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=([^&]+)'
281 # tbr was extracted from com/google/youtube/model/VideoFormat.as in watch_as3.swf and converted from Bytes/S to KBits/S
283 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263', 'tbr': 320},
284 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263', 'tbr': 896},
285 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v', 'tbr': 60},
286 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v', 'tbr': 80},
287 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264', 'tbr': 736},
288 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'tbr': 3192},
289 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'tbr': 928},
290 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'tbr': 1280},
291 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
292 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v', 'tbr': 256},
293 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'tbr': 6192},
294 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'tbr': 10128},
295 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'tbr': 928},
296 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'tbr': 1280},
297 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'tbr': 3192},
298 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
299 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'tbr': 1280},
300 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'tbr': 1280},
304 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20, 'tbr': 800},
305 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20, 'tbr': 1152},
306 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20, 'tbr': 3000},
307 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20, 'tbr': 6000},
308 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
309 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
310 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
312 # Apple HTTP Live Streaming
313 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10, 'tbr': 186.625},
314 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10, 'tbr': 951.5625},
315 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10, 'tbr': 1312.5},
316 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10, 'tbr': 3207.421875},
317 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10, 'tbr': 6349.21875},
318 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
319 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
322 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 261.71875},
323 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 951.5625},
324 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 1312.5},
325 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 3207.421875},
326 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 6349.21875},
327 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 10128.0}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
328 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40, 'tbr': 91.796875},
329 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
330 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
331 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
332 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
335 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash', 'tbr': 32},
336 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash', 'tbr': 128},
337 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash', 'tbr': 320},
340 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
344 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
345 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
346 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
347 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
350 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
354 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
355 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
356 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
357 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
358 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
359 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
360 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
361 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
364 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
365 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
367 # Dash webm audio with opus inside
368 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
369 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
370 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
373 '_rtmp': {'protocol': 'rtmp'},
375 _SUBTITLE_FORMATS = ('ttml', 'vtt')
380 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
384 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
385 'uploader': 'Philipp Hagemeister',
386 'uploader_id': 'phihag',
387 'upload_date': '20121002',
388 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
389 'categories': ['Science & Technology'],
390 'tags': ['youtube-dl'],
392 'dislike_count': int,
398 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
399 'note': 'Test generic use_cipher_signature video (#897)',
403 'upload_date': '20120506',
404 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
405 'alt_title': 'I Love It (feat. Charli XCX)',
406 'description': 'md5:782e8651347686cba06e58f71ab51773',
407 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
408 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
409 'iconic ep', 'iconic', 'love', 'it'],
410 'uploader': 'Icona Pop',
411 'uploader_id': 'IconaPop',
412 'creator': 'Icona Pop',
416 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
417 'note': 'Test VEVO video with age protection (#956)',
421 'upload_date': '20130703',
422 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
423 'alt_title': 'Tunnel Vision',
424 'description': 'md5:64249768eec3bc4276236606ea996373',
425 'uploader': 'justintimberlakeVEVO',
426 'uploader_id': 'justintimberlakeVEVO',
427 'creator': 'Justin Timberlake',
432 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
433 'note': 'Embed-only video (#1746)',
437 'upload_date': '20120608',
438 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
439 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
440 'uploader': 'SET India',
441 'uploader_id': 'setindia',
446 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
447 'note': 'Use the first video ID in the URL',
451 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
452 'uploader': 'Philipp Hagemeister',
453 'uploader_id': 'phihag',
454 'upload_date': '20121002',
455 '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 .',
456 'categories': ['Science & Technology'],
457 'tags': ['youtube-dl'],
459 'dislike_count': int,
462 'skip_download': True,
466 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
467 'note': '256k DASH audio (format 141) via DASH manifest',
471 'upload_date': '20121002',
472 'uploader_id': '8KVIDEO',
474 'uploader': '8KVIDEO',
475 'title': 'UHDTV TEST 8K VIDEO.mp4'
478 'youtube_include_dash_manifest': True,
482 # DASH manifest with encrypted signature
484 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
488 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
489 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
490 'uploader': 'AfrojackVEVO',
491 'uploader_id': 'AfrojackVEVO',
492 'upload_date': '20131011',
495 'youtube_include_dash_manifest': True,
499 # JS player signature function name containing $
501 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
505 'title': 'Taylor Swift - Shake It Off',
506 'alt_title': 'Shake It Off',
507 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
508 'uploader': 'TaylorSwiftVEVO',
509 'uploader_id': 'TaylorSwiftVEVO',
510 'upload_date': '20140818',
511 'creator': 'Taylor Swift',
514 'youtube_include_dash_manifest': True,
520 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
524 'upload_date': '20100909',
525 'uploader': 'The Amazing Atheist',
526 'uploader_id': 'TheAmazingAtheist',
527 'title': 'Burning Everyone\'s Koran',
528 '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',
531 # Normal age-gate video (No vevo, embed allowed)
533 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
537 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
538 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
539 'uploader': 'The Witcher',
540 'uploader_id': 'WitcherGame',
541 'upload_date': '20140605',
545 # Age-gate video with encrypted signature
547 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
551 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
552 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
553 'uploader': 'LloydVEVO',
554 'uploader_id': 'LloydVEVO',
555 'upload_date': '20110629',
559 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
561 'url': '__2ABJjxzNo',
565 'upload_date': '20100430',
566 'uploader_id': 'deadmau5',
567 'creator': 'deadmau5',
568 'description': 'md5:12c56784b8032162bb936a5f76d55360',
569 'uploader': 'deadmau5',
570 'title': 'Deadmau5 - Some Chords (HD)',
571 'alt_title': 'Some Chords',
573 'expected_warnings': [
574 'DASH manifest missing',
577 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
579 'url': 'lqQg6PlCWgI',
583 'upload_date': '20150827',
584 'uploader_id': 'olympic',
585 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
586 'uploader': 'Olympics',
587 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
590 'skip_download': 'requires avconv',
595 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
599 'stretched_ratio': 16 / 9.,
600 'upload_date': '20110310',
601 'uploader_id': 'AllenMeow',
602 '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',
604 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
607 # url_encoded_fmt_stream_map is empty string
609 'url': 'qEJwOuvDf7I',
613 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
615 'upload_date': '20150404',
616 'uploader_id': 'spbelect',
617 'uploader': 'Наблюдатели Петербурга',
620 'skip_download': 'requires avconv',
622 'skip': 'This live event has ended.',
624 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
626 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
630 'title': 'md5:7b81415841e02ecd4313668cde88737a',
631 'description': 'md5:116377fd2963b81ec4ce64b542173306',
632 'upload_date': '20150625',
633 'uploader_id': 'dorappi2000',
634 'uploader': 'dorappi2000',
635 'formats': 'mincount:33',
638 # DASH manifest with segment_list
640 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
641 'md5': '8ce563a1d667b599d21064e982ab9e31',
645 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
646 'uploader': 'Airtek',
647 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
648 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
649 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
652 'youtube_include_dash_manifest': True,
653 'format': '135', # bestvideo
657 # Multifeed videos (multiple cameras), URL is for Main Camera
658 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
661 'title': 'teamPGP: Rocket League Noob Stream',
662 'description': 'md5:dc7872fb300e143831327f1bae3af010',
668 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
669 'description': 'md5:dc7872fb300e143831327f1bae3af010',
670 'upload_date': '20150721',
671 'uploader': 'Beer Games Beer',
672 'uploader_id': 'beergamesbeer',
678 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
679 'description': 'md5:dc7872fb300e143831327f1bae3af010',
680 'upload_date': '20150721',
681 'uploader': 'Beer Games Beer',
682 'uploader_id': 'beergamesbeer',
688 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
689 'description': 'md5:dc7872fb300e143831327f1bae3af010',
690 'upload_date': '20150721',
691 'uploader': 'Beer Games Beer',
692 'uploader_id': 'beergamesbeer',
698 'title': 'teamPGP: Rocket League Noob Stream (zim)',
699 'description': 'md5:dc7872fb300e143831327f1bae3af010',
700 'upload_date': '20150721',
701 'uploader': 'Beer Games Beer',
702 'uploader_id': 'beergamesbeer',
706 'skip_download': True,
710 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
711 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
714 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
719 'url': 'http://vid.plus/FlRa-iH7PGw',
720 'only_matching': True,
723 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
724 # Also tests cut-off URL expansion in video description (see
725 # https://github.com/rg3/youtube-dl/issues/1892,
726 # https://github.com/rg3/youtube-dl/issues/8164)
727 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
731 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
732 'alt_title': 'Dark Walk',
733 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
734 'upload_date': '20151119',
735 'uploader_id': 'IronSoulElf',
736 'uploader': 'IronSoulElf',
737 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
740 'skip_download': True,
744 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
745 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
746 'only_matching': True,
749 # Video with yt:stretch=17:0
750 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
754 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
755 'description': 'md5:ee18a25c350637c8faff806845bddee9',
756 'upload_date': '20151107',
757 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
758 'uploader': 'CH GAMER DROID',
761 'skip_download': True,
765 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
766 'only_matching': True,
770 def __init__(self, *args, **kwargs):
771 super(YoutubeIE, self).__init__(*args, **kwargs)
772 self._player_cache = {}
774 def report_video_info_webpage_download(self, video_id):
775 """Report attempt to download video info webpage."""
776 self.to_screen('%s: Downloading video info webpage' % video_id)
778 def report_information_extraction(self, video_id):
779 """Report attempt to extract video information."""
780 self.to_screen('%s: Extracting video information' % video_id)
782 def report_unavailable_format(self, video_id, format):
783 """Report extracted video URL."""
784 self.to_screen('%s: Format %s not available' % (video_id, format))
786 def report_rtmp_download(self):
787 """Indicate the download will use the RTMP protocol."""
788 self.to_screen('RTMP download detected')
790 def _signature_cache_id(self, example_sig):
791 """ Return a string representation of a signature """
792 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
794 def _extract_signature_function(self, video_id, player_url, example_sig):
796 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
799 raise ExtractorError('Cannot identify player %r' % player_url)
800 player_type = id_m.group('ext')
801 player_id = id_m.group('id')
803 # Read from filesystem cache
804 func_id = '%s_%s_%s' % (
805 player_type, player_id, self._signature_cache_id(example_sig))
806 assert os.path.basename(func_id) == func_id
808 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
809 if cache_spec is not None:
810 return lambda s: ''.join(s[i] for i in cache_spec)
813 'Downloading player %s' % player_url
814 if self._downloader.params.get('verbose') else
815 'Downloading %s player %s' % (player_type, player_id)
817 if player_type == 'js':
818 code = self._download_webpage(
819 player_url, video_id,
821 errnote='Download of %s failed' % player_url)
822 res = self._parse_sig_js(code)
823 elif player_type == 'swf':
824 urlh = self._request_webpage(
825 player_url, video_id,
827 errnote='Download of %s failed' % player_url)
829 res = self._parse_sig_swf(code)
831 assert False, 'Invalid player type %r' % player_type
833 test_string = ''.join(map(compat_chr, range(len(example_sig))))
834 cache_res = res(test_string)
835 cache_spec = [ord(c) for c in cache_res]
837 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
840 def _print_sig_code(self, func, example_sig):
841 def gen_sig_code(idxs):
842 def _genslice(start, end, step):
843 starts = '' if start == 0 else str(start)
844 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
845 steps = '' if step == 1 else (':%d' % step)
846 return 's[%s%s%s]' % (starts, ends, steps)
849 # Quelch pyflakes warnings - start will be set when step is set
850 start = '(Never used)'
851 for i, prev in zip(idxs[1:], idxs[:-1]):
855 yield _genslice(start, prev, step)
858 if i - prev in [-1, 1]:
867 yield _genslice(start, i, step)
869 test_string = ''.join(map(compat_chr, range(len(example_sig))))
870 cache_res = func(test_string)
871 cache_spec = [ord(c) for c in cache_res]
872 expr_code = ' + '.join(gen_sig_code(cache_spec))
873 signature_id_tuple = '(%s)' % (
874 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
875 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
876 ' return %s\n') % (signature_id_tuple, expr_code)
877 self.to_screen('Extracted signature function:\n' + code)
879 def _parse_sig_js(self, jscode):
880 funcname = self._search_regex(
881 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
882 'Initial JS player signature function name')
884 jsi = JSInterpreter(jscode)
885 initial_function = jsi.extract_function(funcname)
886 return lambda s: initial_function([s])
888 def _parse_sig_swf(self, file_contents):
889 swfi = SWFInterpreter(file_contents)
890 TARGET_CLASSNAME = 'SignatureDecipher'
891 searched_class = swfi.extract_class(TARGET_CLASSNAME)
892 initial_function = swfi.extract_function(searched_class, 'decipher')
893 return lambda s: initial_function([s])
895 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
896 """Turn the encrypted s field into a working signature"""
898 if player_url is None:
899 raise ExtractorError('Cannot decrypt signature without player_url')
901 if player_url.startswith('//'):
902 player_url = 'https:' + player_url
904 player_id = (player_url, self._signature_cache_id(s))
905 if player_id not in self._player_cache:
906 func = self._extract_signature_function(
907 video_id, player_url, s
909 self._player_cache[player_id] = func
910 func = self._player_cache[player_id]
911 if self._downloader.params.get('youtube_print_sig_code'):
912 self._print_sig_code(func, s)
914 except Exception as e:
915 tb = traceback.format_exc()
916 raise ExtractorError(
917 'Signature extraction failed: ' + tb, cause=e)
919 def _get_subtitles(self, video_id, webpage):
921 subs_doc = self._download_xml(
922 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
923 video_id, note=False)
924 except ExtractorError as err:
925 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
929 for track in subs_doc.findall('track'):
930 lang = track.attrib['lang_code']
931 if lang in sub_lang_list:
934 for ext in self._SUBTITLE_FORMATS:
935 params = compat_urllib_parse.urlencode({
939 'name': track.attrib['name'].encode('utf-8'),
942 'url': 'https://www.youtube.com/api/timedtext?' + params,
945 sub_lang_list[lang] = sub_formats
946 if not sub_lang_list:
947 self._downloader.report_warning('video doesn\'t have subtitles')
951 def _get_ytplayer_config(self, video_id, webpage):
953 # User data may contain arbitrary character sequences that may affect
954 # JSON extraction with regex, e.g. when '};' is contained the second
955 # regex won't capture the whole JSON. Yet working around by trying more
956 # concrete regex first keeping in mind proper quoted string handling
957 # to be implemented in future that will replace this workaround (see
958 # https://github.com/rg3/youtube-dl/issues/7468,
959 # https://github.com/rg3/youtube-dl/pull/7599)
960 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
961 r';ytplayer\.config\s*=\s*({.+?});',
963 config = self._search_regex(
964 patterns, webpage, 'ytplayer.config', default=None)
966 return self._parse_json(
967 uppercase_escape(config), video_id, fatal=False)
969 def _get_automatic_captions(self, video_id, webpage):
970 """We need the webpage for getting the captions url, pass it as an
971 argument to speed up the process."""
972 self.to_screen('%s: Looking for automatic captions' % video_id)
973 player_config = self._get_ytplayer_config(video_id, webpage)
974 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
975 if not player_config:
976 self._downloader.report_warning(err_msg)
979 args = player_config['args']
980 caption_url = args.get('ttsurl')
982 timestamp = args['timestamp']
983 # We get the available subtitles
984 list_params = compat_urllib_parse.urlencode({
989 list_url = caption_url + '&' + list_params
990 caption_list = self._download_xml(list_url, video_id)
991 original_lang_node = caption_list.find('track')
992 if original_lang_node is None:
993 self._downloader.report_warning('Video doesn\'t have automatic captions')
995 original_lang = original_lang_node.attrib['lang_code']
996 caption_kind = original_lang_node.attrib.get('kind', '')
999 for lang_node in caption_list.findall('target'):
1000 sub_lang = lang_node.attrib['lang_code']
1002 for ext in self._SUBTITLE_FORMATS:
1003 params = compat_urllib_parse.urlencode({
1004 'lang': original_lang,
1008 'kind': caption_kind,
1010 sub_formats.append({
1011 'url': caption_url + '&' + params,
1014 sub_lang_list[sub_lang] = sub_formats
1015 return sub_lang_list
1017 # Some videos don't provide ttsurl but rather caption_tracks and
1018 # caption_translation_languages (e.g. 20LmZk1hakA)
1019 caption_tracks = args['caption_tracks']
1020 caption_translation_languages = args['caption_translation_languages']
1021 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1022 parsed_caption_url = compat_urlparse.urlparse(caption_url)
1023 caption_qs = compat_parse_qs(parsed_caption_url.query)
1026 for lang in caption_translation_languages.split(','):
1027 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1028 sub_lang = lang_qs.get('lc', [None])[0]
1032 for ext in self._SUBTITLE_FORMATS:
1034 'tlang': [sub_lang],
1037 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1038 query=compat_urllib_parse.urlencode(caption_qs, True)))
1039 sub_formats.append({
1043 sub_lang_list[sub_lang] = sub_formats
1044 return sub_lang_list
1045 # An extractor error can be raise by the download process if there are
1046 # no automatic captions but there are subtitles
1047 except (KeyError, ExtractorError):
1048 self._downloader.report_warning(err_msg)
1051 def _mark_watched(self, video_id, video_info):
1052 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1053 if not playback_url:
1055 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1056 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1058 # cpn generation algorithm is reverse engineered from base.js.
1059 # In fact it works even with dummy cpn.
1060 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1061 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1067 playback_url = compat_urlparse.urlunparse(
1068 parsed_playback_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
1070 self._download_webpage(
1071 playback_url, video_id, 'Marking watched',
1072 'Unable to mark watched', fatal=False)
1075 def extract_id(cls, url):
1076 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1078 raise ExtractorError('Invalid URL: %s' % url)
1079 video_id = mobj.group(2)
1082 def _extract_from_m3u8(self, manifest_url, video_id):
1085 def _get_urls(_manifest):
1086 lines = _manifest.split('\n')
1087 urls = filter(lambda l: l and not l.startswith('#'),
1090 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1091 formats_urls = _get_urls(manifest)
1092 for format_url in formats_urls:
1093 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1094 url_map[itag] = format_url
1097 def _extract_annotations(self, video_id):
1098 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1099 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1101 def _real_extract(self, url):
1102 url, smuggled_data = unsmuggle_url(url, {})
1105 'http' if self._downloader.params.get('prefer_insecure', False)
1110 parsed_url = compat_urllib_parse_urlparse(url)
1111 for component in [parsed_url.fragment, parsed_url.query]:
1112 query = compat_parse_qs(component)
1113 if start_time is None and 't' in query:
1114 start_time = parse_duration(query['t'][0])
1115 if start_time is None and 'start' in query:
1116 start_time = parse_duration(query['start'][0])
1117 if end_time is None and 'end' in query:
1118 end_time = parse_duration(query['end'][0])
1120 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1121 mobj = re.search(self._NEXT_URL_RE, url)
1123 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1124 video_id = self.extract_id(url)
1127 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1128 video_webpage = self._download_webpage(url, video_id)
1130 # Attempt to extract SWF player URL
1131 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1132 if mobj is not None:
1133 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1139 def add_dash_mpd(video_info):
1140 dash_mpd = video_info.get('dashmpd')
1141 if dash_mpd and dash_mpd[0] not in dash_mpds:
1142 dash_mpds.append(dash_mpd[0])
1145 embed_webpage = None
1147 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1149 # We simulate the access to the video from www.youtube.com/v/{video_id}
1150 # this can be viewed without login into Youtube
1151 url = proto + '://www.youtube.com/embed/%s' % video_id
1152 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1153 data = compat_urllib_parse.urlencode({
1154 'video_id': video_id,
1155 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1156 'sts': self._search_regex(
1157 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1159 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1160 video_info_webpage = self._download_webpage(
1161 video_info_url, video_id,
1162 note='Refetching age-gated info webpage',
1163 errnote='unable to download video info webpage')
1164 video_info = compat_parse_qs(video_info_webpage)
1165 add_dash_mpd(video_info)
1169 # Try looking directly into the video webpage
1170 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1172 args = ytplayer_config['args']
1173 if args.get('url_encoded_fmt_stream_map'):
1174 # Convert to the same format returned by compat_parse_qs
1175 video_info = dict((k, [v]) for k, v in args.items())
1176 add_dash_mpd(video_info)
1177 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1179 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1180 # We also try looking in get_video_info since it may contain different dashmpd
1181 # URL that points to a DASH manifest with possibly different itag set (some itags
1182 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1183 # manifest pointed by get_video_info's dashmpd).
1184 # The general idea is to take a union of itags of both DASH manifests (for example
1185 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1186 self.report_video_info_webpage_download(video_id)
1187 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1189 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1190 % (proto, video_id, el_type))
1191 video_info_webpage = self._download_webpage(
1193 video_id, note=False,
1194 errnote='unable to download video info webpage')
1195 get_video_info = compat_parse_qs(video_info_webpage)
1196 if get_video_info.get('use_cipher_signature') != ['True']:
1197 add_dash_mpd(get_video_info)
1199 video_info = get_video_info
1200 if 'token' in get_video_info:
1201 # Different get_video_info requests may report different results, e.g.
1202 # some may report video unavailability, but some may serve it without
1203 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1204 # the original webpage as well as el=info and el=embedded get_video_info
1205 # requests report video unavailability due to geo restriction while
1206 # el=detailpage succeeds and returns valid data). This is probably
1207 # due to YouTube measures against IP ranges of hosting providers.
1208 # Working around by preferring the first succeeded video_info containing
1209 # the token if no such video_info yet was found.
1210 if 'token' not in video_info:
1211 video_info = get_video_info
1213 if 'token' not in video_info:
1214 if 'reason' in video_info:
1215 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1216 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1218 raise ExtractorError('YouTube said: This video is available in %s only' % (
1219 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1221 raise ExtractorError(
1222 'YouTube said: %s' % video_info['reason'][0],
1223 expected=True, video_id=video_id)
1225 raise ExtractorError(
1226 '"token" parameter not in video info for unknown reason',
1230 if 'title' in video_info:
1231 video_title = video_info['title'][0]
1233 self._downloader.report_warning('Unable to extract video title')
1237 video_description = get_element_by_id("eow-description", video_webpage)
1238 if video_description:
1239 video_description = re.sub(r'''(?x)
1241 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1242 (?:title|href)="([^"]+)"\s+
1243 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1244 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1247 ''', r'\1', video_description)
1248 video_description = clean_html(video_description)
1250 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1252 video_description = unescapeHTML(fd_mobj.group(1))
1254 video_description = ''
1256 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1257 if not self._downloader.params.get('noplaylist'):
1260 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1261 for feed in multifeed_metadata_list.split(','):
1262 # Unquote should take place before split on comma (,) since textual
1263 # fields may contain comma as well (see
1264 # https://github.com/rg3/youtube-dl/issues/8536)
1265 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1267 '_type': 'url_transparent',
1268 'ie_key': 'Youtube',
1270 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1271 {'force_singlefeed': True}),
1272 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1274 feed_ids.append(feed_data['id'][0])
1276 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1277 % (', '.join(feed_ids), video_id))
1278 return self.playlist_result(entries, video_id, video_title, video_description)
1279 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1281 if 'view_count' in video_info:
1282 view_count = int(video_info['view_count'][0])
1286 # Check for "rental" videos
1287 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1288 raise ExtractorError('"rental" videos not supported')
1290 # Start extracting information
1291 self.report_information_extraction(video_id)
1294 if 'author' not in video_info:
1295 raise ExtractorError('Unable to extract uploader name')
1296 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1299 video_uploader_id = None
1300 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1301 if mobj is not None:
1302 video_uploader_id = mobj.group(1)
1304 self._downloader.report_warning('unable to extract uploader nickname')
1307 # We try first to get a high quality image:
1308 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1309 video_webpage, re.DOTALL)
1310 if m_thumb is not None:
1311 video_thumbnail = m_thumb.group(1)
1312 elif 'thumbnail_url' not in video_info:
1313 self._downloader.report_warning('unable to extract video thumbnail')
1314 video_thumbnail = None
1315 else: # don't panic if we can't find it
1316 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1319 upload_date = self._html_search_meta(
1320 'datePublished', video_webpage, 'upload date', default=None)
1322 upload_date = self._search_regex(
1323 [r'(?s)id="eow-date.*?>(.*?)</span>',
1324 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1325 video_webpage, 'upload date', default=None)
1327 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1328 upload_date = unified_strdate(upload_date)
1330 m_music = re.search(
1331 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1334 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1335 video_creator = clean_html(m_music.group('creator'))
1337 video_alt_title = video_creator = None
1339 m_cat_container = self._search_regex(
1340 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1341 video_webpage, 'categories', default=None)
1343 category = self._html_search_regex(
1344 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1346 video_categories = None if category is None else [category]
1348 video_categories = None
1351 unescapeHTML(m.group('content'))
1352 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1354 def _extract_count(count_name):
1355 return str_to_int(self._search_regex(
1356 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1357 % re.escape(count_name),
1358 video_webpage, count_name, default=None))
1360 like_count = _extract_count('like')
1361 dislike_count = _extract_count('dislike')
1364 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1365 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1367 if 'length_seconds' not in video_info:
1368 self._downloader.report_warning('unable to extract video duration')
1369 video_duration = None
1371 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1374 video_annotations = None
1375 if self._downloader.params.get('writeannotations', False):
1376 video_annotations = self._extract_annotations(video_id)
1378 def _map_to_format_list(urlmap):
1380 for itag, video_real_url in urlmap.items():
1383 'url': video_real_url,
1384 'player_url': player_url,
1386 if itag in self._formats:
1387 dct.update(self._formats[itag])
1391 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1392 self.report_rtmp_download()
1394 'format_id': '_rtmp',
1396 'url': video_info['conn'][0],
1397 'player_url': player_url,
1399 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1400 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1401 if 'rtmpe%3Dyes' in encoded_url_map:
1402 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1403 fmt_list = video_info.get('fmt_list', [''])[0]
1405 for fmt in fmt_list.split(','):
1406 spec = fmt.split('/')
1407 width, height = spec[1].split('x')
1408 self._formats[spec[0]].update({
1409 'resolution': spec[1],
1410 'width': int_or_none(width),
1411 'height': int_or_none(height),
1414 for url_data_str in encoded_url_map.split(','):
1415 url_data = compat_parse_qs(url_data_str)
1416 if 'itag' not in url_data or 'url' not in url_data:
1418 format_id = url_data['itag'][0]
1419 url = url_data['url'][0]
1421 if 'sig' in url_data:
1422 url += '&signature=' + url_data['sig'][0]
1423 elif 's' in url_data:
1424 encrypted_sig = url_data['s'][0]
1425 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1427 jsplayer_url_json = self._search_regex(
1429 embed_webpage if age_gate else video_webpage,
1430 'JS player URL (1)', default=None)
1431 if not jsplayer_url_json and not age_gate:
1432 # We need the embed website after all
1433 if embed_webpage is None:
1434 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1435 embed_webpage = self._download_webpage(
1436 embed_url, video_id, 'Downloading embed webpage')
1437 jsplayer_url_json = self._search_regex(
1438 ASSETS_RE, embed_webpage, 'JS player URL')
1440 player_url = json.loads(jsplayer_url_json)
1441 if player_url is None:
1442 player_url_json = self._search_regex(
1443 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1444 video_webpage, 'age gate player URL')
1445 player_url = json.loads(player_url_json)
1447 if self._downloader.params.get('verbose'):
1448 if player_url is None:
1449 player_version = 'unknown'
1450 player_desc = 'unknown'
1452 if player_url.endswith('swf'):
1453 player_version = self._search_regex(
1454 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1455 'flash player', fatal=False)
1456 player_desc = 'flash player %s' % player_version
1458 player_version = self._search_regex(
1459 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1461 'html5 player', fatal=False)
1462 player_desc = 'html5 player %s' % player_version
1464 parts_sizes = self._signature_cache_id(encrypted_sig)
1465 self.to_screen('{%s} signature length %s, %s' %
1466 (format_id, parts_sizes, player_desc))
1468 signature = self._decrypt_signature(
1469 encrypted_sig, video_id, player_url, age_gate)
1470 url += '&signature=' + signature
1471 if 'ratebypass' not in url:
1472 url += '&ratebypass=yes'
1475 'format_id': format_id,
1477 'player_url': player_url,
1479 if format_id in self._formats:
1480 dct.update(self._formats[format_id])
1482 # Some itags are not included in DASH manifest thus corresponding formats will
1483 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1484 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1485 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1486 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1489 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1490 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1493 'fps': int_or_none(url_data.get('fps', [None])[0]),
1494 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1496 for key, value in more_fields.items():
1499 type_ = url_data.get('type', [None])[0]
1501 type_split = type_.split(';')
1502 kind_ext = type_split[0].split('/')
1503 if len(kind_ext) == 2:
1505 dct['ext'] = mimetype2ext(type_split[0])
1506 if kind in ('audio', 'video'):
1508 for mobj in re.finditer(
1509 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1510 if mobj.group('key') == 'codecs':
1511 codecs = mobj.group('val')
1514 codecs = codecs.split(',')
1515 if len(codecs) == 2:
1516 acodec, vcodec = codecs[1], codecs[0]
1518 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1524 elif video_info.get('hlsvp'):
1525 manifest_url = video_info['hlsvp'][0]
1526 url_map = self._extract_from_m3u8(manifest_url, video_id)
1527 formats = _map_to_format_list(url_map)
1528 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1529 for a_format in formats:
1530 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1532 unavailable_message = self._html_search_regex(
1533 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1534 video_webpage, 'unavailable message', default=None)
1535 if unavailable_message:
1536 raise ExtractorError(unavailable_message, expected=True)
1537 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1539 # Look for the DASH manifest
1540 if self._downloader.params.get('youtube_include_dash_manifest', True):
1541 dash_mpd_fatal = True
1542 for mpd_url in dash_mpds:
1545 def decrypt_sig(mobj):
1547 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1548 return '/signature/%s' % dec_s
1550 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1552 for df in self._extract_mpd_formats(
1553 mpd_url, video_id, fatal=dash_mpd_fatal,
1554 formats_dict=self._formats):
1555 # Do not overwrite DASH format found in some previous DASH manifest
1556 if df['format_id'] not in dash_formats:
1557 dash_formats[df['format_id']] = df
1558 # Additional DASH manifests may end up in HTTP Error 403 therefore
1559 # allow them to fail without bug report message if we already have
1560 # some DASH manifest succeeded. This is temporary workaround to reduce
1561 # burst of bug reports until we figure out the reason and whether it
1562 # can be fixed at all.
1563 dash_mpd_fatal = False
1564 except (ExtractorError, KeyError) as e:
1565 self.report_warning(
1566 'Skipping DASH manifest: %r' % e, video_id)
1568 # Remove the formats we found through non-DASH, they
1569 # contain less info and it can be wrong, because we use
1570 # fixed values (for example the resolution). See
1571 # https://github.com/rg3/youtube-dl/issues/5774 for an
1573 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1574 formats.extend(dash_formats.values())
1576 # Check for malformed aspect ratio
1577 stretched_m = re.search(
1578 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1581 w = float(stretched_m.group('w'))
1582 h = float(stretched_m.group('h'))
1583 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1584 # We will only process correct ratios.
1588 if f.get('vcodec') != 'none':
1589 f['stretched_ratio'] = ratio
1591 self._sort_formats(formats)
1593 self.mark_watched(video_id, video_info)
1597 'uploader': video_uploader,
1598 'uploader_id': video_uploader_id,
1599 'upload_date': upload_date,
1600 'creator': video_creator,
1601 'title': video_title,
1602 'alt_title': video_alt_title,
1603 'thumbnail': video_thumbnail,
1604 'description': video_description,
1605 'categories': video_categories,
1607 'subtitles': video_subtitles,
1608 'automatic_captions': automatic_captions,
1609 'duration': video_duration,
1610 'age_limit': 18 if age_gate else 0,
1611 'annotations': video_annotations,
1612 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1613 'view_count': view_count,
1614 'like_count': like_count,
1615 'dislike_count': dislike_count,
1616 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1619 'start_time': start_time,
1620 'end_time': end_time,
1624 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1625 IE_DESC = 'YouTube.com playlists'
1626 _VALID_URL = r"""(?x)(?:
1631 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1632 \? (?:.*?[&;])*? (?:p|a|list)=
1636 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1637 # Top tracks, they can also include dots
1642 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1644 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1645 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1646 IE_NAME = 'youtube:playlist'
1648 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1650 'title': 'ytdl test PL',
1651 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1653 'playlist_count': 3,
1655 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1657 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1658 'title': 'YDL_Empty_List',
1660 'playlist_count': 0,
1662 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1663 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1665 'title': '29C3: Not my department',
1666 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1668 'playlist_count': 95,
1670 'note': 'issue #673',
1671 'url': 'PLBB231211A4F62143',
1673 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1674 'id': 'PLBB231211A4F62143',
1676 'playlist_mincount': 26,
1678 'note': 'Large playlist',
1679 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1681 'title': 'Uploads from Cauchemar',
1682 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1684 'playlist_mincount': 799,
1686 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1688 'title': 'YDL_safe_search',
1689 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1691 'playlist_count': 2,
1694 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1695 'playlist_count': 4,
1698 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1701 'note': 'Embedded SWF player',
1702 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1703 'playlist_count': 4,
1706 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1709 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1710 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1712 'title': 'Uploads from Interstellar Movie',
1713 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1715 'playlist_mincout': 21,
1718 def _real_initialize(self):
1721 def _extract_mix(self, playlist_id):
1722 # The mixes are generated from a single video
1723 # the id of the playlist is just 'RD' + video_id
1724 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1725 webpage = self._download_webpage(
1726 url, playlist_id, 'Downloading Youtube mix')
1727 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1729 search_title('playlist-title') or
1730 search_title('title long-title') or
1731 search_title('title'))
1732 title = clean_html(title_span)
1733 ids = orderedSet(re.findall(
1734 r'''(?xs)data-video-username=".*?".*?
1735 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1737 url_results = self._ids_to_results(ids)
1739 return self.playlist_result(url_results, playlist_id, title)
1741 def _extract_playlist(self, playlist_id):
1742 url = self._TEMPLATE_URL % playlist_id
1743 page = self._download_webpage(url, playlist_id)
1745 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1746 match = match.strip()
1747 # Check if the playlist exists or is private
1748 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1749 raise ExtractorError(
1750 'The playlist doesn\'t exist or is private, use --username or '
1751 '--netrc to access it.',
1753 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1754 raise ExtractorError(
1755 'Invalid parameters. Maybe URL is incorrect.',
1757 elif re.match(r'[^<]*Choose your language[^<]*', match):
1760 self.report_warning('Youtube gives an alert message: ' + match)
1762 playlist_title = self._html_search_regex(
1763 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1766 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1768 def _check_download_just_video(self, url, playlist_id):
1769 # Check if it's a video-specific URL
1770 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1771 if 'v' in query_dict:
1772 video_id = query_dict['v'][0]
1773 if self._downloader.params.get('noplaylist'):
1774 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1775 return self.url_result(video_id, 'Youtube', video_id=video_id)
1777 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1779 def _real_extract(self, url):
1780 # Extract playlist id
1781 mobj = re.match(self._VALID_URL, url)
1783 raise ExtractorError('Invalid URL: %s' % url)
1784 playlist_id = mobj.group(1) or mobj.group(2)
1786 video = self._check_download_just_video(url, playlist_id)
1790 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1791 # Mixes require a custom extraction process
1792 return self._extract_mix(playlist_id)
1794 return self._extract_playlist(playlist_id)
1797 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1798 IE_DESC = 'YouTube.com channels'
1799 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1800 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1801 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1802 IE_NAME = 'youtube:channel'
1804 'note': 'paginated channel',
1805 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1806 'playlist_mincount': 91,
1808 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1809 'title': 'Uploads from lex will',
1812 'note': 'Age restricted channel',
1813 # from https://www.youtube.com/user/DeusExOfficial
1814 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1815 'playlist_mincount': 64,
1817 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1818 'title': 'Uploads from Deus Ex',
1823 def suitable(cls, url):
1824 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1826 def _real_extract(self, url):
1827 channel_id = self._match_id(url)
1829 url = self._TEMPLATE_URL % channel_id
1831 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1832 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1833 # otherwise fallback on channel by page extraction
1834 channel_page = self._download_webpage(
1835 url + '?view=57', channel_id,
1836 'Downloading channel page', fatal=False)
1837 if channel_page is False:
1838 channel_playlist_id = False
1840 channel_playlist_id = self._html_search_meta(
1841 'channelId', channel_page, 'channel id', default=None)
1842 if not channel_playlist_id:
1843 channel_playlist_id = self._search_regex(
1844 r'data-(?:channel-external-|yt)id="([^"]+)"',
1845 channel_page, 'channel id', default=None)
1846 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1847 playlist_id = 'UU' + channel_playlist_id[2:]
1848 return self.url_result(
1849 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1851 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1852 autogenerated = re.search(r'''(?x)
1854 channel-header-autogenerated-label|
1855 yt-channel-title-autogenerated
1856 )[^"]*"''', channel_page) is not None
1859 # The videos are contained in a single page
1860 # the ajax pages can't be used, they are empty
1863 video_id, 'Youtube', video_id=video_id,
1864 video_title=video_title)
1865 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1866 return self.playlist_result(entries, channel_id)
1868 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1871 class YoutubeUserIE(YoutubeChannelIE):
1872 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1873 _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_-]+)'
1874 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1875 IE_NAME = 'youtube:user'
1878 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1879 'playlist_mincount': 320,
1881 'title': 'TheLinuxFoundation',
1884 'url': 'ytuser:phihag',
1885 'only_matching': True,
1889 def suitable(cls, url):
1890 # Don't return True if the url can be extracted with other youtube
1891 # extractor, the regex would is too permissive and it would match.
1892 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1893 if any(ie.suitable(url) for ie in other_ies):
1896 return super(YoutubeUserIE, cls).suitable(url)
1899 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1900 IE_DESC = 'YouTube.com user/channel playlists'
1901 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1902 IE_NAME = 'youtube:playlists'
1905 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1906 'playlist_mincount': 4,
1908 'id': 'ThirstForScience',
1909 'title': 'Thirst for Science',
1912 # with "Load more" button
1913 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1914 'playlist_mincount': 70,
1917 'title': 'Игорь Клейнер',
1920 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1921 'playlist_mincount': 17,
1923 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1924 'title': 'Chem Player',
1929 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1930 IE_DESC = 'YouTube.com searches'
1931 # there doesn't appear to be a real limit, for example if you search for
1932 # 'python' you get more than 8.000.000 results
1933 _MAX_RESULTS = float('inf')
1934 IE_NAME = 'youtube:search'
1935 _SEARCH_KEY = 'ytsearch'
1936 _EXTRA_QUERY_ARGS = {}
1939 def _get_n_results(self, query, n):
1940 """Get a specified number of results for a query"""
1945 for pagenum in itertools.count(1):
1947 'search_query': query.encode('utf-8'),
1951 url_query.update(self._EXTRA_QUERY_ARGS)
1952 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1953 data = self._download_json(
1954 result_url, video_id='query "%s"' % query,
1955 note='Downloading page %s' % pagenum,
1956 errnote='Unable to download API page')
1957 html_content = data[1]['body']['content']
1959 if 'class="search-message' in html_content:
1960 raise ExtractorError(
1961 '[youtube] No video results', expected=True)
1963 new_videos = self._ids_to_results(orderedSet(re.findall(
1964 r'href="/watch\?v=(.{11})', html_content)))
1965 videos += new_videos
1966 if not new_videos or len(videos) > limit:
1971 return self.playlist_result(videos, query)
1974 class YoutubeSearchDateIE(YoutubeSearchIE):
1975 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1976 _SEARCH_KEY = 'ytsearchdate'
1977 IE_DESC = 'YouTube.com searches, newest videos first'
1978 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1981 class YoutubeSearchURLIE(InfoExtractor):
1982 IE_DESC = 'YouTube.com search URLs'
1983 IE_NAME = 'youtube:search_url'
1984 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
1986 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1987 'playlist_mincount': 5,
1989 'title': 'youtube-dl test video',
1992 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
1993 'only_matching': True,
1996 def _real_extract(self, url):
1997 mobj = re.match(self._VALID_URL, url)
1998 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2000 webpage = self._download_webpage(url, query)
2001 result_code = self._search_regex(
2002 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2004 part_codes = re.findall(
2005 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2007 for part_code in part_codes:
2008 part_title = self._html_search_regex(
2009 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2010 part_url_snippet = self._html_search_regex(
2011 r'(?s)href="([^"]+)"', part_code, 'item URL')
2012 part_url = compat_urlparse.urljoin(
2013 'https://www.youtube.com/', part_url_snippet)
2017 'title': part_title,
2021 '_type': 'playlist',
2027 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2028 IE_DESC = 'YouTube.com (multi-season) shows'
2029 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2030 IE_NAME = 'youtube:show'
2032 'url': 'https://www.youtube.com/show/airdisasters',
2033 'playlist_mincount': 5,
2035 'id': 'airdisasters',
2036 'title': 'Air Disasters',
2040 def _real_extract(self, url):
2041 playlist_id = self._match_id(url)
2042 return super(YoutubeShowIE, self)._real_extract(
2043 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2046 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2048 Base class for feed extractors
2049 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2051 _LOGIN_REQUIRED = True
2055 return 'youtube:%s' % self._FEED_NAME
2057 def _real_initialize(self):
2060 def _real_extract(self, url):
2061 page = self._download_webpage(
2062 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2064 # The extraction process is the same as for playlists, but the regex
2065 # for the video ids doesn't contain an index
2067 more_widget_html = content_html = page
2068 for page_num in itertools.count(1):
2069 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2071 # 'recommended' feed has infinite 'load more' and each new portion spins
2072 # the same videos in (sometimes) slightly different order, so we'll check
2073 # for unicity and break when portion has no new videos
2074 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2080 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2084 more = self._download_json(
2085 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2086 'Downloading page #%s' % page_num,
2087 transform_source=uppercase_escape)
2088 content_html = more['content_html']
2089 more_widget_html = more['load_more_widget_html']
2091 return self.playlist_result(
2092 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2095 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2096 IE_NAME = 'youtube:watchlater'
2097 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2098 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2101 'url': 'https://www.youtube.com/playlist?list=WL',
2102 'only_matching': True,
2104 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2105 'only_matching': True,
2108 def _real_extract(self, url):
2109 video = self._check_download_just_video(url, 'WL')
2112 return self._extract_playlist('WL')
2115 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2116 IE_NAME = 'youtube:favorites'
2117 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2118 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2119 _LOGIN_REQUIRED = True
2121 def _real_extract(self, url):
2122 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2123 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2124 return self.url_result(playlist_id, 'YoutubePlaylist')
2127 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2128 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2129 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2130 _FEED_NAME = 'recommended'
2131 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2134 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2135 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2136 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2137 _FEED_NAME = 'subscriptions'
2138 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2141 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2142 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2143 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2144 _FEED_NAME = 'history'
2145 _PLAYLIST_TITLE = 'Youtube History'
2148 class YoutubeTruncatedURLIE(InfoExtractor):
2149 IE_NAME = 'youtube:truncated_url'
2150 IE_DESC = False # Do not list
2151 _VALID_URL = r'''(?x)
2153 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2156 annotation_id=annotation_[^&]+|
2162 attribution_link\?a=[^&]+
2168 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2169 'only_matching': True,
2171 'url': 'http://www.youtube.com/watch?',
2172 'only_matching': True,
2174 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2175 'only_matching': True,
2177 'url': 'https://www.youtube.com/watch?feature=foo',
2178 'only_matching': True,
2180 'url': 'https://www.youtube.com/watch?hl=en-GB',
2181 'only_matching': True,
2183 'url': 'https://www.youtube.com/watch?t=2372',
2184 'only_matching': True,
2187 def _real_extract(self, url):
2188 raise ExtractorError(
2189 'Did you forget to quote the URL? Remember that & is a meta '
2190 'character in most shells, so you want to put the URL in quotes, '
2192 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2193 ' or simply youtube-dl BaW_jenozKc .',
2197 class YoutubeTruncatedIDIE(InfoExtractor):
2198 IE_NAME = 'youtube:truncated_id'
2199 IE_DESC = False # Do not list
2200 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2203 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2204 'only_matching': True,
2207 def _real_extract(self, url):
2208 video_id = self._match_id(url)
2209 raise ExtractorError(
2210 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),