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 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
387 'upload_date': '20121002',
388 'license': 'Standard YouTube License',
389 '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 .',
390 'categories': ['Science & Technology'],
391 'tags': ['youtube-dl'],
393 'dislike_count': int,
399 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
400 'note': 'Test generic use_cipher_signature video (#897)',
404 'upload_date': '20120506',
405 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
406 'alt_title': 'I Love It (feat. Charli XCX)',
407 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
408 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
409 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
410 'iconic ep', 'iconic', 'love', 'it'],
411 'uploader': 'Icona Pop',
412 'uploader_id': 'IconaPop',
413 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
414 'license': 'Standard YouTube License',
415 'creator': 'Icona Pop',
419 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
420 'note': 'Test VEVO video with age protection (#956)',
424 'upload_date': '20130703',
425 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
426 'alt_title': 'Tunnel Vision',
427 'description': 'md5:64249768eec3bc4276236606ea996373',
428 'uploader': 'justintimberlakeVEVO',
429 'uploader_id': 'justintimberlakeVEVO',
430 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
431 'license': 'Standard YouTube License',
432 'creator': 'Justin Timberlake',
437 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
438 'note': 'Embed-only video (#1746)',
442 'upload_date': '20120608',
443 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
444 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
445 'uploader': 'SET India',
446 'uploader_id': 'setindia',
447 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
448 'license': 'Standard YouTube License',
453 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
454 'note': 'Use the first video ID in the URL',
458 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
459 'uploader': 'Philipp Hagemeister',
460 'uploader_id': 'phihag',
461 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
462 'upload_date': '20121002',
463 'license': 'Standard YouTube License',
464 '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 .',
465 'categories': ['Science & Technology'],
466 'tags': ['youtube-dl'],
468 'dislike_count': int,
471 'skip_download': True,
475 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
476 'note': '256k DASH audio (format 141) via DASH manifest',
480 'upload_date': '20121002',
481 'uploader_id': '8KVIDEO',
482 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
484 'uploader': '8KVIDEO',
485 'license': 'Standard YouTube License',
486 'title': 'UHDTV TEST 8K VIDEO.mp4'
489 'youtube_include_dash_manifest': True,
493 # DASH manifest with encrypted signature
495 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
499 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
500 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
501 'uploader': 'AfrojackVEVO',
502 'uploader_id': 'AfrojackVEVO',
503 'upload_date': '20131011',
504 'license': 'Standard YouTube License',
507 'youtube_include_dash_manifest': True,
511 # JS player signature function name containing $
513 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
517 'title': 'Taylor Swift - Shake It Off',
518 'alt_title': 'Shake It Off',
519 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
520 'uploader': 'TaylorSwiftVEVO',
521 'uploader_id': 'TaylorSwiftVEVO',
522 'upload_date': '20140818',
523 'license': 'Standard YouTube License',
524 'creator': 'Taylor Swift',
527 'youtube_include_dash_manifest': True,
533 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
537 'upload_date': '20100909',
538 'uploader': 'The Amazing Atheist',
539 'uploader_id': 'TheAmazingAtheist',
540 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
541 'license': 'Standard YouTube License',
542 'title': 'Burning Everyone\'s Koran',
543 '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',
546 # Normal age-gate video (No vevo, embed allowed)
548 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
552 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
553 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
554 'uploader': 'The Witcher',
555 'uploader_id': 'WitcherGame',
556 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
557 'upload_date': '20140605',
558 'license': 'Standard YouTube License',
562 # Age-gate video with encrypted signature
564 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
568 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
569 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
570 'uploader': 'LloydVEVO',
571 'uploader_id': 'LloydVEVO',
572 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
573 'upload_date': '20110629',
574 'license': 'Standard YouTube License',
578 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
580 'url': '__2ABJjxzNo',
584 'upload_date': '20100430',
585 'uploader_id': 'deadmau5',
586 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
587 'creator': 'deadmau5',
588 'description': 'md5:12c56784b8032162bb936a5f76d55360',
589 'uploader': 'deadmau5',
590 'license': 'Standard YouTube License',
591 'title': 'Deadmau5 - Some Chords (HD)',
592 'alt_title': 'Some Chords',
594 'expected_warnings': [
595 'DASH manifest missing',
598 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
600 'url': 'lqQg6PlCWgI',
604 'upload_date': '20150827',
605 'uploader_id': 'olympic',
606 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
607 'license': 'Standard YouTube License',
608 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
609 'uploader': 'Olympics',
610 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
613 'skip_download': 'requires avconv',
618 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
622 'stretched_ratio': 16 / 9.,
623 'upload_date': '20110310',
624 'uploader_id': 'AllenMeow',
625 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
626 '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',
628 'license': 'Standard YouTube License',
629 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
632 # url_encoded_fmt_stream_map is empty string
634 'url': 'qEJwOuvDf7I',
638 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
640 'upload_date': '20150404',
641 'uploader_id': 'spbelect',
642 'uploader': 'Наблюдатели Петербурга',
645 'skip_download': 'requires avconv',
647 'skip': 'This live event has ended.',
649 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
651 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
655 'title': 'md5:7b81415841e02ecd4313668cde88737a',
656 'description': 'md5:116377fd2963b81ec4ce64b542173306',
657 'upload_date': '20150625',
658 'uploader_id': 'dorappi2000',
659 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
660 'uploader': 'dorappi2000',
661 'license': 'Standard YouTube License',
662 'formats': 'mincount:33',
665 # DASH manifest with segment_list
667 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
668 'md5': '8ce563a1d667b599d21064e982ab9e31',
672 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
673 'uploader': 'Airtek',
674 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
675 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
676 'license': 'Standard YouTube License',
677 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
680 'youtube_include_dash_manifest': True,
681 'format': '135', # bestvideo
685 # Multifeed videos (multiple cameras), URL is for Main Camera
686 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
689 'title': 'teamPGP: Rocket League Noob Stream',
690 'description': 'md5:dc7872fb300e143831327f1bae3af010',
696 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
697 'description': 'md5:dc7872fb300e143831327f1bae3af010',
698 'upload_date': '20150721',
699 'uploader': 'Beer Games Beer',
700 'uploader_id': 'beergamesbeer',
701 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
702 'license': 'Standard YouTube License',
708 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
709 'description': 'md5:dc7872fb300e143831327f1bae3af010',
710 'upload_date': '20150721',
711 'uploader': 'Beer Games Beer',
712 'uploader_id': 'beergamesbeer',
713 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
714 'license': 'Standard YouTube License',
720 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
721 'description': 'md5:dc7872fb300e143831327f1bae3af010',
722 'upload_date': '20150721',
723 'uploader': 'Beer Games Beer',
724 'uploader_id': 'beergamesbeer',
725 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
726 'license': 'Standard YouTube License',
732 'title': 'teamPGP: Rocket League Noob Stream (zim)',
733 'description': 'md5:dc7872fb300e143831327f1bae3af010',
734 'upload_date': '20150721',
735 'uploader': 'Beer Games Beer',
736 'uploader_id': 'beergamesbeer',
737 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
738 'license': 'Standard YouTube License',
742 'skip_download': True,
746 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
747 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
750 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
755 'url': 'http://vid.plus/FlRa-iH7PGw',
756 'only_matching': True,
759 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
760 # Also tests cut-off URL expansion in video description (see
761 # https://github.com/rg3/youtube-dl/issues/1892,
762 # https://github.com/rg3/youtube-dl/issues/8164)
763 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
767 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
768 'alt_title': 'Dark Walk',
769 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
770 'upload_date': '20151119',
771 'uploader_id': 'IronSoulElf',
772 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
773 'uploader': 'IronSoulElf',
774 'license': 'Standard YouTube License',
775 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
778 'skip_download': True,
782 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
783 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
784 'only_matching': True,
787 # Video with yt:stretch=17:0
788 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
792 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
793 'description': 'md5:ee18a25c350637c8faff806845bddee9',
794 'upload_date': '20151107',
795 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
796 'uploader': 'CH GAMER DROID',
799 'skip_download': True,
803 # Video licensed under Creative Commons
804 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
808 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
809 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
810 'upload_date': '20150127',
811 'uploader_id': 'BerkmanCenter',
812 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
813 'uploader': 'BerkmanCenter',
814 'license': 'Creative Commons Attribution license (reuse allowed)',
817 'skip_download': True,
821 # Channel-like uploader_url
822 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
826 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
827 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
828 'upload_date': '20151119',
829 'uploader': 'Bernie 2016',
830 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
831 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
832 'license': 'Creative Commons Attribution license (reuse allowed)',
835 'skip_download': True,
839 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
840 'only_matching': True,
844 def __init__(self, *args, **kwargs):
845 super(YoutubeIE, self).__init__(*args, **kwargs)
846 self._player_cache = {}
848 def report_video_info_webpage_download(self, video_id):
849 """Report attempt to download video info webpage."""
850 self.to_screen('%s: Downloading video info webpage' % video_id)
852 def report_information_extraction(self, video_id):
853 """Report attempt to extract video information."""
854 self.to_screen('%s: Extracting video information' % video_id)
856 def report_unavailable_format(self, video_id, format):
857 """Report extracted video URL."""
858 self.to_screen('%s: Format %s not available' % (video_id, format))
860 def report_rtmp_download(self):
861 """Indicate the download will use the RTMP protocol."""
862 self.to_screen('RTMP download detected')
864 def _signature_cache_id(self, example_sig):
865 """ Return a string representation of a signature """
866 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
868 def _extract_signature_function(self, video_id, player_url, example_sig):
870 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
873 raise ExtractorError('Cannot identify player %r' % player_url)
874 player_type = id_m.group('ext')
875 player_id = id_m.group('id')
877 # Read from filesystem cache
878 func_id = '%s_%s_%s' % (
879 player_type, player_id, self._signature_cache_id(example_sig))
880 assert os.path.basename(func_id) == func_id
882 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
883 if cache_spec is not None:
884 return lambda s: ''.join(s[i] for i in cache_spec)
887 'Downloading player %s' % player_url
888 if self._downloader.params.get('verbose') else
889 'Downloading %s player %s' % (player_type, player_id)
891 if player_type == 'js':
892 code = self._download_webpage(
893 player_url, video_id,
895 errnote='Download of %s failed' % player_url)
896 res = self._parse_sig_js(code)
897 elif player_type == 'swf':
898 urlh = self._request_webpage(
899 player_url, video_id,
901 errnote='Download of %s failed' % player_url)
903 res = self._parse_sig_swf(code)
905 assert False, 'Invalid player type %r' % player_type
907 test_string = ''.join(map(compat_chr, range(len(example_sig))))
908 cache_res = res(test_string)
909 cache_spec = [ord(c) for c in cache_res]
911 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
914 def _print_sig_code(self, func, example_sig):
915 def gen_sig_code(idxs):
916 def _genslice(start, end, step):
917 starts = '' if start == 0 else str(start)
918 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
919 steps = '' if step == 1 else (':%d' % step)
920 return 's[%s%s%s]' % (starts, ends, steps)
923 # Quelch pyflakes warnings - start will be set when step is set
924 start = '(Never used)'
925 for i, prev in zip(idxs[1:], idxs[:-1]):
929 yield _genslice(start, prev, step)
932 if i - prev in [-1, 1]:
941 yield _genslice(start, i, step)
943 test_string = ''.join(map(compat_chr, range(len(example_sig))))
944 cache_res = func(test_string)
945 cache_spec = [ord(c) for c in cache_res]
946 expr_code = ' + '.join(gen_sig_code(cache_spec))
947 signature_id_tuple = '(%s)' % (
948 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
949 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
950 ' return %s\n') % (signature_id_tuple, expr_code)
951 self.to_screen('Extracted signature function:\n' + code)
953 def _parse_sig_js(self, jscode):
954 funcname = self._search_regex(
955 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
956 'Initial JS player signature function name')
958 jsi = JSInterpreter(jscode)
959 initial_function = jsi.extract_function(funcname)
960 return lambda s: initial_function([s])
962 def _parse_sig_swf(self, file_contents):
963 swfi = SWFInterpreter(file_contents)
964 TARGET_CLASSNAME = 'SignatureDecipher'
965 searched_class = swfi.extract_class(TARGET_CLASSNAME)
966 initial_function = swfi.extract_function(searched_class, 'decipher')
967 return lambda s: initial_function([s])
969 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
970 """Turn the encrypted s field into a working signature"""
972 if player_url is None:
973 raise ExtractorError('Cannot decrypt signature without player_url')
975 if player_url.startswith('//'):
976 player_url = 'https:' + player_url
978 player_id = (player_url, self._signature_cache_id(s))
979 if player_id not in self._player_cache:
980 func = self._extract_signature_function(
981 video_id, player_url, s
983 self._player_cache[player_id] = func
984 func = self._player_cache[player_id]
985 if self._downloader.params.get('youtube_print_sig_code'):
986 self._print_sig_code(func, s)
988 except Exception as e:
989 tb = traceback.format_exc()
990 raise ExtractorError(
991 'Signature extraction failed: ' + tb, cause=e)
993 def _get_subtitles(self, video_id, webpage):
995 subs_doc = self._download_xml(
996 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
997 video_id, note=False)
998 except ExtractorError as err:
999 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1003 for track in subs_doc.findall('track'):
1004 lang = track.attrib['lang_code']
1005 if lang in sub_lang_list:
1008 for ext in self._SUBTITLE_FORMATS:
1009 params = compat_urllib_parse.urlencode({
1013 'name': track.attrib['name'].encode('utf-8'),
1015 sub_formats.append({
1016 'url': 'https://www.youtube.com/api/timedtext?' + params,
1019 sub_lang_list[lang] = sub_formats
1020 if not sub_lang_list:
1021 self._downloader.report_warning('video doesn\'t have subtitles')
1023 return sub_lang_list
1025 def _get_ytplayer_config(self, video_id, webpage):
1027 # User data may contain arbitrary character sequences that may affect
1028 # JSON extraction with regex, e.g. when '};' is contained the second
1029 # regex won't capture the whole JSON. Yet working around by trying more
1030 # concrete regex first keeping in mind proper quoted string handling
1031 # to be implemented in future that will replace this workaround (see
1032 # https://github.com/rg3/youtube-dl/issues/7468,
1033 # https://github.com/rg3/youtube-dl/pull/7599)
1034 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1035 r';ytplayer\.config\s*=\s*({.+?});',
1037 config = self._search_regex(
1038 patterns, webpage, 'ytplayer.config', default=None)
1040 return self._parse_json(
1041 uppercase_escape(config), video_id, fatal=False)
1043 def _get_automatic_captions(self, video_id, webpage):
1044 """We need the webpage for getting the captions url, pass it as an
1045 argument to speed up the process."""
1046 self.to_screen('%s: Looking for automatic captions' % video_id)
1047 player_config = self._get_ytplayer_config(video_id, webpage)
1048 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1049 if not player_config:
1050 self._downloader.report_warning(err_msg)
1053 args = player_config['args']
1054 caption_url = args.get('ttsurl')
1056 timestamp = args['timestamp']
1057 # We get the available subtitles
1058 list_params = compat_urllib_parse.urlencode({
1063 list_url = caption_url + '&' + list_params
1064 caption_list = self._download_xml(list_url, video_id)
1065 original_lang_node = caption_list.find('track')
1066 if original_lang_node is None:
1067 self._downloader.report_warning('Video doesn\'t have automatic captions')
1069 original_lang = original_lang_node.attrib['lang_code']
1070 caption_kind = original_lang_node.attrib.get('kind', '')
1073 for lang_node in caption_list.findall('target'):
1074 sub_lang = lang_node.attrib['lang_code']
1076 for ext in self._SUBTITLE_FORMATS:
1077 params = compat_urllib_parse.urlencode({
1078 'lang': original_lang,
1082 'kind': caption_kind,
1084 sub_formats.append({
1085 'url': caption_url + '&' + params,
1088 sub_lang_list[sub_lang] = sub_formats
1089 return sub_lang_list
1091 # Some videos don't provide ttsurl but rather caption_tracks and
1092 # caption_translation_languages (e.g. 20LmZk1hakA)
1093 caption_tracks = args['caption_tracks']
1094 caption_translation_languages = args['caption_translation_languages']
1095 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1096 parsed_caption_url = compat_urlparse.urlparse(caption_url)
1097 caption_qs = compat_parse_qs(parsed_caption_url.query)
1100 for lang in caption_translation_languages.split(','):
1101 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1102 sub_lang = lang_qs.get('lc', [None])[0]
1106 for ext in self._SUBTITLE_FORMATS:
1108 'tlang': [sub_lang],
1111 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1112 query=compat_urllib_parse.urlencode(caption_qs, True)))
1113 sub_formats.append({
1117 sub_lang_list[sub_lang] = sub_formats
1118 return sub_lang_list
1119 # An extractor error can be raise by the download process if there are
1120 # no automatic captions but there are subtitles
1121 except (KeyError, ExtractorError):
1122 self._downloader.report_warning(err_msg)
1125 def _mark_watched(self, video_id, video_info):
1126 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1127 if not playback_url:
1129 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1130 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1132 # cpn generation algorithm is reverse engineered from base.js.
1133 # In fact it works even with dummy cpn.
1134 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1135 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1141 playback_url = compat_urlparse.urlunparse(
1142 parsed_playback_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
1144 self._download_webpage(
1145 playback_url, video_id, 'Marking watched',
1146 'Unable to mark watched', fatal=False)
1149 def extract_id(cls, url):
1150 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1152 raise ExtractorError('Invalid URL: %s' % url)
1153 video_id = mobj.group(2)
1156 def _extract_from_m3u8(self, manifest_url, video_id):
1159 def _get_urls(_manifest):
1160 lines = _manifest.split('\n')
1161 urls = filter(lambda l: l and not l.startswith('#'),
1164 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1165 formats_urls = _get_urls(manifest)
1166 for format_url in formats_urls:
1167 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1168 url_map[itag] = format_url
1171 def _extract_annotations(self, video_id):
1172 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1173 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1175 def _real_extract(self, url):
1176 url, smuggled_data = unsmuggle_url(url, {})
1179 'http' if self._downloader.params.get('prefer_insecure', False)
1184 parsed_url = compat_urllib_parse_urlparse(url)
1185 for component in [parsed_url.fragment, parsed_url.query]:
1186 query = compat_parse_qs(component)
1187 if start_time is None and 't' in query:
1188 start_time = parse_duration(query['t'][0])
1189 if start_time is None and 'start' in query:
1190 start_time = parse_duration(query['start'][0])
1191 if end_time is None and 'end' in query:
1192 end_time = parse_duration(query['end'][0])
1194 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1195 mobj = re.search(self._NEXT_URL_RE, url)
1197 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1198 video_id = self.extract_id(url)
1201 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1202 video_webpage = self._download_webpage(url, video_id)
1204 # Attempt to extract SWF player URL
1205 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1206 if mobj is not None:
1207 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1213 def add_dash_mpd(video_info):
1214 dash_mpd = video_info.get('dashmpd')
1215 if dash_mpd and dash_mpd[0] not in dash_mpds:
1216 dash_mpds.append(dash_mpd[0])
1219 embed_webpage = None
1221 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1223 # We simulate the access to the video from www.youtube.com/v/{video_id}
1224 # this can be viewed without login into Youtube
1225 url = proto + '://www.youtube.com/embed/%s' % video_id
1226 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1227 data = compat_urllib_parse.urlencode({
1228 'video_id': video_id,
1229 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1230 'sts': self._search_regex(
1231 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1233 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1234 video_info_webpage = self._download_webpage(
1235 video_info_url, video_id,
1236 note='Refetching age-gated info webpage',
1237 errnote='unable to download video info webpage')
1238 video_info = compat_parse_qs(video_info_webpage)
1239 add_dash_mpd(video_info)
1243 # Try looking directly into the video webpage
1244 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1246 args = ytplayer_config['args']
1247 if args.get('url_encoded_fmt_stream_map'):
1248 # Convert to the same format returned by compat_parse_qs
1249 video_info = dict((k, [v]) for k, v in args.items())
1250 add_dash_mpd(video_info)
1251 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1253 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1254 # We also try looking in get_video_info since it may contain different dashmpd
1255 # URL that points to a DASH manifest with possibly different itag set (some itags
1256 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1257 # manifest pointed by get_video_info's dashmpd).
1258 # The general idea is to take a union of itags of both DASH manifests (for example
1259 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1260 self.report_video_info_webpage_download(video_id)
1261 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1263 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1264 % (proto, video_id, el_type))
1265 video_info_webpage = self._download_webpage(
1267 video_id, note=False,
1268 errnote='unable to download video info webpage')
1269 get_video_info = compat_parse_qs(video_info_webpage)
1270 if get_video_info.get('use_cipher_signature') != ['True']:
1271 add_dash_mpd(get_video_info)
1273 video_info = get_video_info
1274 if 'token' in get_video_info:
1275 # Different get_video_info requests may report different results, e.g.
1276 # some may report video unavailability, but some may serve it without
1277 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1278 # the original webpage as well as el=info and el=embedded get_video_info
1279 # requests report video unavailability due to geo restriction while
1280 # el=detailpage succeeds and returns valid data). This is probably
1281 # due to YouTube measures against IP ranges of hosting providers.
1282 # Working around by preferring the first succeeded video_info containing
1283 # the token if no such video_info yet was found.
1284 if 'token' not in video_info:
1285 video_info = get_video_info
1287 if 'token' not in video_info:
1288 if 'reason' in video_info:
1289 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1290 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1292 raise ExtractorError('YouTube said: This video is available in %s only' % (
1293 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1295 raise ExtractorError(
1296 'YouTube said: %s' % video_info['reason'][0],
1297 expected=True, video_id=video_id)
1299 raise ExtractorError(
1300 '"token" parameter not in video info for unknown reason',
1304 if 'title' in video_info:
1305 video_title = video_info['title'][0]
1307 self._downloader.report_warning('Unable to extract video title')
1311 video_description = get_element_by_id("eow-description", video_webpage)
1312 if video_description:
1313 video_description = re.sub(r'''(?x)
1315 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1316 (?:title|href)="([^"]+)"\s+
1317 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1318 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1321 ''', r'\1', video_description)
1322 video_description = clean_html(video_description)
1324 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1326 video_description = unescapeHTML(fd_mobj.group(1))
1328 video_description = ''
1330 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1331 if not self._downloader.params.get('noplaylist'):
1334 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1335 for feed in multifeed_metadata_list.split(','):
1336 # Unquote should take place before split on comma (,) since textual
1337 # fields may contain comma as well (see
1338 # https://github.com/rg3/youtube-dl/issues/8536)
1339 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1341 '_type': 'url_transparent',
1342 'ie_key': 'Youtube',
1344 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1345 {'force_singlefeed': True}),
1346 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1348 feed_ids.append(feed_data['id'][0])
1350 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1351 % (', '.join(feed_ids), video_id))
1352 return self.playlist_result(entries, video_id, video_title, video_description)
1353 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1355 if 'view_count' in video_info:
1356 view_count = int(video_info['view_count'][0])
1360 # Check for "rental" videos
1361 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1362 raise ExtractorError('"rental" videos not supported')
1364 # Start extracting information
1365 self.report_information_extraction(video_id)
1368 if 'author' not in video_info:
1369 raise ExtractorError('Unable to extract uploader name')
1370 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1373 video_uploader_id = None
1374 video_uploader_url = None
1376 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1378 if mobj is not None:
1379 video_uploader_id = mobj.group('uploader_id')
1380 video_uploader_url = mobj.group('uploader_url')
1382 self._downloader.report_warning('unable to extract uploader nickname')
1385 # We try first to get a high quality image:
1386 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1387 video_webpage, re.DOTALL)
1388 if m_thumb is not None:
1389 video_thumbnail = m_thumb.group(1)
1390 elif 'thumbnail_url' not in video_info:
1391 self._downloader.report_warning('unable to extract video thumbnail')
1392 video_thumbnail = None
1393 else: # don't panic if we can't find it
1394 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1397 upload_date = self._html_search_meta(
1398 'datePublished', video_webpage, 'upload date', default=None)
1400 upload_date = self._search_regex(
1401 [r'(?s)id="eow-date.*?>(.*?)</span>',
1402 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1403 video_webpage, 'upload date', default=None)
1405 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1406 upload_date = unified_strdate(upload_date)
1408 video_license = self._html_search_regex(
1409 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1410 video_webpage, 'license', default=None)
1412 m_music = re.search(
1413 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1416 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1417 video_creator = clean_html(m_music.group('creator'))
1419 video_alt_title = video_creator = None
1421 m_cat_container = self._search_regex(
1422 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1423 video_webpage, 'categories', default=None)
1425 category = self._html_search_regex(
1426 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1428 video_categories = None if category is None else [category]
1430 video_categories = None
1433 unescapeHTML(m.group('content'))
1434 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1436 def _extract_count(count_name):
1437 return str_to_int(self._search_regex(
1438 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1439 % re.escape(count_name),
1440 video_webpage, count_name, default=None))
1442 like_count = _extract_count('like')
1443 dislike_count = _extract_count('dislike')
1446 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1447 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1449 if 'length_seconds' not in video_info:
1450 self._downloader.report_warning('unable to extract video duration')
1451 video_duration = None
1453 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1456 video_annotations = None
1457 if self._downloader.params.get('writeannotations', False):
1458 video_annotations = self._extract_annotations(video_id)
1460 def _map_to_format_list(urlmap):
1462 for itag, video_real_url in urlmap.items():
1465 'url': video_real_url,
1466 'player_url': player_url,
1468 if itag in self._formats:
1469 dct.update(self._formats[itag])
1473 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1474 self.report_rtmp_download()
1476 'format_id': '_rtmp',
1478 'url': video_info['conn'][0],
1479 'player_url': player_url,
1481 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1482 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1483 if 'rtmpe%3Dyes' in encoded_url_map:
1484 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1486 fmt_list = video_info.get('fmt_list', [''])[0]
1488 for fmt in fmt_list.split(','):
1489 spec = fmt.split('/')
1491 width_height = spec[1].split('x')
1492 if len(width_height) == 2:
1493 formats_spec[spec[0]] = {
1494 'resolution': spec[1],
1495 'width': int_or_none(width_height[0]),
1496 'height': int_or_none(width_height[1]),
1499 for url_data_str in encoded_url_map.split(','):
1500 url_data = compat_parse_qs(url_data_str)
1501 if 'itag' not in url_data or 'url' not in url_data:
1503 format_id = url_data['itag'][0]
1504 url = url_data['url'][0]
1506 if 'sig' in url_data:
1507 url += '&signature=' + url_data['sig'][0]
1508 elif 's' in url_data:
1509 encrypted_sig = url_data['s'][0]
1510 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1512 jsplayer_url_json = self._search_regex(
1514 embed_webpage if age_gate else video_webpage,
1515 'JS player URL (1)', default=None)
1516 if not jsplayer_url_json and not age_gate:
1517 # We need the embed website after all
1518 if embed_webpage is None:
1519 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1520 embed_webpage = self._download_webpage(
1521 embed_url, video_id, 'Downloading embed webpage')
1522 jsplayer_url_json = self._search_regex(
1523 ASSETS_RE, embed_webpage, 'JS player URL')
1525 player_url = json.loads(jsplayer_url_json)
1526 if player_url is None:
1527 player_url_json = self._search_regex(
1528 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1529 video_webpage, 'age gate player URL')
1530 player_url = json.loads(player_url_json)
1532 if self._downloader.params.get('verbose'):
1533 if player_url is None:
1534 player_version = 'unknown'
1535 player_desc = 'unknown'
1537 if player_url.endswith('swf'):
1538 player_version = self._search_regex(
1539 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1540 'flash player', fatal=False)
1541 player_desc = 'flash player %s' % player_version
1543 player_version = self._search_regex(
1544 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1546 'html5 player', fatal=False)
1547 player_desc = 'html5 player %s' % player_version
1549 parts_sizes = self._signature_cache_id(encrypted_sig)
1550 self.to_screen('{%s} signature length %s, %s' %
1551 (format_id, parts_sizes, player_desc))
1553 signature = self._decrypt_signature(
1554 encrypted_sig, video_id, player_url, age_gate)
1555 url += '&signature=' + signature
1556 if 'ratebypass' not in url:
1557 url += '&ratebypass=yes'
1560 'format_id': format_id,
1562 'player_url': player_url,
1564 if format_id in self._formats:
1565 dct.update(self._formats[format_id])
1566 if format_id in formats_spec:
1567 dct.update(formats_spec[format_id])
1569 # Some itags are not included in DASH manifest thus corresponding formats will
1570 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1571 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1572 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1573 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1576 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1577 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1580 'fps': int_or_none(url_data.get('fps', [None])[0]),
1581 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1583 for key, value in more_fields.items():
1586 type_ = url_data.get('type', [None])[0]
1588 type_split = type_.split(';')
1589 kind_ext = type_split[0].split('/')
1590 if len(kind_ext) == 2:
1592 dct['ext'] = mimetype2ext(type_split[0])
1593 if kind in ('audio', 'video'):
1595 for mobj in re.finditer(
1596 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1597 if mobj.group('key') == 'codecs':
1598 codecs = mobj.group('val')
1601 codecs = codecs.split(',')
1602 if len(codecs) == 2:
1603 acodec, vcodec = codecs[1], codecs[0]
1605 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1611 elif video_info.get('hlsvp'):
1612 manifest_url = video_info['hlsvp'][0]
1613 url_map = self._extract_from_m3u8(manifest_url, video_id)
1614 formats = _map_to_format_list(url_map)
1615 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1616 for a_format in formats:
1617 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1619 unavailable_message = self._html_search_regex(
1620 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1621 video_webpage, 'unavailable message', default=None)
1622 if unavailable_message:
1623 raise ExtractorError(unavailable_message, expected=True)
1624 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1626 # Look for the DASH manifest
1627 if self._downloader.params.get('youtube_include_dash_manifest', True):
1628 dash_mpd_fatal = True
1629 for mpd_url in dash_mpds:
1632 def decrypt_sig(mobj):
1634 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1635 return '/signature/%s' % dec_s
1637 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1639 for df in self._extract_mpd_formats(
1640 mpd_url, video_id, fatal=dash_mpd_fatal,
1641 formats_dict=self._formats):
1642 # Do not overwrite DASH format found in some previous DASH manifest
1643 if df['format_id'] not in dash_formats:
1644 dash_formats[df['format_id']] = df
1645 # Additional DASH manifests may end up in HTTP Error 403 therefore
1646 # allow them to fail without bug report message if we already have
1647 # some DASH manifest succeeded. This is temporary workaround to reduce
1648 # burst of bug reports until we figure out the reason and whether it
1649 # can be fixed at all.
1650 dash_mpd_fatal = False
1651 except (ExtractorError, KeyError) as e:
1652 self.report_warning(
1653 'Skipping DASH manifest: %r' % e, video_id)
1655 # Remove the formats we found through non-DASH, they
1656 # contain less info and it can be wrong, because we use
1657 # fixed values (for example the resolution). See
1658 # https://github.com/rg3/youtube-dl/issues/5774 for an
1660 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1661 formats.extend(dash_formats.values())
1663 # Check for malformed aspect ratio
1664 stretched_m = re.search(
1665 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1668 w = float(stretched_m.group('w'))
1669 h = float(stretched_m.group('h'))
1670 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1671 # We will only process correct ratios.
1675 if f.get('vcodec') != 'none':
1676 f['stretched_ratio'] = ratio
1678 self._sort_formats(formats)
1680 self.mark_watched(video_id, video_info)
1684 'uploader': video_uploader,
1685 'uploader_id': video_uploader_id,
1686 'uploader_url': video_uploader_url,
1687 'upload_date': upload_date,
1688 'license': video_license,
1689 'creator': video_creator,
1690 'title': video_title,
1691 'alt_title': video_alt_title,
1692 'thumbnail': video_thumbnail,
1693 'description': video_description,
1694 'categories': video_categories,
1696 'subtitles': video_subtitles,
1697 'automatic_captions': automatic_captions,
1698 'duration': video_duration,
1699 'age_limit': 18 if age_gate else 0,
1700 'annotations': video_annotations,
1701 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1702 'view_count': view_count,
1703 'like_count': like_count,
1704 'dislike_count': dislike_count,
1705 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1708 'start_time': start_time,
1709 'end_time': end_time,
1713 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1714 IE_DESC = 'YouTube.com playlists'
1715 _VALID_URL = r"""(?x)(?:
1720 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1721 \? (?:.*?[&;])*? (?:p|a|list)=
1725 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1726 # Top tracks, they can also include dots
1731 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1733 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1734 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1735 IE_NAME = 'youtube:playlist'
1737 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1739 'title': 'ytdl test PL',
1740 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1742 'playlist_count': 3,
1744 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1746 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1747 'title': 'YDL_Empty_List',
1749 'playlist_count': 0,
1751 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1752 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1754 'title': '29C3: Not my department',
1755 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1757 'playlist_count': 95,
1759 'note': 'issue #673',
1760 'url': 'PLBB231211A4F62143',
1762 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1763 'id': 'PLBB231211A4F62143',
1765 'playlist_mincount': 26,
1767 'note': 'Large playlist',
1768 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1770 'title': 'Uploads from Cauchemar',
1771 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1773 'playlist_mincount': 799,
1775 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1777 'title': 'YDL_safe_search',
1778 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1780 'playlist_count': 2,
1783 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1784 'playlist_count': 4,
1787 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1790 'note': 'Embedded SWF player',
1791 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1792 'playlist_count': 4,
1795 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1798 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1799 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1801 'title': 'Uploads from Interstellar Movie',
1802 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1804 'playlist_mincout': 21,
1807 def _real_initialize(self):
1810 def _extract_mix(self, playlist_id):
1811 # The mixes are generated from a single video
1812 # the id of the playlist is just 'RD' + video_id
1813 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1814 webpage = self._download_webpage(
1815 url, playlist_id, 'Downloading Youtube mix')
1816 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1818 search_title('playlist-title') or
1819 search_title('title long-title') or
1820 search_title('title'))
1821 title = clean_html(title_span)
1822 ids = orderedSet(re.findall(
1823 r'''(?xs)data-video-username=".*?".*?
1824 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1826 url_results = self._ids_to_results(ids)
1828 return self.playlist_result(url_results, playlist_id, title)
1830 def _extract_playlist(self, playlist_id):
1831 url = self._TEMPLATE_URL % playlist_id
1832 page = self._download_webpage(url, playlist_id)
1834 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1835 match = match.strip()
1836 # Check if the playlist exists or is private
1837 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1838 raise ExtractorError(
1839 'The playlist doesn\'t exist or is private, use --username or '
1840 '--netrc to access it.',
1842 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1843 raise ExtractorError(
1844 'Invalid parameters. Maybe URL is incorrect.',
1846 elif re.match(r'[^<]*Choose your language[^<]*', match):
1849 self.report_warning('Youtube gives an alert message: ' + match)
1851 playlist_title = self._html_search_regex(
1852 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1855 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1857 def _check_download_just_video(self, url, playlist_id):
1858 # Check if it's a video-specific URL
1859 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1860 if 'v' in query_dict:
1861 video_id = query_dict['v'][0]
1862 if self._downloader.params.get('noplaylist'):
1863 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1864 return self.url_result(video_id, 'Youtube', video_id=video_id)
1866 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1868 def _real_extract(self, url):
1869 # Extract playlist id
1870 mobj = re.match(self._VALID_URL, url)
1872 raise ExtractorError('Invalid URL: %s' % url)
1873 playlist_id = mobj.group(1) or mobj.group(2)
1875 video = self._check_download_just_video(url, playlist_id)
1879 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1880 # Mixes require a custom extraction process
1881 return self._extract_mix(playlist_id)
1883 return self._extract_playlist(playlist_id)
1886 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1887 IE_DESC = 'YouTube.com channels'
1888 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1889 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1890 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1891 IE_NAME = 'youtube:channel'
1893 'note': 'paginated channel',
1894 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1895 'playlist_mincount': 91,
1897 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1898 'title': 'Uploads from lex will',
1901 'note': 'Age restricted channel',
1902 # from https://www.youtube.com/user/DeusExOfficial
1903 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1904 'playlist_mincount': 64,
1906 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1907 'title': 'Uploads from Deus Ex',
1912 def suitable(cls, url):
1913 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1915 def _real_extract(self, url):
1916 channel_id = self._match_id(url)
1918 url = self._TEMPLATE_URL % channel_id
1920 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1921 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1922 # otherwise fallback on channel by page extraction
1923 channel_page = self._download_webpage(
1924 url + '?view=57', channel_id,
1925 'Downloading channel page', fatal=False)
1926 if channel_page is False:
1927 channel_playlist_id = False
1929 channel_playlist_id = self._html_search_meta(
1930 'channelId', channel_page, 'channel id', default=None)
1931 if not channel_playlist_id:
1932 channel_playlist_id = self._search_regex(
1933 r'data-(?:channel-external-|yt)id="([^"]+)"',
1934 channel_page, 'channel id', default=None)
1935 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1936 playlist_id = 'UU' + channel_playlist_id[2:]
1937 return self.url_result(
1938 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1940 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1941 autogenerated = re.search(r'''(?x)
1943 channel-header-autogenerated-label|
1944 yt-channel-title-autogenerated
1945 )[^"]*"''', channel_page) is not None
1948 # The videos are contained in a single page
1949 # the ajax pages can't be used, they are empty
1952 video_id, 'Youtube', video_id=video_id,
1953 video_title=video_title)
1954 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1955 return self.playlist_result(entries, channel_id)
1957 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1960 class YoutubeUserIE(YoutubeChannelIE):
1961 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1962 _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_-]+)'
1963 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1964 IE_NAME = 'youtube:user'
1967 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1968 'playlist_mincount': 320,
1970 'title': 'TheLinuxFoundation',
1973 'url': 'ytuser:phihag',
1974 'only_matching': True,
1978 def suitable(cls, url):
1979 # Don't return True if the url can be extracted with other youtube
1980 # extractor, the regex would is too permissive and it would match.
1981 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1982 if any(ie.suitable(url) for ie in other_ies):
1985 return super(YoutubeUserIE, cls).suitable(url)
1988 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1989 IE_DESC = 'YouTube.com user/channel playlists'
1990 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1991 IE_NAME = 'youtube:playlists'
1994 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1995 'playlist_mincount': 4,
1997 'id': 'ThirstForScience',
1998 'title': 'Thirst for Science',
2001 # with "Load more" button
2002 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2003 'playlist_mincount': 70,
2006 'title': 'Игорь Клейнер',
2009 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2010 'playlist_mincount': 17,
2012 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2013 'title': 'Chem Player',
2018 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2019 IE_DESC = 'YouTube.com searches'
2020 # there doesn't appear to be a real limit, for example if you search for
2021 # 'python' you get more than 8.000.000 results
2022 _MAX_RESULTS = float('inf')
2023 IE_NAME = 'youtube:search'
2024 _SEARCH_KEY = 'ytsearch'
2025 _EXTRA_QUERY_ARGS = {}
2028 def _get_n_results(self, query, n):
2029 """Get a specified number of results for a query"""
2034 for pagenum in itertools.count(1):
2036 'search_query': query.encode('utf-8'),
2040 url_query.update(self._EXTRA_QUERY_ARGS)
2041 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
2042 data = self._download_json(
2043 result_url, video_id='query "%s"' % query,
2044 note='Downloading page %s' % pagenum,
2045 errnote='Unable to download API page')
2046 html_content = data[1]['body']['content']
2048 if 'class="search-message' in html_content:
2049 raise ExtractorError(
2050 '[youtube] No video results', expected=True)
2052 new_videos = self._ids_to_results(orderedSet(re.findall(
2053 r'href="/watch\?v=(.{11})', html_content)))
2054 videos += new_videos
2055 if not new_videos or len(videos) > limit:
2060 return self.playlist_result(videos, query)
2063 class YoutubeSearchDateIE(YoutubeSearchIE):
2064 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2065 _SEARCH_KEY = 'ytsearchdate'
2066 IE_DESC = 'YouTube.com searches, newest videos first'
2067 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2070 class YoutubeSearchURLIE(InfoExtractor):
2071 IE_DESC = 'YouTube.com search URLs'
2072 IE_NAME = 'youtube:search_url'
2073 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2075 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2076 'playlist_mincount': 5,
2078 'title': 'youtube-dl test video',
2081 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2082 'only_matching': True,
2085 def _real_extract(self, url):
2086 mobj = re.match(self._VALID_URL, url)
2087 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2089 webpage = self._download_webpage(url, query)
2090 result_code = self._search_regex(
2091 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2093 part_codes = re.findall(
2094 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2096 for part_code in part_codes:
2097 part_title = self._html_search_regex(
2098 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2099 part_url_snippet = self._html_search_regex(
2100 r'(?s)href="([^"]+)"', part_code, 'item URL')
2101 part_url = compat_urlparse.urljoin(
2102 'https://www.youtube.com/', part_url_snippet)
2106 'title': part_title,
2110 '_type': 'playlist',
2116 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2117 IE_DESC = 'YouTube.com (multi-season) shows'
2118 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2119 IE_NAME = 'youtube:show'
2121 'url': 'https://www.youtube.com/show/airdisasters',
2122 'playlist_mincount': 5,
2124 'id': 'airdisasters',
2125 'title': 'Air Disasters',
2129 def _real_extract(self, url):
2130 playlist_id = self._match_id(url)
2131 return super(YoutubeShowIE, self)._real_extract(
2132 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2135 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2137 Base class for feed extractors
2138 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2140 _LOGIN_REQUIRED = True
2144 return 'youtube:%s' % self._FEED_NAME
2146 def _real_initialize(self):
2149 def _real_extract(self, url):
2150 page = self._download_webpage(
2151 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2153 # The extraction process is the same as for playlists, but the regex
2154 # for the video ids doesn't contain an index
2156 more_widget_html = content_html = page
2157 for page_num in itertools.count(1):
2158 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2160 # 'recommended' feed has infinite 'load more' and each new portion spins
2161 # the same videos in (sometimes) slightly different order, so we'll check
2162 # for unicity and break when portion has no new videos
2163 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2169 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2173 more = self._download_json(
2174 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2175 'Downloading page #%s' % page_num,
2176 transform_source=uppercase_escape)
2177 content_html = more['content_html']
2178 more_widget_html = more['load_more_widget_html']
2180 return self.playlist_result(
2181 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2184 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2185 IE_NAME = 'youtube:watchlater'
2186 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2187 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2190 'url': 'https://www.youtube.com/playlist?list=WL',
2191 'only_matching': True,
2193 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2194 'only_matching': True,
2197 def _real_extract(self, url):
2198 video = self._check_download_just_video(url, 'WL')
2201 return self._extract_playlist('WL')
2204 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2205 IE_NAME = 'youtube:favorites'
2206 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2207 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2208 _LOGIN_REQUIRED = True
2210 def _real_extract(self, url):
2211 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2212 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2213 return self.url_result(playlist_id, 'YoutubePlaylist')
2216 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2217 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2218 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2219 _FEED_NAME = 'recommended'
2220 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2223 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2224 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2225 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2226 _FEED_NAME = 'subscriptions'
2227 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2230 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2231 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2232 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2233 _FEED_NAME = 'history'
2234 _PLAYLIST_TITLE = 'Youtube History'
2237 class YoutubeTruncatedURLIE(InfoExtractor):
2238 IE_NAME = 'youtube:truncated_url'
2239 IE_DESC = False # Do not list
2240 _VALID_URL = r'''(?x)
2242 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2245 annotation_id=annotation_[^&]+|
2251 attribution_link\?a=[^&]+
2257 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2258 'only_matching': True,
2260 'url': 'http://www.youtube.com/watch?',
2261 'only_matching': True,
2263 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2264 'only_matching': True,
2266 'url': 'https://www.youtube.com/watch?feature=foo',
2267 'only_matching': True,
2269 'url': 'https://www.youtube.com/watch?hl=en-GB',
2270 'only_matching': True,
2272 'url': 'https://www.youtube.com/watch?t=2372',
2273 'only_matching': True,
2276 def _real_extract(self, url):
2277 raise ExtractorError(
2278 'Did you forget to quote the URL? Remember that & is a meta '
2279 'character in most shells, so you want to put the URL in quotes, '
2281 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2282 ' or simply youtube-dl BaW_jenozKc .',
2286 class YoutubeTruncatedIDIE(InfoExtractor):
2287 IE_NAME = 'youtube:truncated_id'
2288 IE_DESC = False # Do not list
2289 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2292 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2293 'only_matching': True,
2296 def _real_extract(self, url):
2297 video_id = self._match_id(url)
2298 raise ExtractorError(
2299 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),