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 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
313 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
314 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
315 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
316 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
317 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
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},
323 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
325 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
326 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
327 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # 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},
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'},
336 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
337 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
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 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
388 'upload_date': '20121002',
389 'license': 'Standard YouTube License',
390 '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 .',
391 'categories': ['Science & Technology'],
392 'tags': ['youtube-dl'],
394 'dislike_count': int,
400 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
401 'note': 'Test generic use_cipher_signature video (#897)',
405 'upload_date': '20120506',
406 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
407 'alt_title': 'I Love It (feat. Charli XCX)',
408 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
409 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
410 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
411 'iconic ep', 'iconic', 'love', 'it'],
412 'uploader': 'Icona Pop',
413 'uploader_id': 'IconaPop',
414 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
415 'license': 'Standard YouTube License',
416 'creator': 'Icona Pop',
420 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
421 'note': 'Test VEVO video with age protection (#956)',
425 'upload_date': '20130703',
426 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
427 'alt_title': 'Tunnel Vision',
428 'description': 'md5:64249768eec3bc4276236606ea996373',
429 'uploader': 'justintimberlakeVEVO',
430 'uploader_id': 'justintimberlakeVEVO',
431 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
432 'license': 'Standard YouTube License',
433 'creator': 'Justin Timberlake',
438 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
439 'note': 'Embed-only video (#1746)',
443 'upload_date': '20120608',
444 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
445 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
446 'uploader': 'SET India',
447 'uploader_id': 'setindia',
448 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
449 'license': 'Standard YouTube License',
454 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
455 'note': 'Use the first video ID in the URL',
459 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
460 'uploader': 'Philipp Hagemeister',
461 'uploader_id': 'phihag',
462 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
463 'upload_date': '20121002',
464 'license': 'Standard YouTube License',
465 '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 .',
466 'categories': ['Science & Technology'],
467 'tags': ['youtube-dl'],
469 'dislike_count': int,
472 'skip_download': True,
476 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
477 'note': '256k DASH audio (format 141) via DASH manifest',
481 'upload_date': '20121002',
482 'uploader_id': '8KVIDEO',
483 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
485 'uploader': '8KVIDEO',
486 'license': 'Standard YouTube License',
487 'title': 'UHDTV TEST 8K VIDEO.mp4'
490 'youtube_include_dash_manifest': True,
494 # DASH manifest with encrypted signature
496 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
500 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
501 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
502 'uploader': 'AfrojackVEVO',
503 'uploader_id': 'AfrojackVEVO',
504 'upload_date': '20131011',
505 'license': 'Standard YouTube License',
508 'youtube_include_dash_manifest': True,
512 # JS player signature function name containing $
514 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
518 'title': 'Taylor Swift - Shake It Off',
519 'alt_title': 'Shake It Off',
520 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
521 'uploader': 'TaylorSwiftVEVO',
522 'uploader_id': 'TaylorSwiftVEVO',
523 'upload_date': '20140818',
524 'license': 'Standard YouTube License',
525 'creator': 'Taylor Swift',
528 'youtube_include_dash_manifest': True,
534 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
538 'upload_date': '20100909',
539 'uploader': 'The Amazing Atheist',
540 'uploader_id': 'TheAmazingAtheist',
541 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
542 'license': 'Standard YouTube License',
543 'title': 'Burning Everyone\'s Koran',
544 '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',
547 # Normal age-gate video (No vevo, embed allowed)
549 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
553 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
554 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
555 'uploader': 'The Witcher',
556 'uploader_id': 'WitcherGame',
557 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
558 'upload_date': '20140605',
559 'license': 'Standard YouTube License',
563 # Age-gate video with encrypted signature
565 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
569 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
570 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
571 'uploader': 'LloydVEVO',
572 'uploader_id': 'LloydVEVO',
573 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
574 'upload_date': '20110629',
575 'license': 'Standard YouTube License',
579 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
581 'url': '__2ABJjxzNo',
585 'upload_date': '20100430',
586 'uploader_id': 'deadmau5',
587 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
588 'creator': 'deadmau5',
589 'description': 'md5:12c56784b8032162bb936a5f76d55360',
590 'uploader': 'deadmau5',
591 'license': 'Standard YouTube License',
592 'title': 'Deadmau5 - Some Chords (HD)',
593 'alt_title': 'Some Chords',
595 'expected_warnings': [
596 'DASH manifest missing',
599 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
601 'url': 'lqQg6PlCWgI',
605 'upload_date': '20150827',
606 'uploader_id': 'olympic',
607 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
608 'license': 'Standard YouTube License',
609 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
610 'uploader': 'Olympics',
611 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
614 'skip_download': 'requires avconv',
619 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
623 'stretched_ratio': 16 / 9.,
624 'upload_date': '20110310',
625 'uploader_id': 'AllenMeow',
626 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
627 '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',
629 'license': 'Standard YouTube License',
630 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
633 # url_encoded_fmt_stream_map is empty string
635 'url': 'qEJwOuvDf7I',
639 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
641 'upload_date': '20150404',
642 'uploader_id': 'spbelect',
643 'uploader': 'Наблюдатели Петербурга',
646 'skip_download': 'requires avconv',
648 'skip': 'This live event has ended.',
650 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
652 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
656 'title': 'md5:7b81415841e02ecd4313668cde88737a',
657 'description': 'md5:116377fd2963b81ec4ce64b542173306',
658 'upload_date': '20150625',
659 'uploader_id': 'dorappi2000',
660 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
661 'uploader': 'dorappi2000',
662 'license': 'Standard YouTube License',
663 'formats': 'mincount:33',
666 # DASH manifest with segment_list
668 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
669 'md5': '8ce563a1d667b599d21064e982ab9e31',
673 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
674 'uploader': 'Airtek',
675 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
676 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
677 'license': 'Standard YouTube License',
678 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
681 'youtube_include_dash_manifest': True,
682 'format': '135', # bestvideo
686 # Multifeed videos (multiple cameras), URL is for Main Camera
687 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
690 'title': 'teamPGP: Rocket League Noob Stream',
691 'description': 'md5:dc7872fb300e143831327f1bae3af010',
697 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
698 'description': 'md5:dc7872fb300e143831327f1bae3af010',
699 'upload_date': '20150721',
700 'uploader': 'Beer Games Beer',
701 'uploader_id': 'beergamesbeer',
702 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
703 'license': 'Standard YouTube License',
709 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
710 'description': 'md5:dc7872fb300e143831327f1bae3af010',
711 'upload_date': '20150721',
712 'uploader': 'Beer Games Beer',
713 'uploader_id': 'beergamesbeer',
714 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
715 'license': 'Standard YouTube License',
721 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
722 'description': 'md5:dc7872fb300e143831327f1bae3af010',
723 'upload_date': '20150721',
724 'uploader': 'Beer Games Beer',
725 'uploader_id': 'beergamesbeer',
726 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
727 'license': 'Standard YouTube License',
733 'title': 'teamPGP: Rocket League Noob Stream (zim)',
734 'description': 'md5:dc7872fb300e143831327f1bae3af010',
735 'upload_date': '20150721',
736 'uploader': 'Beer Games Beer',
737 'uploader_id': 'beergamesbeer',
738 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
739 'license': 'Standard YouTube License',
743 'skip_download': True,
747 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
748 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
751 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
756 'url': 'http://vid.plus/FlRa-iH7PGw',
757 'only_matching': True,
760 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
761 # Also tests cut-off URL expansion in video description (see
762 # https://github.com/rg3/youtube-dl/issues/1892,
763 # https://github.com/rg3/youtube-dl/issues/8164)
764 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
768 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
769 'alt_title': 'Dark Walk',
770 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
771 'upload_date': '20151119',
772 'uploader_id': 'IronSoulElf',
773 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
774 'uploader': 'IronSoulElf',
775 'license': 'Standard YouTube License',
776 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
779 'skip_download': True,
783 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
784 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
785 'only_matching': True,
788 # Video with yt:stretch=17:0
789 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
793 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
794 'description': 'md5:ee18a25c350637c8faff806845bddee9',
795 'upload_date': '20151107',
796 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
797 'uploader': 'CH GAMER DROID',
800 'skip_download': True,
804 # Video licensed under Creative Commons
805 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
809 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
810 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
811 'upload_date': '20150127',
812 'uploader_id': 'BerkmanCenter',
813 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
814 'uploader': 'BerkmanCenter',
815 'license': 'Creative Commons Attribution license (reuse allowed)',
818 'skip_download': True,
822 # Channel-like uploader_url
823 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
827 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
828 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
829 'upload_date': '20151119',
830 'uploader': 'Bernie 2016',
831 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
833 'license': 'Creative Commons Attribution license (reuse allowed)',
836 'skip_download': True,
840 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
841 'only_matching': True,
845 def __init__(self, *args, **kwargs):
846 super(YoutubeIE, self).__init__(*args, **kwargs)
847 self._player_cache = {}
849 def report_video_info_webpage_download(self, video_id):
850 """Report attempt to download video info webpage."""
851 self.to_screen('%s: Downloading video info webpage' % video_id)
853 def report_information_extraction(self, video_id):
854 """Report attempt to extract video information."""
855 self.to_screen('%s: Extracting video information' % video_id)
857 def report_unavailable_format(self, video_id, format):
858 """Report extracted video URL."""
859 self.to_screen('%s: Format %s not available' % (video_id, format))
861 def report_rtmp_download(self):
862 """Indicate the download will use the RTMP protocol."""
863 self.to_screen('RTMP download detected')
865 def _signature_cache_id(self, example_sig):
866 """ Return a string representation of a signature """
867 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
869 def _extract_signature_function(self, video_id, player_url, example_sig):
871 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
874 raise ExtractorError('Cannot identify player %r' % player_url)
875 player_type = id_m.group('ext')
876 player_id = id_m.group('id')
878 # Read from filesystem cache
879 func_id = '%s_%s_%s' % (
880 player_type, player_id, self._signature_cache_id(example_sig))
881 assert os.path.basename(func_id) == func_id
883 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
884 if cache_spec is not None:
885 return lambda s: ''.join(s[i] for i in cache_spec)
888 'Downloading player %s' % player_url
889 if self._downloader.params.get('verbose') else
890 'Downloading %s player %s' % (player_type, player_id)
892 if player_type == 'js':
893 code = self._download_webpage(
894 player_url, video_id,
896 errnote='Download of %s failed' % player_url)
897 res = self._parse_sig_js(code)
898 elif player_type == 'swf':
899 urlh = self._request_webpage(
900 player_url, video_id,
902 errnote='Download of %s failed' % player_url)
904 res = self._parse_sig_swf(code)
906 assert False, 'Invalid player type %r' % player_type
908 test_string = ''.join(map(compat_chr, range(len(example_sig))))
909 cache_res = res(test_string)
910 cache_spec = [ord(c) for c in cache_res]
912 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
915 def _print_sig_code(self, func, example_sig):
916 def gen_sig_code(idxs):
917 def _genslice(start, end, step):
918 starts = '' if start == 0 else str(start)
919 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
920 steps = '' if step == 1 else (':%d' % step)
921 return 's[%s%s%s]' % (starts, ends, steps)
924 # Quelch pyflakes warnings - start will be set when step is set
925 start = '(Never used)'
926 for i, prev in zip(idxs[1:], idxs[:-1]):
930 yield _genslice(start, prev, step)
933 if i - prev in [-1, 1]:
942 yield _genslice(start, i, step)
944 test_string = ''.join(map(compat_chr, range(len(example_sig))))
945 cache_res = func(test_string)
946 cache_spec = [ord(c) for c in cache_res]
947 expr_code = ' + '.join(gen_sig_code(cache_spec))
948 signature_id_tuple = '(%s)' % (
949 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
950 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
951 ' return %s\n') % (signature_id_tuple, expr_code)
952 self.to_screen('Extracted signature function:\n' + code)
954 def _parse_sig_js(self, jscode):
955 funcname = self._search_regex(
956 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
957 'Initial JS player signature function name')
959 jsi = JSInterpreter(jscode)
960 initial_function = jsi.extract_function(funcname)
961 return lambda s: initial_function([s])
963 def _parse_sig_swf(self, file_contents):
964 swfi = SWFInterpreter(file_contents)
965 TARGET_CLASSNAME = 'SignatureDecipher'
966 searched_class = swfi.extract_class(TARGET_CLASSNAME)
967 initial_function = swfi.extract_function(searched_class, 'decipher')
968 return lambda s: initial_function([s])
970 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
971 """Turn the encrypted s field into a working signature"""
973 if player_url is None:
974 raise ExtractorError('Cannot decrypt signature without player_url')
976 if player_url.startswith('//'):
977 player_url = 'https:' + player_url
979 player_id = (player_url, self._signature_cache_id(s))
980 if player_id not in self._player_cache:
981 func = self._extract_signature_function(
982 video_id, player_url, s
984 self._player_cache[player_id] = func
985 func = self._player_cache[player_id]
986 if self._downloader.params.get('youtube_print_sig_code'):
987 self._print_sig_code(func, s)
989 except Exception as e:
990 tb = traceback.format_exc()
991 raise ExtractorError(
992 'Signature extraction failed: ' + tb, cause=e)
994 def _get_subtitles(self, video_id, webpage):
996 subs_doc = self._download_xml(
997 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
998 video_id, note=False)
999 except ExtractorError as err:
1000 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1004 for track in subs_doc.findall('track'):
1005 lang = track.attrib['lang_code']
1006 if lang in sub_lang_list:
1009 for ext in self._SUBTITLE_FORMATS:
1010 params = compat_urllib_parse.urlencode({
1014 'name': track.attrib['name'].encode('utf-8'),
1016 sub_formats.append({
1017 'url': 'https://www.youtube.com/api/timedtext?' + params,
1020 sub_lang_list[lang] = sub_formats
1021 if not sub_lang_list:
1022 self._downloader.report_warning('video doesn\'t have subtitles')
1024 return sub_lang_list
1026 def _get_ytplayer_config(self, video_id, webpage):
1028 # User data may contain arbitrary character sequences that may affect
1029 # JSON extraction with regex, e.g. when '};' is contained the second
1030 # regex won't capture the whole JSON. Yet working around by trying more
1031 # concrete regex first keeping in mind proper quoted string handling
1032 # to be implemented in future that will replace this workaround (see
1033 # https://github.com/rg3/youtube-dl/issues/7468,
1034 # https://github.com/rg3/youtube-dl/pull/7599)
1035 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1036 r';ytplayer\.config\s*=\s*({.+?});',
1038 config = self._search_regex(
1039 patterns, webpage, 'ytplayer.config', default=None)
1041 return self._parse_json(
1042 uppercase_escape(config), video_id, fatal=False)
1044 def _get_automatic_captions(self, video_id, webpage):
1045 """We need the webpage for getting the captions url, pass it as an
1046 argument to speed up the process."""
1047 self.to_screen('%s: Looking for automatic captions' % video_id)
1048 player_config = self._get_ytplayer_config(video_id, webpage)
1049 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1050 if not player_config:
1051 self._downloader.report_warning(err_msg)
1054 args = player_config['args']
1055 caption_url = args.get('ttsurl')
1057 timestamp = args['timestamp']
1058 # We get the available subtitles
1059 list_params = compat_urllib_parse.urlencode({
1064 list_url = caption_url + '&' + list_params
1065 caption_list = self._download_xml(list_url, video_id)
1066 original_lang_node = caption_list.find('track')
1067 if original_lang_node is None:
1068 self._downloader.report_warning('Video doesn\'t have automatic captions')
1070 original_lang = original_lang_node.attrib['lang_code']
1071 caption_kind = original_lang_node.attrib.get('kind', '')
1074 for lang_node in caption_list.findall('target'):
1075 sub_lang = lang_node.attrib['lang_code']
1077 for ext in self._SUBTITLE_FORMATS:
1078 params = compat_urllib_parse.urlencode({
1079 'lang': original_lang,
1083 'kind': caption_kind,
1085 sub_formats.append({
1086 'url': caption_url + '&' + params,
1089 sub_lang_list[sub_lang] = sub_formats
1090 return sub_lang_list
1092 # Some videos don't provide ttsurl but rather caption_tracks and
1093 # caption_translation_languages (e.g. 20LmZk1hakA)
1094 caption_tracks = args['caption_tracks']
1095 caption_translation_languages = args['caption_translation_languages']
1096 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1097 parsed_caption_url = compat_urlparse.urlparse(caption_url)
1098 caption_qs = compat_parse_qs(parsed_caption_url.query)
1101 for lang in caption_translation_languages.split(','):
1102 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1103 sub_lang = lang_qs.get('lc', [None])[0]
1107 for ext in self._SUBTITLE_FORMATS:
1109 'tlang': [sub_lang],
1112 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1113 query=compat_urllib_parse.urlencode(caption_qs, True)))
1114 sub_formats.append({
1118 sub_lang_list[sub_lang] = sub_formats
1119 return sub_lang_list
1120 # An extractor error can be raise by the download process if there are
1121 # no automatic captions but there are subtitles
1122 except (KeyError, ExtractorError):
1123 self._downloader.report_warning(err_msg)
1126 def _mark_watched(self, video_id, video_info):
1127 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1128 if not playback_url:
1130 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1131 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1133 # cpn generation algorithm is reverse engineered from base.js.
1134 # In fact it works even with dummy cpn.
1135 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1136 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1142 playback_url = compat_urlparse.urlunparse(
1143 parsed_playback_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
1145 self._download_webpage(
1146 playback_url, video_id, 'Marking watched',
1147 'Unable to mark watched', fatal=False)
1150 def extract_id(cls, url):
1151 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1153 raise ExtractorError('Invalid URL: %s' % url)
1154 video_id = mobj.group(2)
1157 def _extract_from_m3u8(self, manifest_url, video_id):
1160 def _get_urls(_manifest):
1161 lines = _manifest.split('\n')
1162 urls = filter(lambda l: l and not l.startswith('#'),
1165 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1166 formats_urls = _get_urls(manifest)
1167 for format_url in formats_urls:
1168 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1169 url_map[itag] = format_url
1172 def _extract_annotations(self, video_id):
1173 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1174 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1176 def _real_extract(self, url):
1177 url, smuggled_data = unsmuggle_url(url, {})
1180 'http' if self._downloader.params.get('prefer_insecure', False)
1185 parsed_url = compat_urllib_parse_urlparse(url)
1186 for component in [parsed_url.fragment, parsed_url.query]:
1187 query = compat_parse_qs(component)
1188 if start_time is None and 't' in query:
1189 start_time = parse_duration(query['t'][0])
1190 if start_time is None and 'start' in query:
1191 start_time = parse_duration(query['start'][0])
1192 if end_time is None and 'end' in query:
1193 end_time = parse_duration(query['end'][0])
1195 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1196 mobj = re.search(self._NEXT_URL_RE, url)
1198 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1199 video_id = self.extract_id(url)
1202 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1203 video_webpage = self._download_webpage(url, video_id)
1205 # Attempt to extract SWF player URL
1206 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1207 if mobj is not None:
1208 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1214 def add_dash_mpd(video_info):
1215 dash_mpd = video_info.get('dashmpd')
1216 if dash_mpd and dash_mpd[0] not in dash_mpds:
1217 dash_mpds.append(dash_mpd[0])
1220 embed_webpage = None
1222 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1224 # We simulate the access to the video from www.youtube.com/v/{video_id}
1225 # this can be viewed without login into Youtube
1226 url = proto + '://www.youtube.com/embed/%s' % video_id
1227 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1228 data = compat_urllib_parse.urlencode({
1229 'video_id': video_id,
1230 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1231 'sts': self._search_regex(
1232 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1234 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1235 video_info_webpage = self._download_webpage(
1236 video_info_url, video_id,
1237 note='Refetching age-gated info webpage',
1238 errnote='unable to download video info webpage')
1239 video_info = compat_parse_qs(video_info_webpage)
1240 add_dash_mpd(video_info)
1244 # Try looking directly into the video webpage
1245 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1247 args = ytplayer_config['args']
1248 if args.get('url_encoded_fmt_stream_map'):
1249 # Convert to the same format returned by compat_parse_qs
1250 video_info = dict((k, [v]) for k, v in args.items())
1251 add_dash_mpd(video_info)
1252 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1254 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1255 # We also try looking in get_video_info since it may contain different dashmpd
1256 # URL that points to a DASH manifest with possibly different itag set (some itags
1257 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1258 # manifest pointed by get_video_info's dashmpd).
1259 # The general idea is to take a union of itags of both DASH manifests (for example
1260 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1261 self.report_video_info_webpage_download(video_id)
1262 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1264 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1265 % (proto, video_id, el_type))
1266 video_info_webpage = self._download_webpage(
1268 video_id, note=False,
1269 errnote='unable to download video info webpage')
1270 get_video_info = compat_parse_qs(video_info_webpage)
1271 if get_video_info.get('use_cipher_signature') != ['True']:
1272 add_dash_mpd(get_video_info)
1274 video_info = get_video_info
1275 if 'token' in get_video_info:
1276 # Different get_video_info requests may report different results, e.g.
1277 # some may report video unavailability, but some may serve it without
1278 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1279 # the original webpage as well as el=info and el=embedded get_video_info
1280 # requests report video unavailability due to geo restriction while
1281 # el=detailpage succeeds and returns valid data). This is probably
1282 # due to YouTube measures against IP ranges of hosting providers.
1283 # Working around by preferring the first succeeded video_info containing
1284 # the token if no such video_info yet was found.
1285 if 'token' not in video_info:
1286 video_info = get_video_info
1288 if 'token' not in video_info:
1289 if 'reason' in video_info:
1290 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1291 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1293 raise ExtractorError('YouTube said: This video is available in %s only' % (
1294 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1296 raise ExtractorError(
1297 'YouTube said: %s' % video_info['reason'][0],
1298 expected=True, video_id=video_id)
1300 raise ExtractorError(
1301 '"token" parameter not in video info for unknown reason',
1305 if 'title' in video_info:
1306 video_title = video_info['title'][0]
1308 self._downloader.report_warning('Unable to extract video title')
1312 video_description = get_element_by_id("eow-description", video_webpage)
1313 if video_description:
1314 video_description = re.sub(r'''(?x)
1316 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1317 (?:title|href)="([^"]+)"\s+
1318 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1319 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1322 ''', r'\1', video_description)
1323 video_description = clean_html(video_description)
1325 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1327 video_description = unescapeHTML(fd_mobj.group(1))
1329 video_description = ''
1331 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1332 if not self._downloader.params.get('noplaylist'):
1335 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1336 for feed in multifeed_metadata_list.split(','):
1337 # Unquote should take place before split on comma (,) since textual
1338 # fields may contain comma as well (see
1339 # https://github.com/rg3/youtube-dl/issues/8536)
1340 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1342 '_type': 'url_transparent',
1343 'ie_key': 'Youtube',
1345 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1346 {'force_singlefeed': True}),
1347 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1349 feed_ids.append(feed_data['id'][0])
1351 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1352 % (', '.join(feed_ids), video_id))
1353 return self.playlist_result(entries, video_id, video_title, video_description)
1354 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1356 if 'view_count' in video_info:
1357 view_count = int(video_info['view_count'][0])
1361 # Check for "rental" videos
1362 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1363 raise ExtractorError('"rental" videos not supported')
1365 # Start extracting information
1366 self.report_information_extraction(video_id)
1369 if 'author' not in video_info:
1370 raise ExtractorError('Unable to extract uploader name')
1371 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1374 video_uploader_id = None
1375 video_uploader_url = None
1377 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1379 if mobj is not None:
1380 video_uploader_id = mobj.group('uploader_id')
1381 video_uploader_url = mobj.group('uploader_url')
1383 self._downloader.report_warning('unable to extract uploader nickname')
1386 # We try first to get a high quality image:
1387 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1388 video_webpage, re.DOTALL)
1389 if m_thumb is not None:
1390 video_thumbnail = m_thumb.group(1)
1391 elif 'thumbnail_url' not in video_info:
1392 self._downloader.report_warning('unable to extract video thumbnail')
1393 video_thumbnail = None
1394 else: # don't panic if we can't find it
1395 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1398 upload_date = self._html_search_meta(
1399 'datePublished', video_webpage, 'upload date', default=None)
1401 upload_date = self._search_regex(
1402 [r'(?s)id="eow-date.*?>(.*?)</span>',
1403 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1404 video_webpage, 'upload date', default=None)
1406 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1407 upload_date = unified_strdate(upload_date)
1409 video_license = self._html_search_regex(
1410 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1411 video_webpage, 'license', default=None)
1413 m_music = re.search(
1414 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1417 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1418 video_creator = clean_html(m_music.group('creator'))
1420 video_alt_title = video_creator = None
1422 m_cat_container = self._search_regex(
1423 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1424 video_webpage, 'categories', default=None)
1426 category = self._html_search_regex(
1427 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1429 video_categories = None if category is None else [category]
1431 video_categories = None
1434 unescapeHTML(m.group('content'))
1435 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1437 def _extract_count(count_name):
1438 return str_to_int(self._search_regex(
1439 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1440 % re.escape(count_name),
1441 video_webpage, count_name, default=None))
1443 like_count = _extract_count('like')
1444 dislike_count = _extract_count('dislike')
1447 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1448 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1450 if 'length_seconds' not in video_info:
1451 self._downloader.report_warning('unable to extract video duration')
1452 video_duration = None
1454 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1457 video_annotations = None
1458 if self._downloader.params.get('writeannotations', False):
1459 video_annotations = self._extract_annotations(video_id)
1461 def _map_to_format_list(urlmap):
1463 for itag, video_real_url in urlmap.items():
1466 'url': video_real_url,
1467 'player_url': player_url,
1469 if itag in self._formats:
1470 dct.update(self._formats[itag])
1474 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1475 self.report_rtmp_download()
1477 'format_id': '_rtmp',
1479 'url': video_info['conn'][0],
1480 'player_url': player_url,
1482 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1483 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1484 if 'rtmpe%3Dyes' in encoded_url_map:
1485 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1487 fmt_list = video_info.get('fmt_list', [''])[0]
1489 for fmt in fmt_list.split(','):
1490 spec = fmt.split('/')
1492 width_height = spec[1].split('x')
1493 if len(width_height) == 2:
1494 formats_spec[spec[0]] = {
1495 'resolution': spec[1],
1496 'width': int_or_none(width_height[0]),
1497 'height': int_or_none(width_height[1]),
1500 for url_data_str in encoded_url_map.split(','):
1501 url_data = compat_parse_qs(url_data_str)
1502 if 'itag' not in url_data or 'url' not in url_data:
1504 format_id = url_data['itag'][0]
1505 url = url_data['url'][0]
1507 if 'sig' in url_data:
1508 url += '&signature=' + url_data['sig'][0]
1509 elif 's' in url_data:
1510 encrypted_sig = url_data['s'][0]
1511 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1513 jsplayer_url_json = self._search_regex(
1515 embed_webpage if age_gate else video_webpage,
1516 'JS player URL (1)', default=None)
1517 if not jsplayer_url_json and not age_gate:
1518 # We need the embed website after all
1519 if embed_webpage is None:
1520 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1521 embed_webpage = self._download_webpage(
1522 embed_url, video_id, 'Downloading embed webpage')
1523 jsplayer_url_json = self._search_regex(
1524 ASSETS_RE, embed_webpage, 'JS player URL')
1526 player_url = json.loads(jsplayer_url_json)
1527 if player_url is None:
1528 player_url_json = self._search_regex(
1529 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1530 video_webpage, 'age gate player URL')
1531 player_url = json.loads(player_url_json)
1533 if self._downloader.params.get('verbose'):
1534 if player_url is None:
1535 player_version = 'unknown'
1536 player_desc = 'unknown'
1538 if player_url.endswith('swf'):
1539 player_version = self._search_regex(
1540 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1541 'flash player', fatal=False)
1542 player_desc = 'flash player %s' % player_version
1544 player_version = self._search_regex(
1545 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1547 'html5 player', fatal=False)
1548 player_desc = 'html5 player %s' % player_version
1550 parts_sizes = self._signature_cache_id(encrypted_sig)
1551 self.to_screen('{%s} signature length %s, %s' %
1552 (format_id, parts_sizes, player_desc))
1554 signature = self._decrypt_signature(
1555 encrypted_sig, video_id, player_url, age_gate)
1556 url += '&signature=' + signature
1557 if 'ratebypass' not in url:
1558 url += '&ratebypass=yes'
1561 'format_id': format_id,
1563 'player_url': player_url,
1565 if format_id in self._formats:
1566 dct.update(self._formats[format_id])
1567 if format_id in formats_spec:
1568 dct.update(formats_spec[format_id])
1570 # Some itags are not included in DASH manifest thus corresponding formats will
1571 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1572 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1573 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1574 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1577 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1578 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1581 'fps': int_or_none(url_data.get('fps', [None])[0]),
1582 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1584 for key, value in more_fields.items():
1587 type_ = url_data.get('type', [None])[0]
1589 type_split = type_.split(';')
1590 kind_ext = type_split[0].split('/')
1591 if len(kind_ext) == 2:
1593 dct['ext'] = mimetype2ext(type_split[0])
1594 if kind in ('audio', 'video'):
1596 for mobj in re.finditer(
1597 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1598 if mobj.group('key') == 'codecs':
1599 codecs = mobj.group('val')
1602 codecs = codecs.split(',')
1603 if len(codecs) == 2:
1604 acodec, vcodec = codecs[1], codecs[0]
1606 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1612 elif video_info.get('hlsvp'):
1613 manifest_url = video_info['hlsvp'][0]
1614 url_map = self._extract_from_m3u8(manifest_url, video_id)
1615 formats = _map_to_format_list(url_map)
1616 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1617 for a_format in formats:
1618 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1620 unavailable_message = self._html_search_regex(
1621 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1622 video_webpage, 'unavailable message', default=None)
1623 if unavailable_message:
1624 raise ExtractorError(unavailable_message, expected=True)
1625 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1627 # Look for the DASH manifest
1628 if self._downloader.params.get('youtube_include_dash_manifest', True):
1629 dash_mpd_fatal = True
1630 for mpd_url in dash_mpds:
1633 def decrypt_sig(mobj):
1635 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1636 return '/signature/%s' % dec_s
1638 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1640 for df in self._extract_mpd_formats(
1641 mpd_url, video_id, fatal=dash_mpd_fatal,
1642 formats_dict=self._formats):
1643 # Do not overwrite DASH format found in some previous DASH manifest
1644 if df['format_id'] not in dash_formats:
1645 dash_formats[df['format_id']] = df
1646 # Additional DASH manifests may end up in HTTP Error 403 therefore
1647 # allow them to fail without bug report message if we already have
1648 # some DASH manifest succeeded. This is temporary workaround to reduce
1649 # burst of bug reports until we figure out the reason and whether it
1650 # can be fixed at all.
1651 dash_mpd_fatal = False
1652 except (ExtractorError, KeyError) as e:
1653 self.report_warning(
1654 'Skipping DASH manifest: %r' % e, video_id)
1656 # Remove the formats we found through non-DASH, they
1657 # contain less info and it can be wrong, because we use
1658 # fixed values (for example the resolution). See
1659 # https://github.com/rg3/youtube-dl/issues/5774 for an
1661 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1662 formats.extend(dash_formats.values())
1664 # Check for malformed aspect ratio
1665 stretched_m = re.search(
1666 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1669 w = float(stretched_m.group('w'))
1670 h = float(stretched_m.group('h'))
1671 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1672 # We will only process correct ratios.
1676 if f.get('vcodec') != 'none':
1677 f['stretched_ratio'] = ratio
1679 self._sort_formats(formats)
1681 self.mark_watched(video_id, video_info)
1685 'uploader': video_uploader,
1686 'uploader_id': video_uploader_id,
1687 'uploader_url': video_uploader_url,
1688 'upload_date': upload_date,
1689 'license': video_license,
1690 'creator': video_creator,
1691 'title': video_title,
1692 'alt_title': video_alt_title,
1693 'thumbnail': video_thumbnail,
1694 'description': video_description,
1695 'categories': video_categories,
1697 'subtitles': video_subtitles,
1698 'automatic_captions': automatic_captions,
1699 'duration': video_duration,
1700 'age_limit': 18 if age_gate else 0,
1701 'annotations': video_annotations,
1702 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1703 'view_count': view_count,
1704 'like_count': like_count,
1705 'dislike_count': dislike_count,
1706 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1709 'start_time': start_time,
1710 'end_time': end_time,
1714 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1715 IE_DESC = 'YouTube.com playlists'
1716 _VALID_URL = r"""(?x)(?:
1721 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1722 \? (?:.*?[&;])*? (?:p|a|list)=
1726 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1727 # Top tracks, they can also include dots
1732 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1734 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1735 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1736 IE_NAME = 'youtube:playlist'
1738 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1740 'title': 'ytdl test PL',
1741 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1743 'playlist_count': 3,
1745 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1747 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1748 'title': 'YDL_Empty_List',
1750 'playlist_count': 0,
1752 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1753 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1755 'title': '29C3: Not my department',
1756 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1758 'playlist_count': 95,
1760 'note': 'issue #673',
1761 'url': 'PLBB231211A4F62143',
1763 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1764 'id': 'PLBB231211A4F62143',
1766 'playlist_mincount': 26,
1768 'note': 'Large playlist',
1769 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1771 'title': 'Uploads from Cauchemar',
1772 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1774 'playlist_mincount': 799,
1776 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1778 'title': 'YDL_safe_search',
1779 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1781 'playlist_count': 2,
1784 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1785 'playlist_count': 4,
1788 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1791 'note': 'Embedded SWF player',
1792 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1793 'playlist_count': 4,
1796 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1799 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1800 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1802 'title': 'Uploads from Interstellar Movie',
1803 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1805 'playlist_mincout': 21,
1808 def _real_initialize(self):
1811 def _extract_mix(self, playlist_id):
1812 # The mixes are generated from a single video
1813 # the id of the playlist is just 'RD' + video_id
1814 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1815 webpage = self._download_webpage(
1816 url, playlist_id, 'Downloading Youtube mix')
1817 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1819 search_title('playlist-title') or
1820 search_title('title long-title') or
1821 search_title('title'))
1822 title = clean_html(title_span)
1823 ids = orderedSet(re.findall(
1824 r'''(?xs)data-video-username=".*?".*?
1825 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1827 url_results = self._ids_to_results(ids)
1829 return self.playlist_result(url_results, playlist_id, title)
1831 def _extract_playlist(self, playlist_id):
1832 url = self._TEMPLATE_URL % playlist_id
1833 page = self._download_webpage(url, playlist_id)
1835 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1836 match = match.strip()
1837 # Check if the playlist exists or is private
1838 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1839 raise ExtractorError(
1840 'The playlist doesn\'t exist or is private, use --username or '
1841 '--netrc to access it.',
1843 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1844 raise ExtractorError(
1845 'Invalid parameters. Maybe URL is incorrect.',
1847 elif re.match(r'[^<]*Choose your language[^<]*', match):
1850 self.report_warning('Youtube gives an alert message: ' + match)
1852 playlist_title = self._html_search_regex(
1853 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1856 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1858 def _check_download_just_video(self, url, playlist_id):
1859 # Check if it's a video-specific URL
1860 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1861 if 'v' in query_dict:
1862 video_id = query_dict['v'][0]
1863 if self._downloader.params.get('noplaylist'):
1864 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1865 return self.url_result(video_id, 'Youtube', video_id=video_id)
1867 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1869 def _real_extract(self, url):
1870 # Extract playlist id
1871 mobj = re.match(self._VALID_URL, url)
1873 raise ExtractorError('Invalid URL: %s' % url)
1874 playlist_id = mobj.group(1) or mobj.group(2)
1876 video = self._check_download_just_video(url, playlist_id)
1880 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1881 # Mixes require a custom extraction process
1882 return self._extract_mix(playlist_id)
1884 return self._extract_playlist(playlist_id)
1887 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1888 IE_DESC = 'YouTube.com channels'
1889 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1890 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1891 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1892 IE_NAME = 'youtube:channel'
1894 'note': 'paginated channel',
1895 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1896 'playlist_mincount': 91,
1898 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1899 'title': 'Uploads from lex will',
1902 'note': 'Age restricted channel',
1903 # from https://www.youtube.com/user/DeusExOfficial
1904 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1905 'playlist_mincount': 64,
1907 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1908 'title': 'Uploads from Deus Ex',
1913 def suitable(cls, url):
1914 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1915 else super(YoutubeChannelIE, cls).suitable(url))
1917 def _real_extract(self, url):
1918 channel_id = self._match_id(url)
1920 url = self._TEMPLATE_URL % channel_id
1922 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1923 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1924 # otherwise fallback on channel by page extraction
1925 channel_page = self._download_webpage(
1926 url + '?view=57', channel_id,
1927 'Downloading channel page', fatal=False)
1928 if channel_page is False:
1929 channel_playlist_id = False
1931 channel_playlist_id = self._html_search_meta(
1932 'channelId', channel_page, 'channel id', default=None)
1933 if not channel_playlist_id:
1934 channel_playlist_id = self._search_regex(
1935 r'data-(?:channel-external-|yt)id="([^"]+)"',
1936 channel_page, 'channel id', default=None)
1937 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1938 playlist_id = 'UU' + channel_playlist_id[2:]
1939 return self.url_result(
1940 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1942 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1943 autogenerated = re.search(r'''(?x)
1945 channel-header-autogenerated-label|
1946 yt-channel-title-autogenerated
1947 )[^"]*"''', channel_page) is not None
1950 # The videos are contained in a single page
1951 # the ajax pages can't be used, they are empty
1954 video_id, 'Youtube', video_id=video_id,
1955 video_title=video_title)
1956 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1957 return self.playlist_result(entries, channel_id)
1959 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1962 class YoutubeUserIE(YoutubeChannelIE):
1963 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1964 _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_-]+)'
1965 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1966 IE_NAME = 'youtube:user'
1969 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1970 'playlist_mincount': 320,
1972 'title': 'TheLinuxFoundation',
1975 'url': 'ytuser:phihag',
1976 'only_matching': True,
1980 def suitable(cls, url):
1981 # Don't return True if the url can be extracted with other youtube
1982 # extractor, the regex would is too permissive and it would match.
1983 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1984 if any(ie.suitable(url) for ie in other_ies):
1987 return super(YoutubeUserIE, cls).suitable(url)
1990 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
1991 IE_DESC = 'YouTube.com live streams'
1992 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
1993 IE_NAME = 'youtube:live'
1996 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
1998 'id': 'a48o2S1cPoo',
2000 'title': 'The Young Turks - Live Main Show',
2001 'uploader': 'The Young Turks',
2002 'uploader_id': 'TheYoungTurks',
2003 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2004 'upload_date': '20150715',
2005 'license': 'Standard YouTube License',
2006 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2007 'categories': ['News & Politics'],
2008 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2010 'dislike_count': int,
2013 'skip_download': True,
2016 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2017 'only_matching': True,
2020 def _real_extract(self, url):
2021 mobj = re.match(self._VALID_URL, url)
2022 channel_id = mobj.group('id')
2023 base_url = mobj.group('base_url')
2024 webpage = self._download_webpage(url, channel_id, fatal=False)
2026 page_type = self._og_search_property(
2027 'type', webpage, 'page type', default=None)
2028 video_id = self._html_search_meta(
2029 'videoId', webpage, 'video id', default=None)
2030 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2031 return self.url_result(video_id, YoutubeIE.ie_key())
2032 return self.url_result(base_url)
2035 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2036 IE_DESC = 'YouTube.com user/channel playlists'
2037 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2038 IE_NAME = 'youtube:playlists'
2041 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2042 'playlist_mincount': 4,
2044 'id': 'ThirstForScience',
2045 'title': 'Thirst for Science',
2048 # with "Load more" button
2049 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2050 'playlist_mincount': 70,
2053 'title': 'Игорь Клейнер',
2056 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2057 'playlist_mincount': 17,
2059 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2060 'title': 'Chem Player',
2065 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2066 IE_DESC = 'YouTube.com searches'
2067 # there doesn't appear to be a real limit, for example if you search for
2068 # 'python' you get more than 8.000.000 results
2069 _MAX_RESULTS = float('inf')
2070 IE_NAME = 'youtube:search'
2071 _SEARCH_KEY = 'ytsearch'
2072 _EXTRA_QUERY_ARGS = {}
2075 def _get_n_results(self, query, n):
2076 """Get a specified number of results for a query"""
2081 for pagenum in itertools.count(1):
2083 'search_query': query.encode('utf-8'),
2087 url_query.update(self._EXTRA_QUERY_ARGS)
2088 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
2089 data = self._download_json(
2090 result_url, video_id='query "%s"' % query,
2091 note='Downloading page %s' % pagenum,
2092 errnote='Unable to download API page')
2093 html_content = data[1]['body']['content']
2095 if 'class="search-message' in html_content:
2096 raise ExtractorError(
2097 '[youtube] No video results', expected=True)
2099 new_videos = self._ids_to_results(orderedSet(re.findall(
2100 r'href="/watch\?v=(.{11})', html_content)))
2101 videos += new_videos
2102 if not new_videos or len(videos) > limit:
2107 return self.playlist_result(videos, query)
2110 class YoutubeSearchDateIE(YoutubeSearchIE):
2111 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2112 _SEARCH_KEY = 'ytsearchdate'
2113 IE_DESC = 'YouTube.com searches, newest videos first'
2114 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2117 class YoutubeSearchURLIE(InfoExtractor):
2118 IE_DESC = 'YouTube.com search URLs'
2119 IE_NAME = 'youtube:search_url'
2120 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2122 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2123 'playlist_mincount': 5,
2125 'title': 'youtube-dl test video',
2128 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2129 'only_matching': True,
2132 def _real_extract(self, url):
2133 mobj = re.match(self._VALID_URL, url)
2134 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2136 webpage = self._download_webpage(url, query)
2137 result_code = self._search_regex(
2138 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2140 part_codes = re.findall(
2141 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2143 for part_code in part_codes:
2144 part_title = self._html_search_regex(
2145 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2146 part_url_snippet = self._html_search_regex(
2147 r'(?s)href="([^"]+)"', part_code, 'item URL')
2148 part_url = compat_urlparse.urljoin(
2149 'https://www.youtube.com/', part_url_snippet)
2153 'title': part_title,
2157 '_type': 'playlist',
2163 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2164 IE_DESC = 'YouTube.com (multi-season) shows'
2165 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2166 IE_NAME = 'youtube:show'
2168 'url': 'https://www.youtube.com/show/airdisasters',
2169 'playlist_mincount': 5,
2171 'id': 'airdisasters',
2172 'title': 'Air Disasters',
2176 def _real_extract(self, url):
2177 playlist_id = self._match_id(url)
2178 return super(YoutubeShowIE, self)._real_extract(
2179 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2182 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2184 Base class for feed extractors
2185 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2187 _LOGIN_REQUIRED = True
2191 return 'youtube:%s' % self._FEED_NAME
2193 def _real_initialize(self):
2196 def _real_extract(self, url):
2197 page = self._download_webpage(
2198 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2200 # The extraction process is the same as for playlists, but the regex
2201 # for the video ids doesn't contain an index
2203 more_widget_html = content_html = page
2204 for page_num in itertools.count(1):
2205 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2207 # 'recommended' feed has infinite 'load more' and each new portion spins
2208 # the same videos in (sometimes) slightly different order, so we'll check
2209 # for unicity and break when portion has no new videos
2210 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2216 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2220 more = self._download_json(
2221 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2222 'Downloading page #%s' % page_num,
2223 transform_source=uppercase_escape)
2224 content_html = more['content_html']
2225 more_widget_html = more['load_more_widget_html']
2227 return self.playlist_result(
2228 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2231 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2232 IE_NAME = 'youtube:watchlater'
2233 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2234 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2237 'url': 'https://www.youtube.com/playlist?list=WL',
2238 'only_matching': True,
2240 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2241 'only_matching': True,
2244 def _real_extract(self, url):
2245 video = self._check_download_just_video(url, 'WL')
2248 return self._extract_playlist('WL')
2251 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2252 IE_NAME = 'youtube:favorites'
2253 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2254 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2255 _LOGIN_REQUIRED = True
2257 def _real_extract(self, url):
2258 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2259 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2260 return self.url_result(playlist_id, 'YoutubePlaylist')
2263 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2264 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2265 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2266 _FEED_NAME = 'recommended'
2267 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2270 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2271 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2272 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2273 _FEED_NAME = 'subscriptions'
2274 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2277 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2278 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2279 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2280 _FEED_NAME = 'history'
2281 _PLAYLIST_TITLE = 'Youtube History'
2284 class YoutubeTruncatedURLIE(InfoExtractor):
2285 IE_NAME = 'youtube:truncated_url'
2286 IE_DESC = False # Do not list
2287 _VALID_URL = r'''(?x)
2289 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2292 annotation_id=annotation_[^&]+|
2298 attribution_link\?a=[^&]+
2304 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2305 'only_matching': True,
2307 'url': 'http://www.youtube.com/watch?',
2308 'only_matching': True,
2310 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2311 'only_matching': True,
2313 'url': 'https://www.youtube.com/watch?feature=foo',
2314 'only_matching': True,
2316 'url': 'https://www.youtube.com/watch?hl=en-GB',
2317 'only_matching': True,
2319 'url': 'https://www.youtube.com/watch?t=2372',
2320 'only_matching': True,
2323 def _real_extract(self, url):
2324 raise ExtractorError(
2325 'Did you forget to quote the URL? Remember that & is a meta '
2326 'character in most shells, so you want to put the URL in quotes, '
2328 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2329 ' or simply youtube-dl BaW_jenozKc .',
2333 class YoutubeTruncatedIDIE(InfoExtractor):
2334 IE_NAME = 'youtube:truncated_id'
2335 IE_DESC = False # Do not list
2336 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2339 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2340 'only_matching': True,
2343 def _real_extract(self, url):
2344 video_id = self._match_id(url)
2345 raise ExtractorError(
2346 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),