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 (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
32 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 = urlencode_postdata(login_form_strs)
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 error_msg = self._html_search_regex(
129 r'<[^>]+id="errormsg_0_Passwd"[^>]*>([^<]+)<',
130 login_results, 'error message', default=None)
132 raise ExtractorError('Unable to login: %s' % error_msg, expected=True)
134 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
135 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
138 # TODO add SMS and phone call support - these require making a request and then prompting the user
140 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
141 tfa_code = self._get_tfa_info('2-step verification code')
144 self._downloader.report_warning(
145 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
146 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
149 tfa_code = remove_start(tfa_code, 'G-')
151 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
153 tfa_form_strs.update({
158 tfa_data = urlencode_postdata(tfa_form_strs)
160 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
161 tfa_results = self._download_webpage(
163 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
165 if tfa_results is False:
168 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
169 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
171 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
172 self._downloader.report_warning('unable to log in - did the page structure change?')
174 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
175 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
178 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
179 self._downloader.report_warning('unable to log in: bad username or password')
183 def _real_initialize(self):
184 if self._downloader is None:
187 if not self._login():
191 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
192 # Extract entries from page with "Load more" button
193 def _entries(self, page, playlist_id):
194 more_widget_html = content_html = page
195 for page_num in itertools.count(1):
196 for entry in self._process_page(content_html):
199 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
203 more = self._download_json(
204 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
205 'Downloading page #%s' % page_num,
206 transform_source=uppercase_escape)
207 content_html = more['content_html']
208 if not content_html.strip():
209 # Some webpages show a "Load more" button but they don't
212 more_widget_html = more['load_more_widget_html']
215 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
216 def _process_page(self, content):
217 for video_id, video_title in self.extract_videos_from_page(content):
218 yield self.url_result(video_id, 'Youtube', video_id, video_title)
220 def extract_videos_from_page(self, page):
223 for mobj in re.finditer(self._VIDEO_RE, page):
224 # The link with index 0 is not the first video of the playlist (not sure if still actual)
225 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
227 video_id = mobj.group('id')
228 video_title = unescapeHTML(mobj.group('title'))
230 video_title = video_title.strip()
232 idx = ids_in_page.index(video_id)
233 if video_title and not titles_in_page[idx]:
234 titles_in_page[idx] = video_title
236 ids_in_page.append(video_id)
237 titles_in_page.append(video_title)
238 return zip(ids_in_page, titles_in_page)
241 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
242 def _process_page(self, content):
243 for playlist_id in orderedSet(re.findall(
244 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
246 yield self.url_result(
247 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
249 def _real_extract(self, url):
250 playlist_id = self._match_id(url)
251 webpage = self._download_webpage(url, playlist_id)
252 title = self._og_search_title(webpage, fatal=False)
253 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
256 class YoutubeIE(YoutubeBaseInfoExtractor):
257 IE_DESC = 'YouTube.com'
258 _VALID_URL = r"""(?x)^
260 (?:https?://|//) # http(s):// or protocol-independent URL
261 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
262 (?:www\.)?deturl\.com/www\.youtube\.com/|
263 (?:www\.)?pwnyoutube\.com/|
264 (?:www\.)?yourepeat\.com/|
265 tube\.majestyc\.net/|
266 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
267 (?:.*?\#/)? # handle anchor (#/) redirect urls
268 (?: # the various things that can precede the ID:
269 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
270 |(?: # or the v= param in all its forms
271 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
272 (?:\?|\#!?) # the params delimiter ? or # or #!
273 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
278 youtu\.be| # just youtu.be/xxxx
279 vid\.plus| # or vid.plus/xxxx
280 zwearz\.com/watch| # or zwearz.com/watch/xxxx
282 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
284 )? # all until now is optional -> you can pass the naked ID
285 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
286 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
287 (?(1).+)? # if we found the ID, everything can follow
289 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
291 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
292 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
293 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
294 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
295 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
296 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
297 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
298 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
300 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
301 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
302 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
303 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
304 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
305 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
306 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
307 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
308 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
312 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
313 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
314 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
315 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
316 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
317 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
318 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
320 # Apple HTTP Live Streaming
321 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
322 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
323 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
324 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
325 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
326 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
327 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
328 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
331 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
332 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
333 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
334 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
335 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
336 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
337 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
338 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
339 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
340 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
341 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
344 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
345 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
346 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
349 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
350 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
351 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
352 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
353 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
354 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
355 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
356 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
357 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
358 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
359 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
361 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
362 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
363 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
364 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
365 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
366 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
367 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
368 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
369 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
370 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
373 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
374 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
376 # Dash webm audio with opus inside
377 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
378 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
379 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
382 '_rtmp': {'protocol': 'rtmp'},
384 _SUBTITLE_FORMATS = ('ttml', 'vtt')
389 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
393 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
394 'uploader': 'Philipp Hagemeister',
395 'uploader_id': 'phihag',
396 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
397 'upload_date': '20121002',
398 'license': 'Standard YouTube License',
399 '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 .',
400 'categories': ['Science & Technology'],
401 'tags': ['youtube-dl'],
403 'dislike_count': int,
409 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
410 'note': 'Test generic use_cipher_signature video (#897)',
414 'upload_date': '20120506',
415 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
416 'alt_title': 'I Love It (feat. Charli XCX)',
417 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
418 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
419 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
420 'iconic ep', 'iconic', 'love', 'it'],
421 'uploader': 'Icona Pop',
422 'uploader_id': 'IconaPop',
423 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
424 'license': 'Standard YouTube License',
425 'creator': 'Icona Pop',
429 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
430 'note': 'Test VEVO video with age protection (#956)',
434 'upload_date': '20130703',
435 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
436 'alt_title': 'Tunnel Vision',
437 'description': 'md5:64249768eec3bc4276236606ea996373',
438 'uploader': 'justintimberlakeVEVO',
439 'uploader_id': 'justintimberlakeVEVO',
440 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
441 'license': 'Standard YouTube License',
442 'creator': 'Justin Timberlake',
447 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
448 'note': 'Embed-only video (#1746)',
452 'upload_date': '20120608',
453 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
454 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
455 'uploader': 'SET India',
456 'uploader_id': 'setindia',
457 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
458 'license': 'Standard YouTube License',
463 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
464 'note': 'Use the first video ID in the URL',
468 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
469 'uploader': 'Philipp Hagemeister',
470 'uploader_id': 'phihag',
471 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
472 'upload_date': '20121002',
473 'license': 'Standard YouTube License',
474 '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 .',
475 'categories': ['Science & Technology'],
476 'tags': ['youtube-dl'],
478 'dislike_count': int,
481 'skip_download': True,
485 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
486 'note': '256k DASH audio (format 141) via DASH manifest',
490 'upload_date': '20121002',
491 'uploader_id': '8KVIDEO',
492 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
494 'uploader': '8KVIDEO',
495 'license': 'Standard YouTube License',
496 'title': 'UHDTV TEST 8K VIDEO.mp4'
499 'youtube_include_dash_manifest': True,
503 # DASH manifest with encrypted signature
505 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
509 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
510 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
511 'uploader': 'AfrojackVEVO',
512 'uploader_id': 'AfrojackVEVO',
513 'upload_date': '20131011',
514 'license': 'Standard YouTube License',
517 'youtube_include_dash_manifest': True,
521 # JS player signature function name containing $
523 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
527 'title': 'Taylor Swift - Shake It Off',
528 'alt_title': 'Shake It Off',
529 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
530 'uploader': 'TaylorSwiftVEVO',
531 'uploader_id': 'TaylorSwiftVEVO',
532 'upload_date': '20140818',
533 'license': 'Standard YouTube License',
534 'creator': 'Taylor Swift',
537 'youtube_include_dash_manifest': True,
543 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
547 'upload_date': '20100909',
548 'uploader': 'The Amazing Atheist',
549 'uploader_id': 'TheAmazingAtheist',
550 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
551 'license': 'Standard YouTube License',
552 'title': 'Burning Everyone\'s Koran',
553 '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',
556 # Normal age-gate video (No vevo, embed allowed)
558 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
562 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
563 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
564 'uploader': 'The Witcher',
565 'uploader_id': 'WitcherGame',
566 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
567 'upload_date': '20140605',
568 'license': 'Standard YouTube License',
572 # Age-gate video with encrypted signature
574 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
578 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
579 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
580 'uploader': 'LloydVEVO',
581 'uploader_id': 'LloydVEVO',
582 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
583 'upload_date': '20110629',
584 'license': 'Standard YouTube License',
588 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
590 'url': '__2ABJjxzNo',
594 'upload_date': '20100430',
595 'uploader_id': 'deadmau5',
596 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
597 'creator': 'deadmau5',
598 'description': 'md5:12c56784b8032162bb936a5f76d55360',
599 'uploader': 'deadmau5',
600 'license': 'Standard YouTube License',
601 'title': 'Deadmau5 - Some Chords (HD)',
602 'alt_title': 'Some Chords',
604 'expected_warnings': [
605 'DASH manifest missing',
608 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
610 'url': 'lqQg6PlCWgI',
614 'upload_date': '20150827',
615 'uploader_id': 'olympic',
616 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
617 'license': 'Standard YouTube License',
618 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
619 'uploader': 'Olympics',
620 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
623 'skip_download': 'requires avconv',
628 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
632 'stretched_ratio': 16 / 9.,
633 'upload_date': '20110310',
634 'uploader_id': 'AllenMeow',
635 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
636 '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',
638 'license': 'Standard YouTube License',
639 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
642 # url_encoded_fmt_stream_map is empty string
644 'url': 'qEJwOuvDf7I',
648 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
650 'upload_date': '20150404',
651 'uploader_id': 'spbelect',
652 'uploader': 'Наблюдатели Петербурга',
655 'skip_download': 'requires avconv',
657 'skip': 'This live event has ended.',
659 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
661 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
665 'title': 'md5:7b81415841e02ecd4313668cde88737a',
666 'description': 'md5:116377fd2963b81ec4ce64b542173306',
667 'upload_date': '20150625',
668 'uploader_id': 'dorappi2000',
669 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
670 'uploader': 'dorappi2000',
671 'license': 'Standard YouTube License',
672 'formats': 'mincount:33',
675 # DASH manifest with segment_list
677 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
678 'md5': '8ce563a1d667b599d21064e982ab9e31',
682 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
683 'uploader': 'Airtek',
684 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
685 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
686 'license': 'Standard YouTube License',
687 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
690 'youtube_include_dash_manifest': True,
691 'format': '135', # bestvideo
695 # Multifeed videos (multiple cameras), URL is for Main Camera
696 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
699 'title': 'teamPGP: Rocket League Noob Stream',
700 'description': 'md5:dc7872fb300e143831327f1bae3af010',
706 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
707 'description': 'md5:dc7872fb300e143831327f1bae3af010',
708 'upload_date': '20150721',
709 'uploader': 'Beer Games Beer',
710 'uploader_id': 'beergamesbeer',
711 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
712 'license': 'Standard YouTube License',
718 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
719 'description': 'md5:dc7872fb300e143831327f1bae3af010',
720 'upload_date': '20150721',
721 'uploader': 'Beer Games Beer',
722 'uploader_id': 'beergamesbeer',
723 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
724 'license': 'Standard YouTube License',
730 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
731 'description': 'md5:dc7872fb300e143831327f1bae3af010',
732 'upload_date': '20150721',
733 'uploader': 'Beer Games Beer',
734 'uploader_id': 'beergamesbeer',
735 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
736 'license': 'Standard YouTube License',
742 'title': 'teamPGP: Rocket League Noob Stream (zim)',
743 'description': 'md5:dc7872fb300e143831327f1bae3af010',
744 'upload_date': '20150721',
745 'uploader': 'Beer Games Beer',
746 'uploader_id': 'beergamesbeer',
747 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
748 'license': 'Standard YouTube License',
752 'skip_download': True,
756 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
757 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
760 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
765 'url': 'http://vid.plus/FlRa-iH7PGw',
766 'only_matching': True,
769 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
770 'only_matching': True,
773 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
774 # Also tests cut-off URL expansion in video description (see
775 # https://github.com/rg3/youtube-dl/issues/1892,
776 # https://github.com/rg3/youtube-dl/issues/8164)
777 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
781 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
782 'alt_title': 'Dark Walk',
783 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
784 'upload_date': '20151119',
785 'uploader_id': 'IronSoulElf',
786 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
787 'uploader': 'IronSoulElf',
788 'license': 'Standard YouTube License',
789 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
792 'skip_download': True,
796 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
797 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
798 'only_matching': True,
801 # Video with yt:stretch=17:0
802 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
806 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
807 'description': 'md5:ee18a25c350637c8faff806845bddee9',
808 'upload_date': '20151107',
809 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
810 'uploader': 'CH GAMER DROID',
813 'skip_download': True,
817 # Video licensed under Creative Commons
818 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
822 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
823 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
824 'upload_date': '20150127',
825 'uploader_id': 'BerkmanCenter',
826 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
827 'uploader': 'BerkmanCenter',
828 'license': 'Creative Commons Attribution license (reuse allowed)',
831 'skip_download': True,
835 # Channel-like uploader_url
836 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
840 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
841 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
842 'upload_date': '20151119',
843 'uploader': 'Bernie 2016',
844 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
845 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
846 'license': 'Creative Commons Attribution license (reuse allowed)',
849 'skip_download': True,
853 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
854 'only_matching': True,
858 def __init__(self, *args, **kwargs):
859 super(YoutubeIE, self).__init__(*args, **kwargs)
860 self._player_cache = {}
862 def report_video_info_webpage_download(self, video_id):
863 """Report attempt to download video info webpage."""
864 self.to_screen('%s: Downloading video info webpage' % video_id)
866 def report_information_extraction(self, video_id):
867 """Report attempt to extract video information."""
868 self.to_screen('%s: Extracting video information' % video_id)
870 def report_unavailable_format(self, video_id, format):
871 """Report extracted video URL."""
872 self.to_screen('%s: Format %s not available' % (video_id, format))
874 def report_rtmp_download(self):
875 """Indicate the download will use the RTMP protocol."""
876 self.to_screen('RTMP download detected')
878 def _signature_cache_id(self, example_sig):
879 """ Return a string representation of a signature """
880 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
882 def _extract_signature_function(self, video_id, player_url, example_sig):
884 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
887 raise ExtractorError('Cannot identify player %r' % player_url)
888 player_type = id_m.group('ext')
889 player_id = id_m.group('id')
891 # Read from filesystem cache
892 func_id = '%s_%s_%s' % (
893 player_type, player_id, self._signature_cache_id(example_sig))
894 assert os.path.basename(func_id) == func_id
896 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
897 if cache_spec is not None:
898 return lambda s: ''.join(s[i] for i in cache_spec)
901 'Downloading player %s' % player_url
902 if self._downloader.params.get('verbose') else
903 'Downloading %s player %s' % (player_type, player_id)
905 if player_type == 'js':
906 code = self._download_webpage(
907 player_url, video_id,
909 errnote='Download of %s failed' % player_url)
910 res = self._parse_sig_js(code)
911 elif player_type == 'swf':
912 urlh = self._request_webpage(
913 player_url, video_id,
915 errnote='Download of %s failed' % player_url)
917 res = self._parse_sig_swf(code)
919 assert False, 'Invalid player type %r' % player_type
921 test_string = ''.join(map(compat_chr, range(len(example_sig))))
922 cache_res = res(test_string)
923 cache_spec = [ord(c) for c in cache_res]
925 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
928 def _print_sig_code(self, func, example_sig):
929 def gen_sig_code(idxs):
930 def _genslice(start, end, step):
931 starts = '' if start == 0 else str(start)
932 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
933 steps = '' if step == 1 else (':%d' % step)
934 return 's[%s%s%s]' % (starts, ends, steps)
937 # Quelch pyflakes warnings - start will be set when step is set
938 start = '(Never used)'
939 for i, prev in zip(idxs[1:], idxs[:-1]):
943 yield _genslice(start, prev, step)
946 if i - prev in [-1, 1]:
955 yield _genslice(start, i, step)
957 test_string = ''.join(map(compat_chr, range(len(example_sig))))
958 cache_res = func(test_string)
959 cache_spec = [ord(c) for c in cache_res]
960 expr_code = ' + '.join(gen_sig_code(cache_spec))
961 signature_id_tuple = '(%s)' % (
962 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
963 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
964 ' return %s\n') % (signature_id_tuple, expr_code)
965 self.to_screen('Extracted signature function:\n' + code)
967 def _parse_sig_js(self, jscode):
968 funcname = self._search_regex(
969 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
970 'Initial JS player signature function name')
972 jsi = JSInterpreter(jscode)
973 initial_function = jsi.extract_function(funcname)
974 return lambda s: initial_function([s])
976 def _parse_sig_swf(self, file_contents):
977 swfi = SWFInterpreter(file_contents)
978 TARGET_CLASSNAME = 'SignatureDecipher'
979 searched_class = swfi.extract_class(TARGET_CLASSNAME)
980 initial_function = swfi.extract_function(searched_class, 'decipher')
981 return lambda s: initial_function([s])
983 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
984 """Turn the encrypted s field into a working signature"""
986 if player_url is None:
987 raise ExtractorError('Cannot decrypt signature without player_url')
989 if player_url.startswith('//'):
990 player_url = 'https:' + player_url
992 player_id = (player_url, self._signature_cache_id(s))
993 if player_id not in self._player_cache:
994 func = self._extract_signature_function(
995 video_id, player_url, s
997 self._player_cache[player_id] = func
998 func = self._player_cache[player_id]
999 if self._downloader.params.get('youtube_print_sig_code'):
1000 self._print_sig_code(func, s)
1002 except Exception as e:
1003 tb = traceback.format_exc()
1004 raise ExtractorError(
1005 'Signature extraction failed: ' + tb, cause=e)
1007 def _get_subtitles(self, video_id, webpage):
1009 subs_doc = self._download_xml(
1010 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1011 video_id, note=False)
1012 except ExtractorError as err:
1013 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1017 for track in subs_doc.findall('track'):
1018 lang = track.attrib['lang_code']
1019 if lang in sub_lang_list:
1022 for ext in self._SUBTITLE_FORMATS:
1023 params = compat_urllib_parse_urlencode({
1027 'name': track.attrib['name'].encode('utf-8'),
1029 sub_formats.append({
1030 'url': 'https://www.youtube.com/api/timedtext?' + params,
1033 sub_lang_list[lang] = sub_formats
1034 if not sub_lang_list:
1035 self._downloader.report_warning('video doesn\'t have subtitles')
1037 return sub_lang_list
1039 def _get_ytplayer_config(self, video_id, webpage):
1041 # User data may contain arbitrary character sequences that may affect
1042 # JSON extraction with regex, e.g. when '};' is contained the second
1043 # regex won't capture the whole JSON. Yet working around by trying more
1044 # concrete regex first keeping in mind proper quoted string handling
1045 # to be implemented in future that will replace this workaround (see
1046 # https://github.com/rg3/youtube-dl/issues/7468,
1047 # https://github.com/rg3/youtube-dl/pull/7599)
1048 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1049 r';ytplayer\.config\s*=\s*({.+?});',
1051 config = self._search_regex(
1052 patterns, webpage, 'ytplayer.config', default=None)
1054 return self._parse_json(
1055 uppercase_escape(config), video_id, fatal=False)
1057 def _get_automatic_captions(self, video_id, webpage):
1058 """We need the webpage for getting the captions url, pass it as an
1059 argument to speed up the process."""
1060 self.to_screen('%s: Looking for automatic captions' % video_id)
1061 player_config = self._get_ytplayer_config(video_id, webpage)
1062 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1063 if not player_config:
1064 self._downloader.report_warning(err_msg)
1067 args = player_config['args']
1068 caption_url = args.get('ttsurl')
1070 timestamp = args['timestamp']
1071 # We get the available subtitles
1072 list_params = compat_urllib_parse_urlencode({
1077 list_url = caption_url + '&' + list_params
1078 caption_list = self._download_xml(list_url, video_id)
1079 original_lang_node = caption_list.find('track')
1080 if original_lang_node is None:
1081 self._downloader.report_warning('Video doesn\'t have automatic captions')
1083 original_lang = original_lang_node.attrib['lang_code']
1084 caption_kind = original_lang_node.attrib.get('kind', '')
1087 for lang_node in caption_list.findall('target'):
1088 sub_lang = lang_node.attrib['lang_code']
1090 for ext in self._SUBTITLE_FORMATS:
1091 params = compat_urllib_parse_urlencode({
1092 'lang': original_lang,
1096 'kind': caption_kind,
1098 sub_formats.append({
1099 'url': caption_url + '&' + params,
1102 sub_lang_list[sub_lang] = sub_formats
1103 return sub_lang_list
1105 # Some videos don't provide ttsurl but rather caption_tracks and
1106 # caption_translation_languages (e.g. 20LmZk1hakA)
1107 caption_tracks = args['caption_tracks']
1108 caption_translation_languages = args['caption_translation_languages']
1109 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1110 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1111 caption_qs = compat_parse_qs(parsed_caption_url.query)
1114 for lang in caption_translation_languages.split(','):
1115 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1116 sub_lang = lang_qs.get('lc', [None])[0]
1120 for ext in self._SUBTITLE_FORMATS:
1122 'tlang': [sub_lang],
1125 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1126 query=compat_urllib_parse_urlencode(caption_qs, True)))
1127 sub_formats.append({
1131 sub_lang_list[sub_lang] = sub_formats
1132 return sub_lang_list
1133 # An extractor error can be raise by the download process if there are
1134 # no automatic captions but there are subtitles
1135 except (KeyError, ExtractorError):
1136 self._downloader.report_warning(err_msg)
1139 def _mark_watched(self, video_id, video_info):
1140 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1141 if not playback_url:
1143 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1144 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1146 # cpn generation algorithm is reverse engineered from base.js.
1147 # In fact it works even with dummy cpn.
1148 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1149 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1155 playback_url = compat_urlparse.urlunparse(
1156 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1158 self._download_webpage(
1159 playback_url, video_id, 'Marking watched',
1160 'Unable to mark watched', fatal=False)
1163 def extract_id(cls, url):
1164 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1166 raise ExtractorError('Invalid URL: %s' % url)
1167 video_id = mobj.group(2)
1170 def _extract_from_m3u8(self, manifest_url, video_id):
1173 def _get_urls(_manifest):
1174 lines = _manifest.split('\n')
1175 urls = filter(lambda l: l and not l.startswith('#'),
1178 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1179 formats_urls = _get_urls(manifest)
1180 for format_url in formats_urls:
1181 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1182 url_map[itag] = format_url
1185 def _extract_annotations(self, video_id):
1186 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1187 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1189 def _real_extract(self, url):
1190 url, smuggled_data = unsmuggle_url(url, {})
1193 'http' if self._downloader.params.get('prefer_insecure', False)
1198 parsed_url = compat_urllib_parse_urlparse(url)
1199 for component in [parsed_url.fragment, parsed_url.query]:
1200 query = compat_parse_qs(component)
1201 if start_time is None and 't' in query:
1202 start_time = parse_duration(query['t'][0])
1203 if start_time is None and 'start' in query:
1204 start_time = parse_duration(query['start'][0])
1205 if end_time is None and 'end' in query:
1206 end_time = parse_duration(query['end'][0])
1208 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1209 mobj = re.search(self._NEXT_URL_RE, url)
1211 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1212 video_id = self.extract_id(url)
1215 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1216 video_webpage = self._download_webpage(url, video_id)
1218 # Attempt to extract SWF player URL
1219 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1220 if mobj is not None:
1221 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1227 def add_dash_mpd(video_info):
1228 dash_mpd = video_info.get('dashmpd')
1229 if dash_mpd and dash_mpd[0] not in dash_mpds:
1230 dash_mpds.append(dash_mpd[0])
1233 embed_webpage = None
1235 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1237 # We simulate the access to the video from www.youtube.com/v/{video_id}
1238 # this can be viewed without login into Youtube
1239 url = proto + '://www.youtube.com/embed/%s' % video_id
1240 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1241 data = compat_urllib_parse_urlencode({
1242 'video_id': video_id,
1243 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1244 'sts': self._search_regex(
1245 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1247 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1248 video_info_webpage = self._download_webpage(
1249 video_info_url, video_id,
1250 note='Refetching age-gated info webpage',
1251 errnote='unable to download video info webpage')
1252 video_info = compat_parse_qs(video_info_webpage)
1253 add_dash_mpd(video_info)
1257 # Try looking directly into the video webpage
1258 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1260 args = ytplayer_config['args']
1261 if args.get('url_encoded_fmt_stream_map'):
1262 # Convert to the same format returned by compat_parse_qs
1263 video_info = dict((k, [v]) for k, v in args.items())
1264 add_dash_mpd(video_info)
1265 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1267 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1268 # We also try looking in get_video_info since it may contain different dashmpd
1269 # URL that points to a DASH manifest with possibly different itag set (some itags
1270 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1271 # manifest pointed by get_video_info's dashmpd).
1272 # The general idea is to take a union of itags of both DASH manifests (for example
1273 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1274 self.report_video_info_webpage_download(video_id)
1275 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1277 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1278 % (proto, video_id, el_type))
1279 video_info_webpage = self._download_webpage(
1281 video_id, note=False,
1282 errnote='unable to download video info webpage')
1283 get_video_info = compat_parse_qs(video_info_webpage)
1284 if get_video_info.get('use_cipher_signature') != ['True']:
1285 add_dash_mpd(get_video_info)
1287 video_info = get_video_info
1288 if 'token' in get_video_info:
1289 # Different get_video_info requests may report different results, e.g.
1290 # some may report video unavailability, but some may serve it without
1291 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1292 # the original webpage as well as el=info and el=embedded get_video_info
1293 # requests report video unavailability due to geo restriction while
1294 # el=detailpage succeeds and returns valid data). This is probably
1295 # due to YouTube measures against IP ranges of hosting providers.
1296 # Working around by preferring the first succeeded video_info containing
1297 # the token if no such video_info yet was found.
1298 if 'token' not in video_info:
1299 video_info = get_video_info
1301 if 'token' not in video_info:
1302 if 'reason' in video_info:
1303 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1304 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1306 raise ExtractorError('YouTube said: This video is available in %s only' % (
1307 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1309 raise ExtractorError(
1310 'YouTube said: %s' % video_info['reason'][0],
1311 expected=True, video_id=video_id)
1313 raise ExtractorError(
1314 '"token" parameter not in video info for unknown reason',
1318 if 'title' in video_info:
1319 video_title = video_info['title'][0]
1321 self._downloader.report_warning('Unable to extract video title')
1325 video_description = get_element_by_id("eow-description", video_webpage)
1326 if video_description:
1327 video_description = re.sub(r'''(?x)
1329 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1330 (?:title|href)="([^"]+)"\s+
1331 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1332 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1335 ''', r'\1', video_description)
1336 video_description = clean_html(video_description)
1338 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1340 video_description = unescapeHTML(fd_mobj.group(1))
1342 video_description = ''
1344 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1345 if not self._downloader.params.get('noplaylist'):
1348 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1349 for feed in multifeed_metadata_list.split(','):
1350 # Unquote should take place before split on comma (,) since textual
1351 # fields may contain comma as well (see
1352 # https://github.com/rg3/youtube-dl/issues/8536)
1353 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1355 '_type': 'url_transparent',
1356 'ie_key': 'Youtube',
1358 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1359 {'force_singlefeed': True}),
1360 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1362 feed_ids.append(feed_data['id'][0])
1364 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1365 % (', '.join(feed_ids), video_id))
1366 return self.playlist_result(entries, video_id, video_title, video_description)
1367 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1369 if 'view_count' in video_info:
1370 view_count = int(video_info['view_count'][0])
1374 # Check for "rental" videos
1375 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1376 raise ExtractorError('"rental" videos not supported')
1378 # Start extracting information
1379 self.report_information_extraction(video_id)
1382 if 'author' not in video_info:
1383 raise ExtractorError('Unable to extract uploader name')
1384 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1387 video_uploader_id = None
1388 video_uploader_url = None
1390 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1392 if mobj is not None:
1393 video_uploader_id = mobj.group('uploader_id')
1394 video_uploader_url = mobj.group('uploader_url')
1396 self._downloader.report_warning('unable to extract uploader nickname')
1399 # We try first to get a high quality image:
1400 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1401 video_webpage, re.DOTALL)
1402 if m_thumb is not None:
1403 video_thumbnail = m_thumb.group(1)
1404 elif 'thumbnail_url' not in video_info:
1405 self._downloader.report_warning('unable to extract video thumbnail')
1406 video_thumbnail = None
1407 else: # don't panic if we can't find it
1408 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1411 upload_date = self._html_search_meta(
1412 'datePublished', video_webpage, 'upload date', default=None)
1414 upload_date = self._search_regex(
1415 [r'(?s)id="eow-date.*?>(.*?)</span>',
1416 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1417 video_webpage, 'upload date', default=None)
1419 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1420 upload_date = unified_strdate(upload_date)
1422 video_license = self._html_search_regex(
1423 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1424 video_webpage, 'license', default=None)
1426 m_music = re.search(
1427 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1430 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1431 video_creator = clean_html(m_music.group('creator'))
1433 video_alt_title = video_creator = None
1435 m_cat_container = self._search_regex(
1436 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1437 video_webpage, 'categories', default=None)
1439 category = self._html_search_regex(
1440 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1442 video_categories = None if category is None else [category]
1444 video_categories = None
1447 unescapeHTML(m.group('content'))
1448 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1450 def _extract_count(count_name):
1451 return str_to_int(self._search_regex(
1452 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1453 % re.escape(count_name),
1454 video_webpage, count_name, default=None))
1456 like_count = _extract_count('like')
1457 dislike_count = _extract_count('dislike')
1460 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1461 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1463 if 'length_seconds' not in video_info:
1464 self._downloader.report_warning('unable to extract video duration')
1465 video_duration = None
1467 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1470 video_annotations = None
1471 if self._downloader.params.get('writeannotations', False):
1472 video_annotations = self._extract_annotations(video_id)
1474 def _map_to_format_list(urlmap):
1476 for itag, video_real_url in urlmap.items():
1479 'url': video_real_url,
1480 'player_url': player_url,
1482 if itag in self._formats:
1483 dct.update(self._formats[itag])
1487 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1488 self.report_rtmp_download()
1490 'format_id': '_rtmp',
1492 'url': video_info['conn'][0],
1493 'player_url': player_url,
1495 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1496 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1497 if 'rtmpe%3Dyes' in encoded_url_map:
1498 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1500 fmt_list = video_info.get('fmt_list', [''])[0]
1502 for fmt in fmt_list.split(','):
1503 spec = fmt.split('/')
1505 width_height = spec[1].split('x')
1506 if len(width_height) == 2:
1507 formats_spec[spec[0]] = {
1508 'resolution': spec[1],
1509 'width': int_or_none(width_height[0]),
1510 'height': int_or_none(width_height[1]),
1513 for url_data_str in encoded_url_map.split(','):
1514 url_data = compat_parse_qs(url_data_str)
1515 if 'itag' not in url_data or 'url' not in url_data:
1517 format_id = url_data['itag'][0]
1518 url = url_data['url'][0]
1520 if 'sig' in url_data:
1521 url += '&signature=' + url_data['sig'][0]
1522 elif 's' in url_data:
1523 encrypted_sig = url_data['s'][0]
1524 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1526 jsplayer_url_json = self._search_regex(
1528 embed_webpage if age_gate else video_webpage,
1529 'JS player URL (1)', default=None)
1530 if not jsplayer_url_json and not age_gate:
1531 # We need the embed website after all
1532 if embed_webpage is None:
1533 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1534 embed_webpage = self._download_webpage(
1535 embed_url, video_id, 'Downloading embed webpage')
1536 jsplayer_url_json = self._search_regex(
1537 ASSETS_RE, embed_webpage, 'JS player URL')
1539 player_url = json.loads(jsplayer_url_json)
1540 if player_url is None:
1541 player_url_json = self._search_regex(
1542 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1543 video_webpage, 'age gate player URL')
1544 player_url = json.loads(player_url_json)
1546 if self._downloader.params.get('verbose'):
1547 if player_url is None:
1548 player_version = 'unknown'
1549 player_desc = 'unknown'
1551 if player_url.endswith('swf'):
1552 player_version = self._search_regex(
1553 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1554 'flash player', fatal=False)
1555 player_desc = 'flash player %s' % player_version
1557 player_version = self._search_regex(
1558 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1560 'html5 player', fatal=False)
1561 player_desc = 'html5 player %s' % player_version
1563 parts_sizes = self._signature_cache_id(encrypted_sig)
1564 self.to_screen('{%s} signature length %s, %s' %
1565 (format_id, parts_sizes, player_desc))
1567 signature = self._decrypt_signature(
1568 encrypted_sig, video_id, player_url, age_gate)
1569 url += '&signature=' + signature
1570 if 'ratebypass' not in url:
1571 url += '&ratebypass=yes'
1574 'format_id': format_id,
1576 'player_url': player_url,
1578 if format_id in self._formats:
1579 dct.update(self._formats[format_id])
1580 if format_id in formats_spec:
1581 dct.update(formats_spec[format_id])
1583 # Some itags are not included in DASH manifest thus corresponding formats will
1584 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1585 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1586 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1587 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1590 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1591 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1594 'fps': int_or_none(url_data.get('fps', [None])[0]),
1595 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1597 for key, value in more_fields.items():
1600 type_ = url_data.get('type', [None])[0]
1602 type_split = type_.split(';')
1603 kind_ext = type_split[0].split('/')
1604 if len(kind_ext) == 2:
1606 dct['ext'] = mimetype2ext(type_split[0])
1607 if kind in ('audio', 'video'):
1609 for mobj in re.finditer(
1610 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1611 if mobj.group('key') == 'codecs':
1612 codecs = mobj.group('val')
1615 codecs = codecs.split(',')
1616 if len(codecs) == 2:
1617 acodec, vcodec = codecs[1], codecs[0]
1619 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1625 elif video_info.get('hlsvp'):
1626 manifest_url = video_info['hlsvp'][0]
1627 url_map = self._extract_from_m3u8(manifest_url, video_id)
1628 formats = _map_to_format_list(url_map)
1629 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1630 for a_format in formats:
1631 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1633 unavailable_message = self._html_search_regex(
1634 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1635 video_webpage, 'unavailable message', default=None)
1636 if unavailable_message:
1637 raise ExtractorError(unavailable_message, expected=True)
1638 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1640 # Look for the DASH manifest
1641 if self._downloader.params.get('youtube_include_dash_manifest', True):
1642 dash_mpd_fatal = True
1643 for mpd_url in dash_mpds:
1646 def decrypt_sig(mobj):
1648 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1649 return '/signature/%s' % dec_s
1651 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1653 for df in self._extract_mpd_formats(
1654 mpd_url, video_id, fatal=dash_mpd_fatal,
1655 formats_dict=self._formats):
1656 # Do not overwrite DASH format found in some previous DASH manifest
1657 if df['format_id'] not in dash_formats:
1658 dash_formats[df['format_id']] = df
1659 # Additional DASH manifests may end up in HTTP Error 403 therefore
1660 # allow them to fail without bug report message if we already have
1661 # some DASH manifest succeeded. This is temporary workaround to reduce
1662 # burst of bug reports until we figure out the reason and whether it
1663 # can be fixed at all.
1664 dash_mpd_fatal = False
1665 except (ExtractorError, KeyError) as e:
1666 self.report_warning(
1667 'Skipping DASH manifest: %r' % e, video_id)
1669 # Remove the formats we found through non-DASH, they
1670 # contain less info and it can be wrong, because we use
1671 # fixed values (for example the resolution). See
1672 # https://github.com/rg3/youtube-dl/issues/5774 for an
1674 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1675 formats.extend(dash_formats.values())
1677 # Check for malformed aspect ratio
1678 stretched_m = re.search(
1679 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1682 w = float(stretched_m.group('w'))
1683 h = float(stretched_m.group('h'))
1684 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1685 # We will only process correct ratios.
1689 if f.get('vcodec') != 'none':
1690 f['stretched_ratio'] = ratio
1692 self._sort_formats(formats)
1694 self.mark_watched(video_id, video_info)
1698 'uploader': video_uploader,
1699 'uploader_id': video_uploader_id,
1700 'uploader_url': video_uploader_url,
1701 'upload_date': upload_date,
1702 'license': video_license,
1703 'creator': video_creator,
1704 'title': video_title,
1705 'alt_title': video_alt_title,
1706 'thumbnail': video_thumbnail,
1707 'description': video_description,
1708 'categories': video_categories,
1710 'subtitles': video_subtitles,
1711 'automatic_captions': automatic_captions,
1712 'duration': video_duration,
1713 'age_limit': 18 if age_gate else 0,
1714 'annotations': video_annotations,
1715 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1716 'view_count': view_count,
1717 'like_count': like_count,
1718 'dislike_count': dislike_count,
1719 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1722 'start_time': start_time,
1723 'end_time': end_time,
1727 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1728 IE_DESC = 'YouTube.com playlists'
1729 _VALID_URL = r"""(?x)(?:
1734 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1735 \? (?:.*?[&;])*? (?:p|a|list)=
1739 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1740 # Top tracks, they can also include dots
1745 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1747 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1748 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1749 IE_NAME = 'youtube:playlist'
1751 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1753 'title': 'ytdl test PL',
1754 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1756 'playlist_count': 3,
1758 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1760 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1761 'title': 'YDL_Empty_List',
1763 'playlist_count': 0,
1765 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1766 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1768 'title': '29C3: Not my department',
1769 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1771 'playlist_count': 95,
1773 'note': 'issue #673',
1774 'url': 'PLBB231211A4F62143',
1776 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1777 'id': 'PLBB231211A4F62143',
1779 'playlist_mincount': 26,
1781 'note': 'Large playlist',
1782 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1784 'title': 'Uploads from Cauchemar',
1785 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1787 'playlist_mincount': 799,
1789 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1791 'title': 'YDL_safe_search',
1792 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1794 'playlist_count': 2,
1797 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1798 'playlist_count': 4,
1801 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1804 'note': 'Embedded SWF player',
1805 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1806 'playlist_count': 4,
1809 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1812 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1813 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1815 'title': 'Uploads from Interstellar Movie',
1816 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1818 'playlist_mincout': 21,
1821 def _real_initialize(self):
1824 def _extract_mix(self, playlist_id):
1825 # The mixes are generated from a single video
1826 # the id of the playlist is just 'RD' + video_id
1828 last_id = playlist_id[-11:]
1829 for n in itertools.count(1):
1830 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1831 webpage = self._download_webpage(
1832 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1833 new_ids = orderedSet(re.findall(
1834 r'''(?xs)data-video-username=".*?".*?
1835 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1837 # Fetch new pages until all the videos are repeated, it seems that
1838 # there are always 51 unique videos.
1839 new_ids = [_id for _id in new_ids if _id not in ids]
1845 url_results = self._ids_to_results(ids)
1847 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1849 search_title('playlist-title') or
1850 search_title('title long-title') or
1851 search_title('title'))
1852 title = clean_html(title_span)
1854 return self.playlist_result(url_results, playlist_id, title)
1856 def _extract_playlist(self, playlist_id):
1857 url = self._TEMPLATE_URL % playlist_id
1858 page = self._download_webpage(url, playlist_id)
1860 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1861 match = match.strip()
1862 # Check if the playlist exists or is private
1863 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1864 raise ExtractorError(
1865 'The playlist doesn\'t exist or is private, use --username or '
1866 '--netrc to access it.',
1868 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1869 raise ExtractorError(
1870 'Invalid parameters. Maybe URL is incorrect.',
1872 elif re.match(r'[^<]*Choose your language[^<]*', match):
1875 self.report_warning('Youtube gives an alert message: ' + match)
1877 playlist_title = self._html_search_regex(
1878 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1881 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1883 def _check_download_just_video(self, url, playlist_id):
1884 # Check if it's a video-specific URL
1885 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1886 if 'v' in query_dict:
1887 video_id = query_dict['v'][0]
1888 if self._downloader.params.get('noplaylist'):
1889 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1890 return self.url_result(video_id, 'Youtube', video_id=video_id)
1892 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1894 def _real_extract(self, url):
1895 # Extract playlist id
1896 mobj = re.match(self._VALID_URL, url)
1898 raise ExtractorError('Invalid URL: %s' % url)
1899 playlist_id = mobj.group(1) or mobj.group(2)
1901 video = self._check_download_just_video(url, playlist_id)
1905 if playlist_id.startswith(('RD', 'UL', 'PU')):
1906 # Mixes require a custom extraction process
1907 return self._extract_mix(playlist_id)
1909 return self._extract_playlist(playlist_id)
1912 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1913 IE_DESC = 'YouTube.com channels'
1914 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1915 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1916 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1917 IE_NAME = 'youtube:channel'
1919 'note': 'paginated channel',
1920 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1921 'playlist_mincount': 91,
1923 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1924 'title': 'Uploads from lex will',
1927 'note': 'Age restricted channel',
1928 # from https://www.youtube.com/user/DeusExOfficial
1929 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1930 'playlist_mincount': 64,
1932 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1933 'title': 'Uploads from Deus Ex',
1938 def suitable(cls, url):
1939 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1940 else super(YoutubeChannelIE, cls).suitable(url))
1942 def _real_extract(self, url):
1943 channel_id = self._match_id(url)
1945 url = self._TEMPLATE_URL % channel_id
1947 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1948 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1949 # otherwise fallback on channel by page extraction
1950 channel_page = self._download_webpage(
1951 url + '?view=57', channel_id,
1952 'Downloading channel page', fatal=False)
1953 if channel_page is False:
1954 channel_playlist_id = False
1956 channel_playlist_id = self._html_search_meta(
1957 'channelId', channel_page, 'channel id', default=None)
1958 if not channel_playlist_id:
1959 channel_playlist_id = self._search_regex(
1960 r'data-(?:channel-external-|yt)id="([^"]+)"',
1961 channel_page, 'channel id', default=None)
1962 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1963 playlist_id = 'UU' + channel_playlist_id[2:]
1964 return self.url_result(
1965 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1967 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1968 autogenerated = re.search(r'''(?x)
1970 channel-header-autogenerated-label|
1971 yt-channel-title-autogenerated
1972 )[^"]*"''', channel_page) is not None
1975 # The videos are contained in a single page
1976 # the ajax pages can't be used, they are empty
1979 video_id, 'Youtube', video_id=video_id,
1980 video_title=video_title)
1981 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1982 return self.playlist_result(entries, channel_id)
1984 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1987 class YoutubeUserIE(YoutubeChannelIE):
1988 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1989 _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_-]+)'
1990 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1991 IE_NAME = 'youtube:user'
1994 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1995 'playlist_mincount': 320,
1997 'title': 'TheLinuxFoundation',
2000 'url': 'ytuser:phihag',
2001 'only_matching': True,
2005 def suitable(cls, url):
2006 # Don't return True if the url can be extracted with other youtube
2007 # extractor, the regex would is too permissive and it would match.
2008 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2009 if any(ie.suitable(url) for ie in other_yt_ies):
2012 return super(YoutubeUserIE, cls).suitable(url)
2015 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2016 IE_DESC = 'YouTube.com live streams'
2017 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2018 IE_NAME = 'youtube:live'
2021 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2023 'id': 'a48o2S1cPoo',
2025 'title': 'The Young Turks - Live Main Show',
2026 'uploader': 'The Young Turks',
2027 'uploader_id': 'TheYoungTurks',
2028 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2029 'upload_date': '20150715',
2030 'license': 'Standard YouTube License',
2031 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2032 'categories': ['News & Politics'],
2033 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2035 'dislike_count': int,
2038 'skip_download': True,
2041 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2042 'only_matching': True,
2045 def _real_extract(self, url):
2046 mobj = re.match(self._VALID_URL, url)
2047 channel_id = mobj.group('id')
2048 base_url = mobj.group('base_url')
2049 webpage = self._download_webpage(url, channel_id, fatal=False)
2051 page_type = self._og_search_property(
2052 'type', webpage, 'page type', default=None)
2053 video_id = self._html_search_meta(
2054 'videoId', webpage, 'video id', default=None)
2055 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2056 return self.url_result(video_id, YoutubeIE.ie_key())
2057 return self.url_result(base_url)
2060 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2061 IE_DESC = 'YouTube.com user/channel playlists'
2062 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2063 IE_NAME = 'youtube:playlists'
2066 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2067 'playlist_mincount': 4,
2069 'id': 'ThirstForScience',
2070 'title': 'Thirst for Science',
2073 # with "Load more" button
2074 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2075 'playlist_mincount': 70,
2078 'title': 'Игорь Клейнер',
2081 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2082 'playlist_mincount': 17,
2084 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2085 'title': 'Chem Player',
2090 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2091 IE_DESC = 'YouTube.com searches'
2092 # there doesn't appear to be a real limit, for example if you search for
2093 # 'python' you get more than 8.000.000 results
2094 _MAX_RESULTS = float('inf')
2095 IE_NAME = 'youtube:search'
2096 _SEARCH_KEY = 'ytsearch'
2097 _EXTRA_QUERY_ARGS = {}
2100 def _get_n_results(self, query, n):
2101 """Get a specified number of results for a query"""
2106 for pagenum in itertools.count(1):
2108 'search_query': query.encode('utf-8'),
2112 url_query.update(self._EXTRA_QUERY_ARGS)
2113 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2114 data = self._download_json(
2115 result_url, video_id='query "%s"' % query,
2116 note='Downloading page %s' % pagenum,
2117 errnote='Unable to download API page')
2118 html_content = data[1]['body']['content']
2120 if 'class="search-message' in html_content:
2121 raise ExtractorError(
2122 '[youtube] No video results', expected=True)
2124 new_videos = self._ids_to_results(orderedSet(re.findall(
2125 r'href="/watch\?v=(.{11})', html_content)))
2126 videos += new_videos
2127 if not new_videos or len(videos) > limit:
2132 return self.playlist_result(videos, query)
2135 class YoutubeSearchDateIE(YoutubeSearchIE):
2136 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2137 _SEARCH_KEY = 'ytsearchdate'
2138 IE_DESC = 'YouTube.com searches, newest videos first'
2139 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2142 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2143 IE_DESC = 'YouTube.com search URLs'
2144 IE_NAME = 'youtube:search_url'
2145 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2146 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2148 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2149 'playlist_mincount': 5,
2151 'title': 'youtube-dl test video',
2154 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2155 'only_matching': True,
2158 def _real_extract(self, url):
2159 mobj = re.match(self._VALID_URL, url)
2160 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2161 webpage = self._download_webpage(url, query)
2162 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2165 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2166 IE_DESC = 'YouTube.com (multi-season) shows'
2167 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2168 IE_NAME = 'youtube:show'
2170 'url': 'https://www.youtube.com/show/airdisasters',
2171 'playlist_mincount': 5,
2173 'id': 'airdisasters',
2174 'title': 'Air Disasters',
2178 def _real_extract(self, url):
2179 playlist_id = self._match_id(url)
2180 return super(YoutubeShowIE, self)._real_extract(
2181 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2184 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2186 Base class for feed extractors
2187 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2189 _LOGIN_REQUIRED = True
2193 return 'youtube:%s' % self._FEED_NAME
2195 def _real_initialize(self):
2198 def _real_extract(self, url):
2199 page = self._download_webpage(
2200 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2202 # The extraction process is the same as for playlists, but the regex
2203 # for the video ids doesn't contain an index
2205 more_widget_html = content_html = page
2206 for page_num in itertools.count(1):
2207 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2209 # 'recommended' feed has infinite 'load more' and each new portion spins
2210 # the same videos in (sometimes) slightly different order, so we'll check
2211 # for unicity and break when portion has no new videos
2212 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2218 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2222 more = self._download_json(
2223 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2224 'Downloading page #%s' % page_num,
2225 transform_source=uppercase_escape)
2226 content_html = more['content_html']
2227 more_widget_html = more['load_more_widget_html']
2229 return self.playlist_result(
2230 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2233 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2234 IE_NAME = 'youtube:watchlater'
2235 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2236 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2239 'url': 'https://www.youtube.com/playlist?list=WL',
2240 'only_matching': True,
2242 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2243 'only_matching': True,
2246 def _real_extract(self, url):
2247 video = self._check_download_just_video(url, 'WL')
2250 return self._extract_playlist('WL')
2253 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2254 IE_NAME = 'youtube:favorites'
2255 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2256 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2257 _LOGIN_REQUIRED = True
2259 def _real_extract(self, url):
2260 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2261 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2262 return self.url_result(playlist_id, 'YoutubePlaylist')
2265 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2266 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2267 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2268 _FEED_NAME = 'recommended'
2269 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2272 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2273 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2274 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2275 _FEED_NAME = 'subscriptions'
2276 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2279 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2280 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2281 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2282 _FEED_NAME = 'history'
2283 _PLAYLIST_TITLE = 'Youtube History'
2286 class YoutubeTruncatedURLIE(InfoExtractor):
2287 IE_NAME = 'youtube:truncated_url'
2288 IE_DESC = False # Do not list
2289 _VALID_URL = r'''(?x)
2291 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2294 annotation_id=annotation_[^&]+|
2300 attribution_link\?a=[^&]+
2306 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2307 'only_matching': True,
2309 'url': 'http://www.youtube.com/watch?',
2310 'only_matching': True,
2312 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2313 'only_matching': True,
2315 'url': 'https://www.youtube.com/watch?feature=foo',
2316 'only_matching': True,
2318 'url': 'https://www.youtube.com/watch?hl=en-GB',
2319 'only_matching': True,
2321 'url': 'https://www.youtube.com/watch?t=2372',
2322 'only_matching': True,
2325 def _real_extract(self, url):
2326 raise ExtractorError(
2327 'Did you forget to quote the URL? Remember that & is a meta '
2328 'character in most shells, so you want to put the URL in quotes, '
2330 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2331 ' or simply youtube-dl BaW_jenozKc .',
2335 class YoutubeTruncatedIDIE(InfoExtractor):
2336 IE_NAME = 'youtube:truncated_id'
2337 IE_DESC = False # Do not list
2338 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2341 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2342 'only_matching': True,
2345 def _real_extract(self, url):
2346 video_id = self._match_id(url)
2347 raise ExtractorError(
2348 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),