 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1216 video_uploader_id = None
1217 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1218 if mobj is not None:
1219 video_uploader_id = mobj.group(1)
1221 self._downloader.report_warning('unable to extract uploader nickname')
1224 # We try first to get a high quality image:
1225 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1226 video_webpage, re.DOTALL)
1227 if m_thumb is not None:
1228 video_thumbnail = m_thumb.group(1)
1229 elif 'thumbnail_url' not in video_info:
1230 self._downloader.report_warning('unable to extract video thumbnail')
1231 video_thumbnail = None
1232 else: # don't panic if we can't find it
1233 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1236 upload_date = self._html_search_meta(
1237 'datePublished', video_webpage, 'upload date', default=None)
1239 upload_date = self._search_regex(
1240 [r'(?s)id="eow-date.*?>(.*?)</span>',
1241 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1242 video_webpage, 'upload date', default=None)
1244 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1245 upload_date = unified_strdate(upload_date)
1247 m_cat_container = self._search_regex(
1248 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1249 video_webpage, 'categories', default=None)
1251 category = self._html_search_regex(
1252 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1254 video_categories = None if category is None else [category]
1256 video_categories = None
1259 unescapeHTML(m.group('content'))
1260 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1262 def _extract_count(count_name):
1263 return str_to_int(self._search_regex(
1264 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1265 % re.escape(count_name),
1266 video_webpage, count_name, default=None))
1268 like_count = _extract_count('like')
1269 dislike_count = _extract_count('dislike')
1272 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1273 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1275 if 'length_seconds' not in video_info:
1276 self._downloader.report_warning('unable to extract video duration')
1277 video_duration = None
1279 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1282 video_annotations = None
1283 if self._downloader.params.get('writeannotations', False):
1284 video_annotations = self._extract_annotations(video_id)
1286 def _map_to_format_list(urlmap):
1288 for itag, video_real_url in urlmap.items():
1291 'url': video_real_url,
1292 'player_url': player_url,
1294 if itag in self._formats:
1295 dct.update(self._formats[itag])
1299 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1300 self.report_rtmp_download()
1302 'format_id': '_rtmp',
1304 'url': video_info['conn'][0],
1305 'player_url': player_url,
1307 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1308 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1309 if 'rtmpe%3Dyes' in encoded_url_map:
1310 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1312 for url_data_str in encoded_url_map.split(','):
1313 url_data = compat_parse_qs(url_data_str)
1314 if 'itag' not in url_data or 'url' not in url_data:
1316 format_id = url_data['itag'][0]
1317 url = url_data['url'][0]
1319 if 'sig' in url_data:
1320 url += '&signature=' + url_data['sig'][0]
1321 elif 's' in url_data:
1322 encrypted_sig = url_data['s'][0]
1323 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1325 jsplayer_url_json = self._search_regex(
1327 embed_webpage if age_gate else video_webpage,
1328 'JS player URL (1)', default=None)
1329 if not jsplayer_url_json and not age_gate:
1330 # We need the embed website after all
1331 if embed_webpage is None:
1332 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1333 embed_webpage = self._download_webpage(
1334 embed_url, video_id, 'Downloading embed webpage')
1335 jsplayer_url_json = self._search_regex(
1336 ASSETS_RE, embed_webpage, 'JS player URL')
1338 player_url = json.loads(jsplayer_url_json)
1339 if player_url is None:
1340 player_url_json = self._search_regex(
1341 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1342 video_webpage, 'age gate player URL')
1343 player_url = json.loads(player_url_json)
1345 if self._downloader.params.get('verbose'):
1346 if player_url is None:
1347 player_version = 'unknown'
1348 player_desc = 'unknown'
1350 if player_url.endswith('swf'):
1351 player_version = self._search_regex(
1352 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1353 'flash player', fatal=False)
1354 player_desc = 'flash player %s' % player_version
1356 player_version = self._search_regex(
1357 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1359 'html5 player', fatal=False)
1360 player_desc = 'html5 player %s' % player_version
1362 parts_sizes = self._signature_cache_id(encrypted_sig)
1363 self.to_screen('{%s} signature length %s, %s' %
1364 (format_id, parts_sizes, player_desc))
1366 signature = self._decrypt_signature(
1367 encrypted_sig, video_id, player_url, age_gate)
1368 url += '&signature=' + signature
1369 if 'ratebypass' not in url:
1370 url += '&ratebypass=yes'
1372 # Some itags are not included in DASH manifest thus corresponding formats will
1373 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1374 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1375 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1376 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1378 'format_id': format_id,
1380 'player_url': player_url,
1381 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1382 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1385 'fps': int_or_none(url_data.get('fps', [None])[0]),
1386 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1388 type_ = url_data.get('type', [None])[0]
1390 type_split = type_.split(';')
1391 kind_ext = type_split[0].split('/')
1392 if len(kind_ext) == 2:
1393 kind, ext = kind_ext
1395 if kind in ('audio', 'video'):
1397 for mobj in re.finditer(
1398 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1399 if mobj.group('key') == 'codecs':
1400 codecs = mobj.group('val')
1403 codecs = codecs.split(',')
1404 if len(codecs) == 2:
1405 acodec, vcodec = codecs[0], codecs[1]
1407 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1412 if format_id in self._formats:
1413 dct.update(self._formats[format_id])
1415 elif video_info.get('hlsvp'):
1416 manifest_url = video_info['hlsvp'][0]
1417 url_map = self._extract_from_m3u8(manifest_url, video_id)
1418 formats = _map_to_format_list(url_map)
1420 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1422 # Look for the DASH manifest
1423 if self._downloader.params.get('youtube_include_dash_manifest', True):
1424 dash_mpd_fatal = True
1425 for dash_manifest_url in dash_mpds:
1428 for df in self._parse_dash_manifest(
1429 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1430 # Do not overwrite DASH format found in some previous DASH manifest
1431 if df['format_id'] not in dash_formats:
1432 dash_formats[df['format_id']] = df
1433 # Additional DASH manifests may end up in HTTP Error 403 therefore
1434 # allow them to fail without bug report message if we already have
1435 # some DASH manifest succeeded. This is temporary workaround to reduce
1436 # burst of bug reports until we figure out the reason and whether it
1437 # can be fixed at all.
1438 dash_mpd_fatal = False
1439 except (ExtractorError, KeyError) as e:
1440 self.report_warning(
1441 'Skipping DASH manifest: %r' % e, video_id)
1443 # Remove the formats we found through non-DASH, they
1444 # contain less info and it can be wrong, because we use
1445 # fixed values (for example the resolution). See
1446 # https://github.com/rg3/youtube-dl/issues/5774 for an
1448 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1449 formats.extend(dash_formats.values())
1451 # Check for malformed aspect ratio
1452 stretched_m = re.search(
1453 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1456 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1458 if f.get('vcodec') != 'none':
1459 f['stretched_ratio'] = ratio
1461 self._sort_formats(formats)
1465 'uploader': video_uploader,
1466 'uploader_id': video_uploader_id,
1467 'upload_date': upload_date,
1468 'title': video_title,
1469 'thumbnail': video_thumbnail,
1470 'description': video_description,
1471 'categories': video_categories,
1473 'subtitles': video_subtitles,
1474 'automatic_captions': automatic_captions,
1475 'duration': video_duration,
1476 'age_limit': 18 if age_gate else 0,
1477 'annotations': video_annotations,
1478 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1479 'view_count': view_count,
1480 'like_count': like_count,
1481 'dislike_count': dislike_count,
1482 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1485 'start_time': start_time,
1486 'end_time': end_time,
1490 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1491 IE_DESC = 'YouTube.com playlists'
1492 _VALID_URL = r"""(?x)(?:
1497 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1498 \? (?:.*?&)*? (?:p|a|list)=
1502 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1503 # Top tracks, they can also include dots
1508 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1510 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1511 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1512 IE_NAME = 'youtube:playlist'
1514 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1516 'title': 'ytdl test PL',
1517 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1519 'playlist_count': 3,
1521 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1523 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1524 'title': 'YDL_Empty_List',
1526 'playlist_count': 0,
1528 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1529 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1531 'title': '29C3: Not my department',
1532 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1534 'playlist_count': 95,
1536 'note': 'issue #673',
1537 'url': 'PLBB231211A4F62143',
1539 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1540 'id': 'PLBB231211A4F62143',
1542 'playlist_mincount': 26,
1544 'note': 'Large playlist',
1545 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1547 'title': 'Uploads from Cauchemar',
1548 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1550 'playlist_mincount': 799,
1552 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1554 'title': 'YDL_safe_search',
1555 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1557 'playlist_count': 2,
1560 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1561 'playlist_count': 4,
1564 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1567 'note': 'Embedded SWF player',
1568 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1569 'playlist_count': 4,
1572 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1575 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1576 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1578 'title': 'Uploads from Interstellar Movie',
1579 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1581 'playlist_mincout': 21,
1584 def _real_initialize(self):
1587 def _extract_mix(self, playlist_id):
1588 # The mixes are generated from a single video
1589 # the id of the playlist is just 'RD' + video_id
1590 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1591 webpage = self._download_webpage(
1592 url, playlist_id, 'Downloading Youtube mix')
1593 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1595 search_title('playlist-title') or
1596 search_title('title long-title') or
1597 search_title('title'))
1598 title = clean_html(title_span)
1599 ids = orderedSet(re.findall(
1600 r'''(?xs)data-video-username=".*?".*?
1601 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1603 url_results = self._ids_to_results(ids)
1605 return self.playlist_result(url_results, playlist_id, title)
1607 def _extract_playlist(self, playlist_id):
1608 url = self._TEMPLATE_URL % playlist_id
1609 page = self._download_webpage(url, playlist_id)
1611 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1612 match = match.strip()
1613 # Check if the playlist exists or is private
1614 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1615 raise ExtractorError(
1616 'The playlist doesn\'t exist or is private, use --username or '
1617 '--netrc to access it.',
1619 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1620 raise ExtractorError(
1621 'Invalid parameters. Maybe URL is incorrect.',
1623 elif re.match(r'[^<]*Choose your language[^<]*', match):
1626 self.report_warning('Youtube gives an alert message: ' + match)
1628 playlist_title = self._html_search_regex(
1629 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1632 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1634 def _real_extract(self, url):
1635 # Extract playlist id
1636 mobj = re.match(self._VALID_URL, url)
1638 raise ExtractorError('Invalid URL: %s' % url)
1639 playlist_id = mobj.group(1) or mobj.group(2)
1641 # Check if it's a video-specific URL
1642 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1643 if 'v' in query_dict:
1644 video_id = query_dict['v'][0]
1645 if self._downloader.params.get('noplaylist'):
1646 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1647 return self.url_result(video_id, 'Youtube', video_id=video_id)
1649 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1651 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1652 # Mixes require a custom extraction process
1653 return self._extract_mix(playlist_id)
1655 return self._extract_playlist(playlist_id)
1658 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1659 IE_DESC = 'YouTube.com channels'
1660 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1661 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1662 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1663 IE_NAME = 'youtube:channel'
1665 'note': 'paginated channel',
1666 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1667 'playlist_mincount': 91,
1669 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1670 'title': 'Uploads from lex will',
1673 'note': 'Age restricted channel',
1674 # from https://www.youtube.com/user/DeusExOfficial
1675 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1676 'playlist_mincount': 64,
1678 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1679 'title': 'Uploads from Deus Ex',
1683 def _real_extract(self, url):
1684 channel_id = self._match_id(url)
1686 url = self._TEMPLATE_URL % channel_id
1688 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1689 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1690 # otherwise fallback on channel by page extraction
1691 channel_page = self._download_webpage(
1692 url + '?view=57', channel_id,
1693 'Downloading channel page', fatal=False)
1694 if channel_page is False:
1695 channel_playlist_id = False
1697 channel_playlist_id = self._html_search_meta(
1698 'channelId', channel_page, 'channel id', default=None)
1699 if not channel_playlist_id:
1700 channel_playlist_id = self._search_regex(
1701 r'data-(?:channel-external-|yt)id="([^"]+)"',
1702 channel_page, 'channel id', default=None)
1703 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1704 playlist_id = 'UU' + channel_playlist_id[2:]
1705 return self.url_result(
1706 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1708 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1709 autogenerated = re.search(r'''(?x)
1711 channel-header-autogenerated-label|
1712 yt-channel-title-autogenerated
1713 )[^"]*"''', channel_page) is not None
1716 # The videos are contained in a single page
1717 # the ajax pages can't be used, they are empty
1720 video_id, 'Youtube', video_id=video_id,
1721 video_title=video_title)
1722 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1723 return self.playlist_result(entries, channel_id)
1725 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1728 class YoutubeUserIE(YoutubeChannelIE):
1729 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1730 _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_-]+)'
1731 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1732 IE_NAME = 'youtube:user'
1735 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1736 'playlist_mincount': 320,
1738 'title': 'TheLinuxFoundation',
1741 'url': 'ytuser:phihag',
1742 'only_matching': True,
1746 def suitable(cls, url):
1747 # Don't return True if the url can be extracted with other youtube
1748 # extractor, the regex would is too permissive and it would match.
1749 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1750 if any(ie.suitable(url) for ie in other_ies):
1753 return super(YoutubeUserIE, cls).suitable(url)
1756 class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1757 IE_DESC = 'YouTube.com user playlists'
1758 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists'
1759 IE_NAME = 'youtube:user:playlists'
1762 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1763 'playlist_mincount': 4,
1765 'id': 'ThirstForScience',
1766 'title': 'Thirst for Science',
1771 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1772 IE_DESC = 'YouTube.com searches'
1773 # there doesn't appear to be a real limit, for example if you search for
1774 # 'python' you get more than 8.000.000 results
1775 _MAX_RESULTS = float('inf')
1776 IE_NAME = 'youtube:search'
1777 _SEARCH_KEY = 'ytsearch'
1778 _EXTRA_QUERY_ARGS = {}
1781 def _get_n_results(self, query, n):
1782 """Get a specified number of results for a query"""
1787 for pagenum in itertools.count(1):
1789 'search_query': query.encode('utf-8'),
1793 url_query.update(self._EXTRA_QUERY_ARGS)
1794 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1795 data = self._download_json(
1796 result_url, video_id='query "%s"' % query,
1797 note='Downloading page %s' % pagenum,
1798 errnote='Unable to download API page')
1799 html_content = data[1]['body']['content']
1801 if 'class="search-message' in html_content:
1802 raise ExtractorError(
1803 '[youtube] No video results', expected=True)
1805 new_videos = self._ids_to_results(orderedSet(re.findall(
1806 r'href="/watch\?v=(.{11})', html_content)))
1807 videos += new_videos
1808 if not new_videos or len(videos) > limit:
1813 return self.playlist_result(videos, query)
1816 class YoutubeSearchDateIE(YoutubeSearchIE):
1817 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1818 _SEARCH_KEY = 'ytsearchdate'
1819 IE_DESC = 'YouTube.com searches, newest videos first'
1820 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1823 class YoutubeSearchURLIE(InfoExtractor):
1824 IE_DESC = 'YouTube.com search URLs'
1825 IE_NAME = 'youtube:search_url'
1826 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1828 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1829 'playlist_mincount': 5,
1831 'title': 'youtube-dl test video',
1835 def _real_extract(self, url):
1836 mobj = re.match(self._VALID_URL, url)
1837 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1839 webpage = self._download_webpage(url, query)
1840 result_code = self._search_regex(
1841 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1843 part_codes = re.findall(
1844 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1846 for part_code in part_codes:
1847 part_title = self._html_search_regex(
1848 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1849 part_url_snippet = self._html_search_regex(
1850 r'(?s)href="([^"]+)"', part_code, 'item URL')
1851 part_url = compat_urlparse.urljoin(
1852 'https://www.youtube.com/', part_url_snippet)
1856 'title': part_title,
1860 '_type': 'playlist',
1866 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1867 IE_DESC = 'YouTube.com (multi-season) shows'
1868 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1869 IE_NAME = 'youtube:show'
1871 'url': 'https://www.youtube.com/show/airdisasters',
1872 'playlist_mincount': 5,
1874 'id': 'airdisasters',
1875 'title': 'Air Disasters',
1879 def _real_extract(self, url):
1880 playlist_id = self._match_id(url)
1881 return super(YoutubeShowIE, self)._real_extract(
1882 'https://www.youtube.com/show/%s/playlists' % playlist_id)
1885 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1887 Base class for feed extractors
1888 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1890 _LOGIN_REQUIRED = True
1894 return 'youtube:%s' % self._FEED_NAME
1896 def _real_initialize(self):
1899 def _real_extract(self, url):
1900 page = self._download_webpage(
1901 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1903 # The extraction process is the same as for playlists, but the regex
1904 # for the video ids doesn't contain an index
1906 more_widget_html = content_html = page
1907 for page_num in itertools.count(1):
1908 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1910 # 'recommended' feed has infinite 'load more' and each new portion spins
1911 # the same videos in (sometimes) slightly different order, so we'll check
1912 # for unicity and break when portion has no new videos
1913 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1919 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1923 more = self._download_json(
1924 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1925 'Downloading page #%s' % page_num,
1926 transform_source=uppercase_escape)
1927 content_html = more['content_html']
1928 more_widget_html = more['load_more_widget_html']
1930 return self.playlist_result(
1931 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1934 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1935 IE_NAME = 'youtube:watchlater'
1936 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1937 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1939 _TESTS = [] # override PlaylistIE tests
1941 def _real_extract(self, url):
1942 return self._extract_playlist('WL')
1945 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1946 IE_NAME = 'youtube:favorites'
1947 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1948 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1949 _LOGIN_REQUIRED = True
1951 def _real_extract(self, url):
1952 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1953 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
1954 return self.url_result(playlist_id, 'YoutubePlaylist')
1957 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
1958 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1959 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1960 _FEED_NAME = 'recommended'
1961 _PLAYLIST_TITLE = 'Youtube Recommended videos'
1964 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
1965 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1966 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1967 _FEED_NAME = 'subscriptions'
1968 _PLAYLIST_TITLE = 'Youtube Subscriptions'
1971 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
1972 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
1973 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
1974 _FEED_NAME = 'history'
1975 _PLAYLIST_TITLE = 'Youtube History'
1978 class YoutubeTruncatedURLIE(InfoExtractor):
1979 IE_NAME = 'youtube:truncated_url'
1980 IE_DESC = False # Do not list
1981 _VALID_URL = r'''(?x)
1983 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1986 annotation_id=annotation_[^&]+|
1992 attribution_link\?a=[^&]+
1998 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1999 'only_matching': True,
2001 'url': 'http://www.youtube.com/watch?',
2002 'only_matching': True,
2004 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2005 'only_matching': True,
2007 'url': 'https://www.youtube.com/watch?feature=foo',
2008 'only_matching': True,
2010 'url': 'https://www.youtube.com/watch?hl=en-GB',
2011 'only_matching': True,
2013 'url': 'https://www.youtube.com/watch?t=2372',
2014 'only_matching': True,
2017 def _real_extract(self, url):
2018 raise ExtractorError(
2019 'Did you forget to quote the URL? Remember that & is a meta '
2020 'character in most shells, so you want to put the URL in quotes, '
2022 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2023 ' or simply youtube-dl BaW_jenozKc .',
2027 class YoutubeTruncatedIDIE(InfoExtractor):
2028 IE_NAME = 'youtube:truncated_id'
2029 IE_DESC = False # Do not list
2030 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2033 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2034 'only_matching': True,
2037 def _real_extract(self, url):
2038 video_id = self._match_id(url)
2039 raise ExtractorError(
2040 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),