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'},
347 '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
348 '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'preference': -50, 'container': 'm4a_dash'},
351 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
352 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
353 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
354 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
355 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
356 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
357 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
358 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
359 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
361 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
362 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
363 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
364 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
365 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
366 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
367 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
368 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
369 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
370 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
371 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
372 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
375 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
376 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
378 # Dash webm audio with opus inside
379 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
380 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
381 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
384 '_rtmp': {'protocol': 'rtmp'},
386 _SUBTITLE_FORMATS = ('ttml', 'vtt')
391 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
395 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
396 'uploader': 'Philipp Hagemeister',
397 'uploader_id': 'phihag',
398 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
399 'upload_date': '20121002',
400 'license': 'Standard YouTube License',
401 '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 .',
402 'categories': ['Science & Technology'],
403 'tags': ['youtube-dl'],
405 'dislike_count': int,
411 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
412 'note': 'Test generic use_cipher_signature video (#897)',
416 'upload_date': '20120506',
417 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
418 'alt_title': 'I Love It (feat. Charli XCX)',
419 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
420 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
421 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
422 'iconic ep', 'iconic', 'love', 'it'],
423 'uploader': 'Icona Pop',
424 'uploader_id': 'IconaPop',
425 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
426 'license': 'Standard YouTube License',
427 'creator': 'Icona Pop',
431 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
432 'note': 'Test VEVO video with age protection (#956)',
436 'upload_date': '20130703',
437 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
438 'alt_title': 'Tunnel Vision',
439 'description': 'md5:64249768eec3bc4276236606ea996373',
440 'uploader': 'justintimberlakeVEVO',
441 'uploader_id': 'justintimberlakeVEVO',
442 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
443 'license': 'Standard YouTube License',
444 'creator': 'Justin Timberlake',
449 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
450 'note': 'Embed-only video (#1746)',
454 'upload_date': '20120608',
455 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
456 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
457 'uploader': 'SET India',
458 'uploader_id': 'setindia',
459 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
460 'license': 'Standard YouTube License',
465 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
466 'note': 'Use the first video ID in the URL',
470 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
471 'uploader': 'Philipp Hagemeister',
472 'uploader_id': 'phihag',
473 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
474 'upload_date': '20121002',
475 'license': 'Standard YouTube License',
476 '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 .',
477 'categories': ['Science & Technology'],
478 'tags': ['youtube-dl'],
480 'dislike_count': int,
483 'skip_download': True,
487 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
488 'note': '256k DASH audio (format 141) via DASH manifest',
492 'upload_date': '20121002',
493 'uploader_id': '8KVIDEO',
494 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
496 'uploader': '8KVIDEO',
497 'license': 'Standard YouTube License',
498 'title': 'UHDTV TEST 8K VIDEO.mp4'
501 'youtube_include_dash_manifest': True,
504 'skip': 'format 141 not served anymore',
506 # DASH manifest with encrypted signature
508 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
512 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
513 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
514 'uploader': 'AfrojackVEVO',
515 'uploader_id': 'AfrojackVEVO',
516 'upload_date': '20131011',
517 'license': 'Standard YouTube License',
520 'youtube_include_dash_manifest': True,
521 'format': '141/bestaudio[ext=m4a]',
524 # JS player signature function name containing $
526 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
530 'title': 'Taylor Swift - Shake It Off',
531 'alt_title': 'Shake It Off',
532 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
533 'uploader': 'TaylorSwiftVEVO',
534 'uploader_id': 'TaylorSwiftVEVO',
535 'upload_date': '20140818',
536 'license': 'Standard YouTube License',
537 'creator': 'Taylor Swift',
540 'youtube_include_dash_manifest': True,
541 'format': '141/bestaudio[ext=m4a]',
546 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
550 'upload_date': '20100909',
551 'uploader': 'The Amazing Atheist',
552 'uploader_id': 'TheAmazingAtheist',
553 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
554 'license': 'Standard YouTube License',
555 'title': 'Burning Everyone\'s Koran',
556 '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',
559 # Normal age-gate video (No vevo, embed allowed)
561 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
565 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
566 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
567 'uploader': 'The Witcher',
568 'uploader_id': 'WitcherGame',
569 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
570 'upload_date': '20140605',
571 'license': 'Standard YouTube License',
575 # Age-gate video with encrypted signature
577 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
581 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
582 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
583 'uploader': 'LloydVEVO',
584 'uploader_id': 'LloydVEVO',
585 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
586 'upload_date': '20110629',
587 'license': 'Standard YouTube License',
591 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
593 'url': '__2ABJjxzNo',
597 'upload_date': '20100430',
598 'uploader_id': 'deadmau5',
599 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
600 'creator': 'deadmau5',
601 'description': 'md5:12c56784b8032162bb936a5f76d55360',
602 'uploader': 'deadmau5',
603 'license': 'Standard YouTube License',
604 'title': 'Deadmau5 - Some Chords (HD)',
605 'alt_title': 'Some Chords',
607 'expected_warnings': [
608 'DASH manifest missing',
611 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
613 'url': 'lqQg6PlCWgI',
617 'upload_date': '20150827',
618 'uploader_id': 'olympic',
619 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
620 'license': 'Standard YouTube License',
621 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
622 'uploader': 'Olympic',
623 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
626 'skip_download': 'requires avconv',
631 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
635 'stretched_ratio': 16 / 9.,
636 'upload_date': '20110310',
637 'uploader_id': 'AllenMeow',
638 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
639 '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',
641 'license': 'Standard YouTube License',
642 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
645 # url_encoded_fmt_stream_map is empty string
647 'url': 'qEJwOuvDf7I',
651 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
653 'upload_date': '20150404',
654 'uploader_id': 'spbelect',
655 'uploader': 'Наблюдатели Петербурга',
658 'skip_download': 'requires avconv',
660 'skip': 'This live event has ended.',
662 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
664 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
668 'title': 'md5:7b81415841e02ecd4313668cde88737a',
669 'description': 'md5:116377fd2963b81ec4ce64b542173306',
670 'upload_date': '20150625',
671 'uploader_id': 'dorappi2000',
672 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
673 'uploader': 'dorappi2000',
674 'license': 'Standard YouTube License',
675 'formats': 'mincount:32',
678 # DASH manifest with segment_list
680 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
681 'md5': '8ce563a1d667b599d21064e982ab9e31',
685 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
686 'uploader': 'Airtek',
687 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
688 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
689 'license': 'Standard YouTube License',
690 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
693 'youtube_include_dash_manifest': True,
694 'format': '135', # bestvideo
696 'skip': 'This live event has ended.',
699 # Multifeed videos (multiple cameras), URL is for Main Camera
700 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
703 'title': 'teamPGP: Rocket League Noob Stream',
704 'description': 'md5:dc7872fb300e143831327f1bae3af010',
710 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
711 'description': 'md5:dc7872fb300e143831327f1bae3af010',
712 'upload_date': '20150721',
713 'uploader': 'Beer Games Beer',
714 'uploader_id': 'beergamesbeer',
715 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
716 'license': 'Standard YouTube License',
722 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
723 'description': 'md5:dc7872fb300e143831327f1bae3af010',
724 'upload_date': '20150721',
725 'uploader': 'Beer Games Beer',
726 'uploader_id': 'beergamesbeer',
727 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
728 'license': 'Standard YouTube License',
734 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
735 'description': 'md5:dc7872fb300e143831327f1bae3af010',
736 'upload_date': '20150721',
737 'uploader': 'Beer Games Beer',
738 'uploader_id': 'beergamesbeer',
739 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
740 'license': 'Standard YouTube License',
746 'title': 'teamPGP: Rocket League Noob Stream (zim)',
747 'description': 'md5:dc7872fb300e143831327f1bae3af010',
748 'upload_date': '20150721',
749 'uploader': 'Beer Games Beer',
750 'uploader_id': 'beergamesbeer',
751 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
752 'license': 'Standard YouTube License',
756 'skip_download': True,
760 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
761 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
764 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
767 'skip': 'Not multifeed anymore',
770 'url': 'http://vid.plus/FlRa-iH7PGw',
771 'only_matching': True,
774 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
775 'only_matching': True,
778 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
779 # Also tests cut-off URL expansion in video description (see
780 # https://github.com/rg3/youtube-dl/issues/1892,
781 # https://github.com/rg3/youtube-dl/issues/8164)
782 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
786 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
787 'alt_title': 'Dark Walk',
788 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
789 'upload_date': '20151119',
790 'uploader_id': 'IronSoulElf',
791 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
792 'uploader': 'IronSoulElf',
793 'license': 'Standard YouTube License',
794 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
797 'skip_download': True,
801 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
802 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
803 'only_matching': True,
806 # Video with yt:stretch=17:0
807 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
811 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
812 'description': 'md5:ee18a25c350637c8faff806845bddee9',
813 'upload_date': '20151107',
814 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
815 'uploader': 'CH GAMER DROID',
818 'skip_download': True,
820 'skip': 'This video does not exist.',
823 # Video licensed under Creative Commons
824 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
828 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
829 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
830 'upload_date': '20150127',
831 'uploader_id': 'BerkmanCenter',
832 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
833 'uploader': 'BerkmanCenter',
834 'license': 'Creative Commons Attribution license (reuse allowed)',
837 'skip_download': True,
841 # Channel-like uploader_url
842 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
846 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
847 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
848 'upload_date': '20151119',
849 'uploader': 'Bernie 2016',
850 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
851 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
852 'license': 'Creative Commons Attribution license (reuse allowed)',
855 'skip_download': True,
859 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
860 'only_matching': True,
864 def __init__(self, *args, **kwargs):
865 super(YoutubeIE, self).__init__(*args, **kwargs)
866 self._player_cache = {}
868 def report_video_info_webpage_download(self, video_id):
869 """Report attempt to download video info webpage."""
870 self.to_screen('%s: Downloading video info webpage' % video_id)
872 def report_information_extraction(self, video_id):
873 """Report attempt to extract video information."""
874 self.to_screen('%s: Extracting video information' % video_id)
876 def report_unavailable_format(self, video_id, format):
877 """Report extracted video URL."""
878 self.to_screen('%s: Format %s not available' % (video_id, format))
880 def report_rtmp_download(self):
881 """Indicate the download will use the RTMP protocol."""
882 self.to_screen('RTMP download detected')
884 def _signature_cache_id(self, example_sig):
885 """ Return a string representation of a signature """
886 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
888 def _extract_signature_function(self, video_id, player_url, example_sig):
890 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
893 raise ExtractorError('Cannot identify player %r' % player_url)
894 player_type = id_m.group('ext')
895 player_id = id_m.group('id')
897 # Read from filesystem cache
898 func_id = '%s_%s_%s' % (
899 player_type, player_id, self._signature_cache_id(example_sig))
900 assert os.path.basename(func_id) == func_id
902 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
903 if cache_spec is not None:
904 return lambda s: ''.join(s[i] for i in cache_spec)
907 'Downloading player %s' % player_url
908 if self._downloader.params.get('verbose') else
909 'Downloading %s player %s' % (player_type, player_id)
911 if player_type == 'js':
912 code = self._download_webpage(
913 player_url, video_id,
915 errnote='Download of %s failed' % player_url)
916 res = self._parse_sig_js(code)
917 elif player_type == 'swf':
918 urlh = self._request_webpage(
919 player_url, video_id,
921 errnote='Download of %s failed' % player_url)
923 res = self._parse_sig_swf(code)
925 assert False, 'Invalid player type %r' % player_type
927 test_string = ''.join(map(compat_chr, range(len(example_sig))))
928 cache_res = res(test_string)
929 cache_spec = [ord(c) for c in cache_res]
931 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
934 def _print_sig_code(self, func, example_sig):
935 def gen_sig_code(idxs):
936 def _genslice(start, end, step):
937 starts = '' if start == 0 else str(start)
938 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
939 steps = '' if step == 1 else (':%d' % step)
940 return 's[%s%s%s]' % (starts, ends, steps)
943 # Quelch pyflakes warnings - start will be set when step is set
944 start = '(Never used)'
945 for i, prev in zip(idxs[1:], idxs[:-1]):
949 yield _genslice(start, prev, step)
952 if i - prev in [-1, 1]:
961 yield _genslice(start, i, step)
963 test_string = ''.join(map(compat_chr, range(len(example_sig))))
964 cache_res = func(test_string)
965 cache_spec = [ord(c) for c in cache_res]
966 expr_code = ' + '.join(gen_sig_code(cache_spec))
967 signature_id_tuple = '(%s)' % (
968 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
969 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
970 ' return %s\n') % (signature_id_tuple, expr_code)
971 self.to_screen('Extracted signature function:\n' + code)
973 def _parse_sig_js(self, jscode):
974 funcname = self._search_regex(
975 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
976 'Initial JS player signature function name')
978 jsi = JSInterpreter(jscode)
979 initial_function = jsi.extract_function(funcname)
980 return lambda s: initial_function([s])
982 def _parse_sig_swf(self, file_contents):
983 swfi = SWFInterpreter(file_contents)
984 TARGET_CLASSNAME = 'SignatureDecipher'
985 searched_class = swfi.extract_class(TARGET_CLASSNAME)
986 initial_function = swfi.extract_function(searched_class, 'decipher')
987 return lambda s: initial_function([s])
989 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
990 """Turn the encrypted s field into a working signature"""
992 if player_url is None:
993 raise ExtractorError('Cannot decrypt signature without player_url')
995 if player_url.startswith('//'):
996 player_url = 'https:' + player_url
998 player_id = (player_url, self._signature_cache_id(s))
999 if player_id not in self._player_cache:
1000 func = self._extract_signature_function(
1001 video_id, player_url, s
1003 self._player_cache[player_id] = func
1004 func = self._player_cache[player_id]
1005 if self._downloader.params.get('youtube_print_sig_code'):
1006 self._print_sig_code(func, s)
1008 except Exception as e:
1009 tb = traceback.format_exc()
1010 raise ExtractorError(
1011 'Signature extraction failed: ' + tb, cause=e)
1013 def _get_subtitles(self, video_id, webpage):
1015 subs_doc = self._download_xml(
1016 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1017 video_id, note=False)
1018 except ExtractorError as err:
1019 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1023 for track in subs_doc.findall('track'):
1024 lang = track.attrib['lang_code']
1025 if lang in sub_lang_list:
1028 for ext in self._SUBTITLE_FORMATS:
1029 params = compat_urllib_parse_urlencode({
1033 'name': track.attrib['name'].encode('utf-8'),
1035 sub_formats.append({
1036 'url': 'https://www.youtube.com/api/timedtext?' + params,
1039 sub_lang_list[lang] = sub_formats
1040 if not sub_lang_list:
1041 self._downloader.report_warning('video doesn\'t have subtitles')
1043 return sub_lang_list
1045 def _get_ytplayer_config(self, video_id, webpage):
1047 # User data may contain arbitrary character sequences that may affect
1048 # JSON extraction with regex, e.g. when '};' is contained the second
1049 # regex won't capture the whole JSON. Yet working around by trying more
1050 # concrete regex first keeping in mind proper quoted string handling
1051 # to be implemented in future that will replace this workaround (see
1052 # https://github.com/rg3/youtube-dl/issues/7468,
1053 # https://github.com/rg3/youtube-dl/pull/7599)
1054 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1055 r';ytplayer\.config\s*=\s*({.+?});',
1057 config = self._search_regex(
1058 patterns, webpage, 'ytplayer.config', default=None)
1060 return self._parse_json(
1061 uppercase_escape(config), video_id, fatal=False)
1063 def _get_automatic_captions(self, video_id, webpage):
1064 """We need the webpage for getting the captions url, pass it as an
1065 argument to speed up the process."""
1066 self.to_screen('%s: Looking for automatic captions' % video_id)
1067 player_config = self._get_ytplayer_config(video_id, webpage)
1068 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1069 if not player_config:
1070 self._downloader.report_warning(err_msg)
1073 args = player_config['args']
1074 caption_url = args.get('ttsurl')
1076 timestamp = args['timestamp']
1077 # We get the available subtitles
1078 list_params = compat_urllib_parse_urlencode({
1083 list_url = caption_url + '&' + list_params
1084 caption_list = self._download_xml(list_url, video_id)
1085 original_lang_node = caption_list.find('track')
1086 if original_lang_node is None:
1087 self._downloader.report_warning('Video doesn\'t have automatic captions')
1089 original_lang = original_lang_node.attrib['lang_code']
1090 caption_kind = original_lang_node.attrib.get('kind', '')
1093 for lang_node in caption_list.findall('target'):
1094 sub_lang = lang_node.attrib['lang_code']
1096 for ext in self._SUBTITLE_FORMATS:
1097 params = compat_urllib_parse_urlencode({
1098 'lang': original_lang,
1102 'kind': caption_kind,
1104 sub_formats.append({
1105 'url': caption_url + '&' + params,
1108 sub_lang_list[sub_lang] = sub_formats
1109 return sub_lang_list
1111 # Some videos don't provide ttsurl but rather caption_tracks and
1112 # caption_translation_languages (e.g. 20LmZk1hakA)
1113 caption_tracks = args['caption_tracks']
1114 caption_translation_languages = args['caption_translation_languages']
1115 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1116 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1117 caption_qs = compat_parse_qs(parsed_caption_url.query)
1120 for lang in caption_translation_languages.split(','):
1121 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1122 sub_lang = lang_qs.get('lc', [None])[0]
1126 for ext in self._SUBTITLE_FORMATS:
1128 'tlang': [sub_lang],
1131 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1132 query=compat_urllib_parse_urlencode(caption_qs, True)))
1133 sub_formats.append({
1137 sub_lang_list[sub_lang] = sub_formats
1138 return sub_lang_list
1139 # An extractor error can be raise by the download process if there are
1140 # no automatic captions but there are subtitles
1141 except (KeyError, ExtractorError):
1142 self._downloader.report_warning(err_msg)
1145 def _mark_watched(self, video_id, video_info):
1146 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1147 if not playback_url:
1149 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1150 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1152 # cpn generation algorithm is reverse engineered from base.js.
1153 # In fact it works even with dummy cpn.
1154 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1155 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1161 playback_url = compat_urlparse.urlunparse(
1162 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1164 self._download_webpage(
1165 playback_url, video_id, 'Marking watched',
1166 'Unable to mark watched', fatal=False)
1169 def extract_id(cls, url):
1170 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1172 raise ExtractorError('Invalid URL: %s' % url)
1173 video_id = mobj.group(2)
1176 def _extract_from_m3u8(self, manifest_url, video_id):
1179 def _get_urls(_manifest):
1180 lines = _manifest.split('\n')
1181 urls = filter(lambda l: l and not l.startswith('#'),
1184 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1185 formats_urls = _get_urls(manifest)
1186 for format_url in formats_urls:
1187 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1188 url_map[itag] = format_url
1191 def _extract_annotations(self, video_id):
1192 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1193 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1195 def _real_extract(self, url):
1196 url, smuggled_data = unsmuggle_url(url, {})
1199 'http' if self._downloader.params.get('prefer_insecure', False)
1204 parsed_url = compat_urllib_parse_urlparse(url)
1205 for component in [parsed_url.fragment, parsed_url.query]:
1206 query = compat_parse_qs(component)
1207 if start_time is None and 't' in query:
1208 start_time = parse_duration(query['t'][0])
1209 if start_time is None and 'start' in query:
1210 start_time = parse_duration(query['start'][0])
1211 if end_time is None and 'end' in query:
1212 end_time = parse_duration(query['end'][0])
1214 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1215 mobj = re.search(self._NEXT_URL_RE, url)
1217 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1218 video_id = self.extract_id(url)
1221 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1222 video_webpage = self._download_webpage(url, video_id)
1224 # Attempt to extract SWF player URL
1225 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1226 if mobj is not None:
1227 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1233 def add_dash_mpd(video_info):
1234 dash_mpd = video_info.get('dashmpd')
1235 if dash_mpd and dash_mpd[0] not in dash_mpds:
1236 dash_mpds.append(dash_mpd[0])
1239 embed_webpage = None
1241 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1243 # We simulate the access to the video from www.youtube.com/v/{video_id}
1244 # this can be viewed without login into Youtube
1245 url = proto + '://www.youtube.com/embed/%s' % video_id
1246 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1247 data = compat_urllib_parse_urlencode({
1248 'video_id': video_id,
1249 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1250 'sts': self._search_regex(
1251 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1253 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1254 video_info_webpage = self._download_webpage(
1255 video_info_url, video_id,
1256 note='Refetching age-gated info webpage',
1257 errnote='unable to download video info webpage')
1258 video_info = compat_parse_qs(video_info_webpage)
1259 add_dash_mpd(video_info)
1263 # Try looking directly into the video webpage
1264 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1266 args = ytplayer_config['args']
1267 if args.get('url_encoded_fmt_stream_map'):
1268 # Convert to the same format returned by compat_parse_qs
1269 video_info = dict((k, [v]) for k, v in args.items())
1270 add_dash_mpd(video_info)
1271 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1273 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1274 # We also try looking in get_video_info since it may contain different dashmpd
1275 # URL that points to a DASH manifest with possibly different itag set (some itags
1276 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1277 # manifest pointed by get_video_info's dashmpd).
1278 # The general idea is to take a union of itags of both DASH manifests (for example
1279 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1280 self.report_video_info_webpage_download(video_id)
1281 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1283 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1284 % (proto, video_id, el_type))
1285 video_info_webpage = self._download_webpage(
1287 video_id, note=False,
1288 errnote='unable to download video info webpage')
1289 get_video_info = compat_parse_qs(video_info_webpage)
1290 if get_video_info.get('use_cipher_signature') != ['True']:
1291 add_dash_mpd(get_video_info)
1293 video_info = get_video_info
1294 if 'token' in get_video_info:
1295 # Different get_video_info requests may report different results, e.g.
1296 # some may report video unavailability, but some may serve it without
1297 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1298 # the original webpage as well as el=info and el=embedded get_video_info
1299 # requests report video unavailability due to geo restriction while
1300 # el=detailpage succeeds and returns valid data). This is probably
1301 # due to YouTube measures against IP ranges of hosting providers.
1302 # Working around by preferring the first succeeded video_info containing
1303 # the token if no such video_info yet was found.
1304 if 'token' not in video_info:
1305 video_info = get_video_info
1307 if 'token' not in video_info:
1308 if 'reason' in video_info:
1309 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1310 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1312 raise ExtractorError('YouTube said: This video is available in %s only' % (
1313 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1315 raise ExtractorError(
1316 'YouTube said: %s' % video_info['reason'][0],
1317 expected=True, video_id=video_id)
1319 raise ExtractorError(
1320 '"token" parameter not in video info for unknown reason',
1324 if 'title' in video_info:
1325 video_title = video_info['title'][0]
1327 self._downloader.report_warning('Unable to extract video title')
1331 video_description = get_element_by_id("eow-description", video_webpage)
1332 if video_description:
1333 video_description = re.sub(r'''(?x)
1335 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1336 (?:title|href)="([^"]+)"\s+
1337 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1341 ''', r'\1', video_description)
1342 video_description = clean_html(video_description)
1344 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1346 video_description = unescapeHTML(fd_mobj.group(1))
1348 video_description = ''
1350 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1351 if not self._downloader.params.get('noplaylist'):
1354 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1355 for feed in multifeed_metadata_list.split(','):
1356 # Unquote should take place before split on comma (,) since textual
1357 # fields may contain comma as well (see
1358 # https://github.com/rg3/youtube-dl/issues/8536)
1359 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1361 '_type': 'url_transparent',
1362 'ie_key': 'Youtube',
1364 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1365 {'force_singlefeed': True}),
1366 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1368 feed_ids.append(feed_data['id'][0])
1370 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1371 % (', '.join(feed_ids), video_id))
1372 return self.playlist_result(entries, video_id, video_title, video_description)
1373 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1375 if 'view_count' in video_info:
1376 view_count = int(video_info['view_count'][0])
1380 # Check for "rental" videos
1381 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1382 raise ExtractorError('"rental" videos not supported')
1384 # Start extracting information
1385 self.report_information_extraction(video_id)
1388 if 'author' not in video_info:
1389 raise ExtractorError('Unable to extract uploader name')
1390 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1393 video_uploader_id = None
1394 video_uploader_url = None
1396 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1398 if mobj is not None:
1399 video_uploader_id = mobj.group('uploader_id')
1400 video_uploader_url = mobj.group('uploader_url')
1402 self._downloader.report_warning('unable to extract uploader nickname')
1405 # We try first to get a high quality image:
1406 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1407 video_webpage, re.DOTALL)
1408 if m_thumb is not None:
1409 video_thumbnail = m_thumb.group(1)
1410 elif 'thumbnail_url' not in video_info:
1411 self._downloader.report_warning('unable to extract video thumbnail')
1412 video_thumbnail = None
1413 else: # don't panic if we can't find it
1414 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1417 upload_date = self._html_search_meta(
1418 'datePublished', video_webpage, 'upload date', default=None)
1420 upload_date = self._search_regex(
1421 [r'(?s)id="eow-date.*?>(.*?)</span>',
1422 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1423 video_webpage, 'upload date', default=None)
1425 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1426 upload_date = unified_strdate(upload_date)
1428 video_license = self._html_search_regex(
1429 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1430 video_webpage, 'license', default=None)
1432 m_music = re.search(
1433 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1436 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1437 video_creator = clean_html(m_music.group('creator'))
1439 video_alt_title = video_creator = None
1441 m_cat_container = self._search_regex(
1442 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1443 video_webpage, 'categories', default=None)
1445 category = self._html_search_regex(
1446 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1448 video_categories = None if category is None else [category]
1450 video_categories = None
1453 unescapeHTML(m.group('content'))
1454 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1456 def _extract_count(count_name):
1457 return str_to_int(self._search_regex(
1458 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1459 % re.escape(count_name),
1460 video_webpage, count_name, default=None))
1462 like_count = _extract_count('like')
1463 dislike_count = _extract_count('dislike')
1466 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1467 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1469 if 'length_seconds' not in video_info:
1470 self._downloader.report_warning('unable to extract video duration')
1471 video_duration = None
1473 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1476 video_annotations = None
1477 if self._downloader.params.get('writeannotations', False):
1478 video_annotations = self._extract_annotations(video_id)
1480 def _map_to_format_list(urlmap):
1482 for itag, video_real_url in urlmap.items():
1485 'url': video_real_url,
1486 'player_url': player_url,
1488 if itag in self._formats:
1489 dct.update(self._formats[itag])
1493 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1494 self.report_rtmp_download()
1496 'format_id': '_rtmp',
1498 'url': video_info['conn'][0],
1499 'player_url': player_url,
1501 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1502 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1503 if 'rtmpe%3Dyes' in encoded_url_map:
1504 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1506 fmt_list = video_info.get('fmt_list', [''])[0]
1508 for fmt in fmt_list.split(','):
1509 spec = fmt.split('/')
1511 width_height = spec[1].split('x')
1512 if len(width_height) == 2:
1513 formats_spec[spec[0]] = {
1514 'resolution': spec[1],
1515 'width': int_or_none(width_height[0]),
1516 'height': int_or_none(width_height[1]),
1519 for url_data_str in encoded_url_map.split(','):
1520 url_data = compat_parse_qs(url_data_str)
1521 if 'itag' not in url_data or 'url' not in url_data:
1523 format_id = url_data['itag'][0]
1524 url = url_data['url'][0]
1526 if 'sig' in url_data:
1527 url += '&signature=' + url_data['sig'][0]
1528 elif 's' in url_data:
1529 encrypted_sig = url_data['s'][0]
1530 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1532 jsplayer_url_json = self._search_regex(
1534 embed_webpage if age_gate else video_webpage,
1535 'JS player URL (1)', default=None)
1536 if not jsplayer_url_json and not age_gate:
1537 # We need the embed website after all
1538 if embed_webpage is None:
1539 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1540 embed_webpage = self._download_webpage(
1541 embed_url, video_id, 'Downloading embed webpage')
1542 jsplayer_url_json = self._search_regex(
1543 ASSETS_RE, embed_webpage, 'JS player URL')
1545 player_url = json.loads(jsplayer_url_json)
1546 if player_url is None:
1547 player_url_json = self._search_regex(
1548 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1549 video_webpage, 'age gate player URL')
1550 player_url = json.loads(player_url_json)
1552 if self._downloader.params.get('verbose'):
1553 if player_url is None:
1554 player_version = 'unknown'
1555 player_desc = 'unknown'
1557 if player_url.endswith('swf'):
1558 player_version = self._search_regex(
1559 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1560 'flash player', fatal=False)
1561 player_desc = 'flash player %s' % player_version
1563 player_version = self._search_regex(
1564 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1566 'html5 player', fatal=False)
1567 player_desc = 'html5 player %s' % player_version
1569 parts_sizes = self._signature_cache_id(encrypted_sig)
1570 self.to_screen('{%s} signature length %s, %s' %
1571 (format_id, parts_sizes, player_desc))
1573 signature = self._decrypt_signature(
1574 encrypted_sig, video_id, player_url, age_gate)
1575 url += '&signature=' + signature
1576 if 'ratebypass' not in url:
1577 url += '&ratebypass=yes'
1580 'format_id': format_id,
1582 'player_url': player_url,
1584 if format_id in self._formats:
1585 dct.update(self._formats[format_id])
1586 if format_id in formats_spec:
1587 dct.update(formats_spec[format_id])
1589 # Some itags are not included in DASH manifest thus corresponding formats will
1590 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1591 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1592 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1593 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1596 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1597 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1600 'fps': int_or_none(url_data.get('fps', [None])[0]),
1601 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1603 for key, value in more_fields.items():
1606 type_ = url_data.get('type', [None])[0]
1608 type_split = type_.split(';')
1609 kind_ext = type_split[0].split('/')
1610 if len(kind_ext) == 2:
1612 dct['ext'] = mimetype2ext(type_split[0])
1613 if kind in ('audio', 'video'):
1615 for mobj in re.finditer(
1616 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1617 if mobj.group('key') == 'codecs':
1618 codecs = mobj.group('val')
1621 codecs = codecs.split(',')
1622 if len(codecs) == 2:
1623 acodec, vcodec = codecs[1], codecs[0]
1625 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1631 elif video_info.get('hlsvp'):
1632 manifest_url = video_info['hlsvp'][0]
1633 url_map = self._extract_from_m3u8(manifest_url, video_id)
1634 formats = _map_to_format_list(url_map)
1635 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1636 for a_format in formats:
1637 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1639 unavailable_message = self._html_search_regex(
1640 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1641 video_webpage, 'unavailable message', default=None)
1642 if unavailable_message:
1643 raise ExtractorError(unavailable_message, expected=True)
1644 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1646 # Look for the DASH manifest
1647 if self._downloader.params.get('youtube_include_dash_manifest', True):
1648 dash_mpd_fatal = True
1649 for mpd_url in dash_mpds:
1652 def decrypt_sig(mobj):
1654 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1655 return '/signature/%s' % dec_s
1657 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1659 for df in self._extract_mpd_formats(
1660 mpd_url, video_id, fatal=dash_mpd_fatal,
1661 formats_dict=self._formats):
1662 # Do not overwrite DASH format found in some previous DASH manifest
1663 if df['format_id'] not in dash_formats:
1664 dash_formats[df['format_id']] = df
1665 # Additional DASH manifests may end up in HTTP Error 403 therefore
1666 # allow them to fail without bug report message if we already have
1667 # some DASH manifest succeeded. This is temporary workaround to reduce
1668 # burst of bug reports until we figure out the reason and whether it
1669 # can be fixed at all.
1670 dash_mpd_fatal = False
1671 except (ExtractorError, KeyError) as e:
1672 self.report_warning(
1673 'Skipping DASH manifest: %r' % e, video_id)
1675 # Remove the formats we found through non-DASH, they
1676 # contain less info and it can be wrong, because we use
1677 # fixed values (for example the resolution). See
1678 # https://github.com/rg3/youtube-dl/issues/5774 for an
1680 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1681 formats.extend(dash_formats.values())
1683 # Check for malformed aspect ratio
1684 stretched_m = re.search(
1685 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1688 w = float(stretched_m.group('w'))
1689 h = float(stretched_m.group('h'))
1690 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1691 # We will only process correct ratios.
1695 if f.get('vcodec') != 'none':
1696 f['stretched_ratio'] = ratio
1698 self._sort_formats(formats)
1700 self.mark_watched(video_id, video_info)
1704 'uploader': video_uploader,
1705 'uploader_id': video_uploader_id,
1706 'uploader_url': video_uploader_url,
1707 'upload_date': upload_date,
1708 'license': video_license,
1709 'creator': video_creator,
1710 'title': video_title,
1711 'alt_title': video_alt_title,
1712 'thumbnail': video_thumbnail,
1713 'description': video_description,
1714 'categories': video_categories,
1716 'subtitles': video_subtitles,
1717 'automatic_captions': automatic_captions,
1718 'duration': video_duration,
1719 'age_limit': 18 if age_gate else 0,
1720 'annotations': video_annotations,
1721 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1722 'view_count': view_count,
1723 'like_count': like_count,
1724 'dislike_count': dislike_count,
1725 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1728 'start_time': start_time,
1729 'end_time': end_time,
1733 class YoutubeSharedVideoIE(InfoExtractor):
1734 _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?ci=(?P<id>[0-9A-Za-z_-]{11})'
1735 IE_NAME = 'youtube:shared'
1738 'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
1740 'id': 'uPDB5I9wfp8',
1742 'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
1743 'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
1744 'upload_date': '20160219',
1745 'uploader': 'Pocoyo - Português (BR)',
1746 'uploader_id': 'PocoyoBrazil',
1748 'add_ie': ['Youtube'],
1750 # There are already too many Youtube downloads
1751 'skip_download': True,
1755 def _real_extract(self, url):
1756 video_id = self._match_id(url)
1758 webpage = self._download_webpage(url, video_id)
1760 real_video_id = self._html_search_meta(
1761 'videoId', webpage, 'YouTube video id', fatal=True)
1763 return self.url_result(real_video_id, YoutubeIE.ie_key())
1766 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1767 IE_DESC = 'YouTube.com playlists'
1768 _VALID_URL = r"""(?x)(?:
1773 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1774 \? (?:.*?[&;])*? (?:p|a|list)=
1778 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1779 # Top tracks, they can also include dots
1784 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1786 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1787 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1788 IE_NAME = 'youtube:playlist'
1790 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1792 'title': 'ytdl test PL',
1793 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1795 'playlist_count': 3,
1797 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1799 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1800 'title': 'YDL_Empty_List',
1802 'playlist_count': 0,
1804 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1805 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1807 'title': '29C3: Not my department',
1808 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1810 'playlist_count': 95,
1812 'note': 'issue #673',
1813 'url': 'PLBB231211A4F62143',
1815 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1816 'id': 'PLBB231211A4F62143',
1818 'playlist_mincount': 26,
1820 'note': 'Large playlist',
1821 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1823 'title': 'Uploads from Cauchemar',
1824 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1826 'playlist_mincount': 799,
1828 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1830 'title': 'YDL_safe_search',
1831 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1833 'playlist_count': 2,
1836 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1837 'playlist_count': 4,
1840 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1843 'note': 'Embedded SWF player',
1844 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1845 'playlist_count': 4,
1848 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1851 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1852 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1854 'title': 'Uploads from Interstellar Movie',
1855 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1857 'playlist_mincout': 21,
1860 def _real_initialize(self):
1863 def _extract_mix(self, playlist_id):
1864 # The mixes are generated from a single video
1865 # the id of the playlist is just 'RD' + video_id
1867 last_id = playlist_id[-11:]
1868 for n in itertools.count(1):
1869 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1870 webpage = self._download_webpage(
1871 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1872 new_ids = orderedSet(re.findall(
1873 r'''(?xs)data-video-username=".*?".*?
1874 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1876 # Fetch new pages until all the videos are repeated, it seems that
1877 # there are always 51 unique videos.
1878 new_ids = [_id for _id in new_ids if _id not in ids]
1884 url_results = self._ids_to_results(ids)
1886 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1888 search_title('playlist-title') or
1889 search_title('title long-title') or
1890 search_title('title'))
1891 title = clean_html(title_span)
1893 return self.playlist_result(url_results, playlist_id, title)
1895 def _extract_playlist(self, playlist_id):
1896 url = self._TEMPLATE_URL % playlist_id
1897 page = self._download_webpage(url, playlist_id)
1899 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1900 match = match.strip()
1901 # Check if the playlist exists or is private
1902 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1903 raise ExtractorError(
1904 'The playlist doesn\'t exist or is private, use --username or '
1905 '--netrc to access it.',
1907 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1908 raise ExtractorError(
1909 'Invalid parameters. Maybe URL is incorrect.',
1911 elif re.match(r'[^<]*Choose your language[^<]*', match):
1914 self.report_warning('Youtube gives an alert message: ' + match)
1916 playlist_title = self._html_search_regex(
1917 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1920 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1922 def _check_download_just_video(self, url, playlist_id):
1923 # Check if it's a video-specific URL
1924 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1925 if 'v' in query_dict:
1926 video_id = query_dict['v'][0]
1927 if self._downloader.params.get('noplaylist'):
1928 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1929 return self.url_result(video_id, 'Youtube', video_id=video_id)
1931 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1933 def _real_extract(self, url):
1934 # Extract playlist id
1935 mobj = re.match(self._VALID_URL, url)
1937 raise ExtractorError('Invalid URL: %s' % url)
1938 playlist_id = mobj.group(1) or mobj.group(2)
1940 video = self._check_download_just_video(url, playlist_id)
1944 if playlist_id.startswith(('RD', 'UL', 'PU')):
1945 # Mixes require a custom extraction process
1946 return self._extract_mix(playlist_id)
1948 return self._extract_playlist(playlist_id)
1951 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1952 IE_DESC = 'YouTube.com channels'
1953 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1954 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1955 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1956 IE_NAME = 'youtube:channel'
1958 'note': 'paginated channel',
1959 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1960 'playlist_mincount': 91,
1962 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1963 'title': 'Uploads from lex will',
1966 'note': 'Age restricted channel',
1967 # from https://www.youtube.com/user/DeusExOfficial
1968 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1969 'playlist_mincount': 64,
1971 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1972 'title': 'Uploads from Deus Ex',
1977 def suitable(cls, url):
1978 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1979 else super(YoutubeChannelIE, cls).suitable(url))
1981 def _build_template_url(self, url, channel_id):
1982 return self._TEMPLATE_URL % channel_id
1984 def _real_extract(self, url):
1985 channel_id = self._match_id(url)
1987 url = self._build_template_url(url, channel_id)
1989 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1990 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1991 # otherwise fallback on channel by page extraction
1992 channel_page = self._download_webpage(
1993 url + '?view=57', channel_id,
1994 'Downloading channel page', fatal=False)
1995 if channel_page is False:
1996 channel_playlist_id = False
1998 channel_playlist_id = self._html_search_meta(
1999 'channelId', channel_page, 'channel id', default=None)
2000 if not channel_playlist_id:
2001 channel_url = self._html_search_meta(
2002 ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2003 channel_page, 'channel url', default=None)
2005 channel_playlist_id = self._search_regex(
2006 r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2007 channel_url, 'channel id', default=None)
2008 if channel_playlist_id and channel_playlist_id.startswith('UC'):
2009 playlist_id = 'UU' + channel_playlist_id[2:]
2010 return self.url_result(
2011 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2013 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2014 autogenerated = re.search(r'''(?x)
2016 channel-header-autogenerated-label|
2017 yt-channel-title-autogenerated
2018 )[^"]*"''', channel_page) is not None
2021 # The videos are contained in a single page
2022 # the ajax pages can't be used, they are empty
2025 video_id, 'Youtube', video_id=video_id,
2026 video_title=video_title)
2027 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2028 return self.playlist_result(entries, channel_id)
2031 next(self._entries(channel_page, channel_id))
2032 except StopIteration:
2033 alert_message = self._html_search_regex(
2034 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2035 channel_page, 'alert', default=None, group='alert')
2037 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2039 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2042 class YoutubeUserIE(YoutubeChannelIE):
2043 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2044 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2045 _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2046 IE_NAME = 'youtube:user'
2049 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2050 'playlist_mincount': 320,
2052 'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2053 'title': 'Uploads from The Linux Foundation',
2056 # Only available via https://www.youtube.com/c/12minuteathlete/videos
2057 # but not https://www.youtube.com/user/12minuteathlete/videos
2058 'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2059 'playlist_mincount': 249,
2061 'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2062 'title': 'Uploads from 12 Minute Athlete',
2065 'url': 'ytuser:phihag',
2066 'only_matching': True,
2068 'url': 'https://www.youtube.com/c/gametrailers',
2069 'only_matching': True,
2071 'url': 'https://www.youtube.com/gametrailers',
2072 'only_matching': True,
2074 # This channel is not available.
2075 'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2076 'only_matching': True,
2080 def suitable(cls, url):
2081 # Don't return True if the url can be extracted with other youtube
2082 # extractor, the regex would is too permissive and it would match.
2083 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2084 if any(ie.suitable(url) for ie in other_yt_ies):
2087 return super(YoutubeUserIE, cls).suitable(url)
2089 def _build_template_url(self, url, channel_id):
2090 mobj = re.match(self._VALID_URL, url)
2091 return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2094 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2095 IE_DESC = 'YouTube.com live streams'
2096 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2097 IE_NAME = 'youtube:live'
2100 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2102 'id': 'a48o2S1cPoo',
2104 'title': 'The Young Turks - Live Main Show',
2105 'uploader': 'The Young Turks',
2106 'uploader_id': 'TheYoungTurks',
2107 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2108 'upload_date': '20150715',
2109 'license': 'Standard YouTube License',
2110 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2111 'categories': ['News & Politics'],
2112 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2114 'dislike_count': int,
2117 'skip_download': True,
2120 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2121 'only_matching': True,
2124 def _real_extract(self, url):
2125 mobj = re.match(self._VALID_URL, url)
2126 channel_id = mobj.group('id')
2127 base_url = mobj.group('base_url')
2128 webpage = self._download_webpage(url, channel_id, fatal=False)
2130 page_type = self._og_search_property(
2131 'type', webpage, 'page type', default=None)
2132 video_id = self._html_search_meta(
2133 'videoId', webpage, 'video id', default=None)
2134 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2135 return self.url_result(video_id, YoutubeIE.ie_key())
2136 return self.url_result(base_url)
2139 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2140 IE_DESC = 'YouTube.com user/channel playlists'
2141 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2142 IE_NAME = 'youtube:playlists'
2145 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2146 'playlist_mincount': 4,
2148 'id': 'ThirstForScience',
2149 'title': 'Thirst for Science',
2152 # with "Load more" button
2153 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2154 'playlist_mincount': 70,
2157 'title': 'Игорь Клейнер',
2160 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2161 'playlist_mincount': 17,
2163 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2164 'title': 'Chem Player',
2169 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2170 IE_DESC = 'YouTube.com searches'
2171 # there doesn't appear to be a real limit, for example if you search for
2172 # 'python' you get more than 8.000.000 results
2173 _MAX_RESULTS = float('inf')
2174 IE_NAME = 'youtube:search'
2175 _SEARCH_KEY = 'ytsearch'
2176 _EXTRA_QUERY_ARGS = {}
2179 def _get_n_results(self, query, n):
2180 """Get a specified number of results for a query"""
2185 for pagenum in itertools.count(1):
2187 'search_query': query.encode('utf-8'),
2191 url_query.update(self._EXTRA_QUERY_ARGS)
2192 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2193 data = self._download_json(
2194 result_url, video_id='query "%s"' % query,
2195 note='Downloading page %s' % pagenum,
2196 errnote='Unable to download API page')
2197 html_content = data[1]['body']['content']
2199 if 'class="search-message' in html_content:
2200 raise ExtractorError(
2201 '[youtube] No video results', expected=True)
2203 new_videos = self._ids_to_results(orderedSet(re.findall(
2204 r'href="/watch\?v=(.{11})', html_content)))
2205 videos += new_videos
2206 if not new_videos or len(videos) > limit:
2211 return self.playlist_result(videos, query)
2214 class YoutubeSearchDateIE(YoutubeSearchIE):
2215 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2216 _SEARCH_KEY = 'ytsearchdate'
2217 IE_DESC = 'YouTube.com searches, newest videos first'
2218 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2221 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2222 IE_DESC = 'YouTube.com search URLs'
2223 IE_NAME = 'youtube:search_url'
2224 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2225 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2227 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2228 'playlist_mincount': 5,
2230 'title': 'youtube-dl test video',
2233 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2234 'only_matching': True,
2237 def _real_extract(self, url):
2238 mobj = re.match(self._VALID_URL, url)
2239 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2240 webpage = self._download_webpage(url, query)
2241 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2244 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2245 IE_DESC = 'YouTube.com (multi-season) shows'
2246 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2247 IE_NAME = 'youtube:show'
2249 'url': 'https://www.youtube.com/show/airdisasters',
2250 'playlist_mincount': 5,
2252 'id': 'airdisasters',
2253 'title': 'Air Disasters',
2257 def _real_extract(self, url):
2258 playlist_id = self._match_id(url)
2259 return super(YoutubeShowIE, self)._real_extract(
2260 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2263 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2265 Base class for feed extractors
2266 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2268 _LOGIN_REQUIRED = True
2272 return 'youtube:%s' % self._FEED_NAME
2274 def _real_initialize(self):
2277 def _real_extract(self, url):
2278 page = self._download_webpage(
2279 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2281 # The extraction process is the same as for playlists, but the regex
2282 # for the video ids doesn't contain an index
2284 more_widget_html = content_html = page
2285 for page_num in itertools.count(1):
2286 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2288 # 'recommended' feed has infinite 'load more' and each new portion spins
2289 # the same videos in (sometimes) slightly different order, so we'll check
2290 # for unicity and break when portion has no new videos
2291 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2297 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2301 more = self._download_json(
2302 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2303 'Downloading page #%s' % page_num,
2304 transform_source=uppercase_escape)
2305 content_html = more['content_html']
2306 more_widget_html = more['load_more_widget_html']
2308 return self.playlist_result(
2309 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2312 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2313 IE_NAME = 'youtube:watchlater'
2314 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2315 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2318 'url': 'https://www.youtube.com/playlist?list=WL',
2319 'only_matching': True,
2321 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2322 'only_matching': True,
2325 def _real_extract(self, url):
2326 video = self._check_download_just_video(url, 'WL')
2329 return self._extract_playlist('WL')
2332 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2333 IE_NAME = 'youtube:favorites'
2334 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2335 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2336 _LOGIN_REQUIRED = True
2338 def _real_extract(self, url):
2339 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2340 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2341 return self.url_result(playlist_id, 'YoutubePlaylist')
2344 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2345 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2346 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2347 _FEED_NAME = 'recommended'
2348 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2351 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2352 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2353 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2354 _FEED_NAME = 'subscriptions'
2355 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2358 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2359 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2360 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2361 _FEED_NAME = 'history'
2362 _PLAYLIST_TITLE = 'Youtube History'
2365 class YoutubeTruncatedURLIE(InfoExtractor):
2366 IE_NAME = 'youtube:truncated_url'
2367 IE_DESC = False # Do not list
2368 _VALID_URL = r'''(?x)
2370 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2373 annotation_id=annotation_[^&]+|
2379 attribution_link\?a=[^&]+
2385 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2386 'only_matching': True,
2388 'url': 'http://www.youtube.com/watch?',
2389 'only_matching': True,
2391 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2392 'only_matching': True,
2394 'url': 'https://www.youtube.com/watch?feature=foo',
2395 'only_matching': True,
2397 'url': 'https://www.youtube.com/watch?hl=en-GB',
2398 'only_matching': True,
2400 'url': 'https://www.youtube.com/watch?t=2372',
2401 'only_matching': True,
2404 def _real_extract(self, url):
2405 raise ExtractorError(
2406 'Did you forget to quote the URL? Remember that & is a meta '
2407 'character in most shells, so you want to put the URL in quotes, '
2409 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2410 ' or simply youtube-dl BaW_jenozKc .',
2414 class YoutubeTruncatedIDIE(InfoExtractor):
2415 IE_NAME = 'youtube:truncated_id'
2416 IE_DESC = False # Do not list
2417 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2420 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2421 'only_matching': True,
2424 def _real_extract(self, url):
2425 video_id = self._match_id(url)
2426 raise ExtractorError(
2427 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),