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': 'Olympics',
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:33',
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
698 # Multifeed videos (multiple cameras), URL is for Main Camera
699 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
702 'title': 'teamPGP: Rocket League Noob Stream',
703 'description': 'md5:dc7872fb300e143831327f1bae3af010',
709 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
710 'description': 'md5:dc7872fb300e143831327f1bae3af010',
711 'upload_date': '20150721',
712 'uploader': 'Beer Games Beer',
713 'uploader_id': 'beergamesbeer',
714 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
715 'license': 'Standard YouTube License',
721 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
722 'description': 'md5:dc7872fb300e143831327f1bae3af010',
723 'upload_date': '20150721',
724 'uploader': 'Beer Games Beer',
725 'uploader_id': 'beergamesbeer',
726 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
727 'license': 'Standard YouTube License',
733 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
734 'description': 'md5:dc7872fb300e143831327f1bae3af010',
735 'upload_date': '20150721',
736 'uploader': 'Beer Games Beer',
737 'uploader_id': 'beergamesbeer',
738 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
739 'license': 'Standard YouTube License',
745 'title': 'teamPGP: Rocket League Noob Stream (zim)',
746 'description': 'md5:dc7872fb300e143831327f1bae3af010',
747 'upload_date': '20150721',
748 'uploader': 'Beer Games Beer',
749 'uploader_id': 'beergamesbeer',
750 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
751 'license': 'Standard YouTube License',
755 'skip_download': True,
759 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
760 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
763 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
768 'url': 'http://vid.plus/FlRa-iH7PGw',
769 'only_matching': True,
772 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
773 'only_matching': True,
776 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
777 # Also tests cut-off URL expansion in video description (see
778 # https://github.com/rg3/youtube-dl/issues/1892,
779 # https://github.com/rg3/youtube-dl/issues/8164)
780 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
784 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
785 'alt_title': 'Dark Walk',
786 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
787 'upload_date': '20151119',
788 'uploader_id': 'IronSoulElf',
789 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
790 'uploader': 'IronSoulElf',
791 'license': 'Standard YouTube License',
792 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
795 'skip_download': True,
799 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
800 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
801 'only_matching': True,
804 # Video with yt:stretch=17:0
805 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
809 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
810 'description': 'md5:ee18a25c350637c8faff806845bddee9',
811 'upload_date': '20151107',
812 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
813 'uploader': 'CH GAMER DROID',
816 'skip_download': True,
820 # Video licensed under Creative Commons
821 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
825 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
826 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
827 'upload_date': '20150127',
828 'uploader_id': 'BerkmanCenter',
829 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
830 'uploader': 'BerkmanCenter',
831 'license': 'Creative Commons Attribution license (reuse allowed)',
834 'skip_download': True,
838 # Channel-like uploader_url
839 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
843 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
844 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
845 'upload_date': '20151119',
846 'uploader': 'Bernie 2016',
847 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
848 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
849 'license': 'Creative Commons Attribution license (reuse allowed)',
852 'skip_download': True,
856 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
857 'only_matching': True,
861 def __init__(self, *args, **kwargs):
862 super(YoutubeIE, self).__init__(*args, **kwargs)
863 self._player_cache = {}
865 def report_video_info_webpage_download(self, video_id):
866 """Report attempt to download video info webpage."""
867 self.to_screen('%s: Downloading video info webpage' % video_id)
869 def report_information_extraction(self, video_id):
870 """Report attempt to extract video information."""
871 self.to_screen('%s: Extracting video information' % video_id)
873 def report_unavailable_format(self, video_id, format):
874 """Report extracted video URL."""
875 self.to_screen('%s: Format %s not available' % (video_id, format))
877 def report_rtmp_download(self):
878 """Indicate the download will use the RTMP protocol."""
879 self.to_screen('RTMP download detected')
881 def _signature_cache_id(self, example_sig):
882 """ Return a string representation of a signature """
883 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
885 def _extract_signature_function(self, video_id, player_url, example_sig):
887 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
890 raise ExtractorError('Cannot identify player %r' % player_url)
891 player_type = id_m.group('ext')
892 player_id = id_m.group('id')
894 # Read from filesystem cache
895 func_id = '%s_%s_%s' % (
896 player_type, player_id, self._signature_cache_id(example_sig))
897 assert os.path.basename(func_id) == func_id
899 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
900 if cache_spec is not None:
901 return lambda s: ''.join(s[i] for i in cache_spec)
904 'Downloading player %s' % player_url
905 if self._downloader.params.get('verbose') else
906 'Downloading %s player %s' % (player_type, player_id)
908 if player_type == 'js':
909 code = self._download_webpage(
910 player_url, video_id,
912 errnote='Download of %s failed' % player_url)
913 res = self._parse_sig_js(code)
914 elif player_type == 'swf':
915 urlh = self._request_webpage(
916 player_url, video_id,
918 errnote='Download of %s failed' % player_url)
920 res = self._parse_sig_swf(code)
922 assert False, 'Invalid player type %r' % player_type
924 test_string = ''.join(map(compat_chr, range(len(example_sig))))
925 cache_res = res(test_string)
926 cache_spec = [ord(c) for c in cache_res]
928 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
931 def _print_sig_code(self, func, example_sig):
932 def gen_sig_code(idxs):
933 def _genslice(start, end, step):
934 starts = '' if start == 0 else str(start)
935 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
936 steps = '' if step == 1 else (':%d' % step)
937 return 's[%s%s%s]' % (starts, ends, steps)
940 # Quelch pyflakes warnings - start will be set when step is set
941 start = '(Never used)'
942 for i, prev in zip(idxs[1:], idxs[:-1]):
946 yield _genslice(start, prev, step)
949 if i - prev in [-1, 1]:
958 yield _genslice(start, i, step)
960 test_string = ''.join(map(compat_chr, range(len(example_sig))))
961 cache_res = func(test_string)
962 cache_spec = [ord(c) for c in cache_res]
963 expr_code = ' + '.join(gen_sig_code(cache_spec))
964 signature_id_tuple = '(%s)' % (
965 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
966 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
967 ' return %s\n') % (signature_id_tuple, expr_code)
968 self.to_screen('Extracted signature function:\n' + code)
970 def _parse_sig_js(self, jscode):
971 funcname = self._search_regex(
972 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
973 'Initial JS player signature function name')
975 jsi = JSInterpreter(jscode)
976 initial_function = jsi.extract_function(funcname)
977 return lambda s: initial_function([s])
979 def _parse_sig_swf(self, file_contents):
980 swfi = SWFInterpreter(file_contents)
981 TARGET_CLASSNAME = 'SignatureDecipher'
982 searched_class = swfi.extract_class(TARGET_CLASSNAME)
983 initial_function = swfi.extract_function(searched_class, 'decipher')
984 return lambda s: initial_function([s])
986 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
987 """Turn the encrypted s field into a working signature"""
989 if player_url is None:
990 raise ExtractorError('Cannot decrypt signature without player_url')
992 if player_url.startswith('//'):
993 player_url = 'https:' + player_url
995 player_id = (player_url, self._signature_cache_id(s))
996 if player_id not in self._player_cache:
997 func = self._extract_signature_function(
998 video_id, player_url, s
1000 self._player_cache[player_id] = func
1001 func = self._player_cache[player_id]
1002 if self._downloader.params.get('youtube_print_sig_code'):
1003 self._print_sig_code(func, s)
1005 except Exception as e:
1006 tb = traceback.format_exc()
1007 raise ExtractorError(
1008 'Signature extraction failed: ' + tb, cause=e)
1010 def _get_subtitles(self, video_id, webpage):
1012 subs_doc = self._download_xml(
1013 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1014 video_id, note=False)
1015 except ExtractorError as err:
1016 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1020 for track in subs_doc.findall('track'):
1021 lang = track.attrib['lang_code']
1022 if lang in sub_lang_list:
1025 for ext in self._SUBTITLE_FORMATS:
1026 params = compat_urllib_parse_urlencode({
1030 'name': track.attrib['name'].encode('utf-8'),
1032 sub_formats.append({
1033 'url': 'https://www.youtube.com/api/timedtext?' + params,
1036 sub_lang_list[lang] = sub_formats
1037 if not sub_lang_list:
1038 self._downloader.report_warning('video doesn\'t have subtitles')
1040 return sub_lang_list
1042 def _get_ytplayer_config(self, video_id, webpage):
1044 # User data may contain arbitrary character sequences that may affect
1045 # JSON extraction with regex, e.g. when '};' is contained the second
1046 # regex won't capture the whole JSON. Yet working around by trying more
1047 # concrete regex first keeping in mind proper quoted string handling
1048 # to be implemented in future that will replace this workaround (see
1049 # https://github.com/rg3/youtube-dl/issues/7468,
1050 # https://github.com/rg3/youtube-dl/pull/7599)
1051 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1052 r';ytplayer\.config\s*=\s*({.+?});',
1054 config = self._search_regex(
1055 patterns, webpage, 'ytplayer.config', default=None)
1057 return self._parse_json(
1058 uppercase_escape(config), video_id, fatal=False)
1060 def _get_automatic_captions(self, video_id, webpage):
1061 """We need the webpage for getting the captions url, pass it as an
1062 argument to speed up the process."""
1063 self.to_screen('%s: Looking for automatic captions' % video_id)
1064 player_config = self._get_ytplayer_config(video_id, webpage)
1065 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1066 if not player_config:
1067 self._downloader.report_warning(err_msg)
1070 args = player_config['args']
1071 caption_url = args.get('ttsurl')
1073 timestamp = args['timestamp']
1074 # We get the available subtitles
1075 list_params = compat_urllib_parse_urlencode({
1080 list_url = caption_url + '&' + list_params
1081 caption_list = self._download_xml(list_url, video_id)
1082 original_lang_node = caption_list.find('track')
1083 if original_lang_node is None:
1084 self._downloader.report_warning('Video doesn\'t have automatic captions')
1086 original_lang = original_lang_node.attrib['lang_code']
1087 caption_kind = original_lang_node.attrib.get('kind', '')
1090 for lang_node in caption_list.findall('target'):
1091 sub_lang = lang_node.attrib['lang_code']
1093 for ext in self._SUBTITLE_FORMATS:
1094 params = compat_urllib_parse_urlencode({
1095 'lang': original_lang,
1099 'kind': caption_kind,
1101 sub_formats.append({
1102 'url': caption_url + '&' + params,
1105 sub_lang_list[sub_lang] = sub_formats
1106 return sub_lang_list
1108 # Some videos don't provide ttsurl but rather caption_tracks and
1109 # caption_translation_languages (e.g. 20LmZk1hakA)
1110 caption_tracks = args['caption_tracks']
1111 caption_translation_languages = args['caption_translation_languages']
1112 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1113 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1114 caption_qs = compat_parse_qs(parsed_caption_url.query)
1117 for lang in caption_translation_languages.split(','):
1118 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1119 sub_lang = lang_qs.get('lc', [None])[0]
1123 for ext in self._SUBTITLE_FORMATS:
1125 'tlang': [sub_lang],
1128 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1129 query=compat_urllib_parse_urlencode(caption_qs, True)))
1130 sub_formats.append({
1134 sub_lang_list[sub_lang] = sub_formats
1135 return sub_lang_list
1136 # An extractor error can be raise by the download process if there are
1137 # no automatic captions but there are subtitles
1138 except (KeyError, ExtractorError):
1139 self._downloader.report_warning(err_msg)
1142 def _mark_watched(self, video_id, video_info):
1143 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1144 if not playback_url:
1146 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1147 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1149 # cpn generation algorithm is reverse engineered from base.js.
1150 # In fact it works even with dummy cpn.
1151 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1152 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1158 playback_url = compat_urlparse.urlunparse(
1159 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1161 self._download_webpage(
1162 playback_url, video_id, 'Marking watched',
1163 'Unable to mark watched', fatal=False)
1166 def extract_id(cls, url):
1167 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1169 raise ExtractorError('Invalid URL: %s' % url)
1170 video_id = mobj.group(2)
1173 def _extract_from_m3u8(self, manifest_url, video_id):
1176 def _get_urls(_manifest):
1177 lines = _manifest.split('\n')
1178 urls = filter(lambda l: l and not l.startswith('#'),
1181 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1182 formats_urls = _get_urls(manifest)
1183 for format_url in formats_urls:
1184 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1185 url_map[itag] = format_url
1188 def _extract_annotations(self, video_id):
1189 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1190 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1192 def _real_extract(self, url):
1193 url, smuggled_data = unsmuggle_url(url, {})
1196 'http' if self._downloader.params.get('prefer_insecure', False)
1201 parsed_url = compat_urllib_parse_urlparse(url)
1202 for component in [parsed_url.fragment, parsed_url.query]:
1203 query = compat_parse_qs(component)
1204 if start_time is None and 't' in query:
1205 start_time = parse_duration(query['t'][0])
1206 if start_time is None and 'start' in query:
1207 start_time = parse_duration(query['start'][0])
1208 if end_time is None and 'end' in query:
1209 end_time = parse_duration(query['end'][0])
1211 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1212 mobj = re.search(self._NEXT_URL_RE, url)
1214 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1215 video_id = self.extract_id(url)
1218 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1219 video_webpage = self._download_webpage(url, video_id)
1221 # Attempt to extract SWF player URL
1222 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1223 if mobj is not None:
1224 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1230 def add_dash_mpd(video_info):
1231 dash_mpd = video_info.get('dashmpd')
1232 if dash_mpd and dash_mpd[0] not in dash_mpds:
1233 dash_mpds.append(dash_mpd[0])
1236 embed_webpage = None
1238 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1240 # We simulate the access to the video from www.youtube.com/v/{video_id}
1241 # this can be viewed without login into Youtube
1242 url = proto + '://www.youtube.com/embed/%s' % video_id
1243 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1244 data = compat_urllib_parse_urlencode({
1245 'video_id': video_id,
1246 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1247 'sts': self._search_regex(
1248 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1250 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1251 video_info_webpage = self._download_webpage(
1252 video_info_url, video_id,
1253 note='Refetching age-gated info webpage',
1254 errnote='unable to download video info webpage')
1255 video_info = compat_parse_qs(video_info_webpage)
1256 add_dash_mpd(video_info)
1260 # Try looking directly into the video webpage
1261 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1263 args = ytplayer_config['args']
1264 if args.get('url_encoded_fmt_stream_map'):
1265 # Convert to the same format returned by compat_parse_qs
1266 video_info = dict((k, [v]) for k, v in args.items())
1267 add_dash_mpd(video_info)
1268 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1270 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1271 # We also try looking in get_video_info since it may contain different dashmpd
1272 # URL that points to a DASH manifest with possibly different itag set (some itags
1273 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1274 # manifest pointed by get_video_info's dashmpd).
1275 # The general idea is to take a union of itags of both DASH manifests (for example
1276 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1277 self.report_video_info_webpage_download(video_id)
1278 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1280 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1281 % (proto, video_id, el_type))
1282 video_info_webpage = self._download_webpage(
1284 video_id, note=False,
1285 errnote='unable to download video info webpage')
1286 get_video_info = compat_parse_qs(video_info_webpage)
1287 if get_video_info.get('use_cipher_signature') != ['True']:
1288 add_dash_mpd(get_video_info)
1290 video_info = get_video_info
1291 if 'token' in get_video_info:
1292 # Different get_video_info requests may report different results, e.g.
1293 # some may report video unavailability, but some may serve it without
1294 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1295 # the original webpage as well as el=info and el=embedded get_video_info
1296 # requests report video unavailability due to geo restriction while
1297 # el=detailpage succeeds and returns valid data). This is probably
1298 # due to YouTube measures against IP ranges of hosting providers.
1299 # Working around by preferring the first succeeded video_info containing
1300 # the token if no such video_info yet was found.
1301 if 'token' not in video_info:
1302 video_info = get_video_info
1304 if 'token' not in video_info:
1305 if 'reason' in video_info:
1306 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1307 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1309 raise ExtractorError('YouTube said: This video is available in %s only' % (
1310 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1312 raise ExtractorError(
1313 'YouTube said: %s' % video_info['reason'][0],
1314 expected=True, video_id=video_id)
1316 raise ExtractorError(
1317 '"token" parameter not in video info for unknown reason',
1321 if 'title' in video_info:
1322 video_title = video_info['title'][0]
1324 self._downloader.report_warning('Unable to extract video title')
1328 video_description = get_element_by_id("eow-description", video_webpage)
1329 if video_description:
1330 video_description = re.sub(r'''(?x)
1332 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1333 (?:title|href)="([^"]+)"\s+
1334 (?:[a-zA-Z-]+="[^"]*"\s+)*?
1338 ''', r'\1', video_description)
1339 video_description = clean_html(video_description)
1341 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1343 video_description = unescapeHTML(fd_mobj.group(1))
1345 video_description = ''
1347 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1348 if not self._downloader.params.get('noplaylist'):
1351 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1352 for feed in multifeed_metadata_list.split(','):
1353 # Unquote should take place before split on comma (,) since textual
1354 # fields may contain comma as well (see
1355 # https://github.com/rg3/youtube-dl/issues/8536)
1356 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1358 '_type': 'url_transparent',
1359 'ie_key': 'Youtube',
1361 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1362 {'force_singlefeed': True}),
1363 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1365 feed_ids.append(feed_data['id'][0])
1367 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1368 % (', '.join(feed_ids), video_id))
1369 return self.playlist_result(entries, video_id, video_title, video_description)
1370 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1372 if 'view_count' in video_info:
1373 view_count = int(video_info['view_count'][0])
1377 # Check for "rental" videos
1378 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1379 raise ExtractorError('"rental" videos not supported')
1381 # Start extracting information
1382 self.report_information_extraction(video_id)
1385 if 'author' not in video_info:
1386 raise ExtractorError('Unable to extract uploader name')
1387 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1390 video_uploader_id = None
1391 video_uploader_url = None
1393 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1395 if mobj is not None:
1396 video_uploader_id = mobj.group('uploader_id')
1397 video_uploader_url = mobj.group('uploader_url')
1399 self._downloader.report_warning('unable to extract uploader nickname')
1402 # We try first to get a high quality image:
1403 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1404 video_webpage, re.DOTALL)
1405 if m_thumb is not None:
1406 video_thumbnail = m_thumb.group(1)
1407 elif 'thumbnail_url' not in video_info:
1408 self._downloader.report_warning('unable to extract video thumbnail')
1409 video_thumbnail = None
1410 else: # don't panic if we can't find it
1411 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1414 upload_date = self._html_search_meta(
1415 'datePublished', video_webpage, 'upload date', default=None)
1417 upload_date = self._search_regex(
1418 [r'(?s)id="eow-date.*?>(.*?)</span>',
1419 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1420 video_webpage, 'upload date', default=None)
1422 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1423 upload_date = unified_strdate(upload_date)
1425 video_license = self._html_search_regex(
1426 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1427 video_webpage, 'license', default=None)
1429 m_music = re.search(
1430 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1433 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1434 video_creator = clean_html(m_music.group('creator'))
1436 video_alt_title = video_creator = None
1438 m_cat_container = self._search_regex(
1439 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1440 video_webpage, 'categories', default=None)
1442 category = self._html_search_regex(
1443 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1445 video_categories = None if category is None else [category]
1447 video_categories = None
1450 unescapeHTML(m.group('content'))
1451 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1453 def _extract_count(count_name):
1454 return str_to_int(self._search_regex(
1455 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1456 % re.escape(count_name),
1457 video_webpage, count_name, default=None))
1459 like_count = _extract_count('like')
1460 dislike_count = _extract_count('dislike')
1463 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1464 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1466 if 'length_seconds' not in video_info:
1467 self._downloader.report_warning('unable to extract video duration')
1468 video_duration = None
1470 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1473 video_annotations = None
1474 if self._downloader.params.get('writeannotations', False):
1475 video_annotations = self._extract_annotations(video_id)
1477 def _map_to_format_list(urlmap):
1479 for itag, video_real_url in urlmap.items():
1482 'url': video_real_url,
1483 'player_url': player_url,
1485 if itag in self._formats:
1486 dct.update(self._formats[itag])
1490 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1491 self.report_rtmp_download()
1493 'format_id': '_rtmp',
1495 'url': video_info['conn'][0],
1496 'player_url': player_url,
1498 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1499 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1500 if 'rtmpe%3Dyes' in encoded_url_map:
1501 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1503 fmt_list = video_info.get('fmt_list', [''])[0]
1505 for fmt in fmt_list.split(','):
1506 spec = fmt.split('/')
1508 width_height = spec[1].split('x')
1509 if len(width_height) == 2:
1510 formats_spec[spec[0]] = {
1511 'resolution': spec[1],
1512 'width': int_or_none(width_height[0]),
1513 'height': int_or_none(width_height[1]),
1516 for url_data_str in encoded_url_map.split(','):
1517 url_data = compat_parse_qs(url_data_str)
1518 if 'itag' not in url_data or 'url' not in url_data:
1520 format_id = url_data['itag'][0]
1521 url = url_data['url'][0]
1523 if 'sig' in url_data:
1524 url += '&signature=' + url_data['sig'][0]
1525 elif 's' in url_data:
1526 encrypted_sig = url_data['s'][0]
1527 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1529 jsplayer_url_json = self._search_regex(
1531 embed_webpage if age_gate else video_webpage,
1532 'JS player URL (1)', default=None)
1533 if not jsplayer_url_json and not age_gate:
1534 # We need the embed website after all
1535 if embed_webpage is None:
1536 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1537 embed_webpage = self._download_webpage(
1538 embed_url, video_id, 'Downloading embed webpage')
1539 jsplayer_url_json = self._search_regex(
1540 ASSETS_RE, embed_webpage, 'JS player URL')
1542 player_url = json.loads(jsplayer_url_json)
1543 if player_url is None:
1544 player_url_json = self._search_regex(
1545 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1546 video_webpage, 'age gate player URL')
1547 player_url = json.loads(player_url_json)
1549 if self._downloader.params.get('verbose'):
1550 if player_url is None:
1551 player_version = 'unknown'
1552 player_desc = 'unknown'
1554 if player_url.endswith('swf'):
1555 player_version = self._search_regex(
1556 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1557 'flash player', fatal=False)
1558 player_desc = 'flash player %s' % player_version
1560 player_version = self._search_regex(
1561 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1563 'html5 player', fatal=False)
1564 player_desc = 'html5 player %s' % player_version
1566 parts_sizes = self._signature_cache_id(encrypted_sig)
1567 self.to_screen('{%s} signature length %s, %s' %
1568 (format_id, parts_sizes, player_desc))
1570 signature = self._decrypt_signature(
1571 encrypted_sig, video_id, player_url, age_gate)
1572 url += '&signature=' + signature
1573 if 'ratebypass' not in url:
1574 url += '&ratebypass=yes'
1577 'format_id': format_id,
1579 'player_url': player_url,
1581 if format_id in self._formats:
1582 dct.update(self._formats[format_id])
1583 if format_id in formats_spec:
1584 dct.update(formats_spec[format_id])
1586 # Some itags are not included in DASH manifest thus corresponding formats will
1587 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1588 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1589 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1590 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1593 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1594 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1597 'fps': int_or_none(url_data.get('fps', [None])[0]),
1598 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1600 for key, value in more_fields.items():
1603 type_ = url_data.get('type', [None])[0]
1605 type_split = type_.split(';')
1606 kind_ext = type_split[0].split('/')
1607 if len(kind_ext) == 2:
1609 dct['ext'] = mimetype2ext(type_split[0])
1610 if kind in ('audio', 'video'):
1612 for mobj in re.finditer(
1613 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1614 if mobj.group('key') == 'codecs':
1615 codecs = mobj.group('val')
1618 codecs = codecs.split(',')
1619 if len(codecs) == 2:
1620 acodec, vcodec = codecs[1], codecs[0]
1622 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1628 elif video_info.get('hlsvp'):
1629 manifest_url = video_info['hlsvp'][0]
1630 url_map = self._extract_from_m3u8(manifest_url, video_id)
1631 formats = _map_to_format_list(url_map)
1632 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1633 for a_format in formats:
1634 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1636 unavailable_message = self._html_search_regex(
1637 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1638 video_webpage, 'unavailable message', default=None)
1639 if unavailable_message:
1640 raise ExtractorError(unavailable_message, expected=True)
1641 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1643 # Look for the DASH manifest
1644 if self._downloader.params.get('youtube_include_dash_manifest', True):
1645 dash_mpd_fatal = True
1646 for mpd_url in dash_mpds:
1649 def decrypt_sig(mobj):
1651 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1652 return '/signature/%s' % dec_s
1654 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1656 for df in self._extract_mpd_formats(
1657 mpd_url, video_id, fatal=dash_mpd_fatal,
1658 formats_dict=self._formats):
1659 # Do not overwrite DASH format found in some previous DASH manifest
1660 if df['format_id'] not in dash_formats:
1661 dash_formats[df['format_id']] = df
1662 # Additional DASH manifests may end up in HTTP Error 403 therefore
1663 # allow them to fail without bug report message if we already have
1664 # some DASH manifest succeeded. This is temporary workaround to reduce
1665 # burst of bug reports until we figure out the reason and whether it
1666 # can be fixed at all.
1667 dash_mpd_fatal = False
1668 except (ExtractorError, KeyError) as e:
1669 self.report_warning(
1670 'Skipping DASH manifest: %r' % e, video_id)
1672 # Remove the formats we found through non-DASH, they
1673 # contain less info and it can be wrong, because we use
1674 # fixed values (for example the resolution). See
1675 # https://github.com/rg3/youtube-dl/issues/5774 for an
1677 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1678 formats.extend(dash_formats.values())
1680 # Check for malformed aspect ratio
1681 stretched_m = re.search(
1682 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1685 w = float(stretched_m.group('w'))
1686 h = float(stretched_m.group('h'))
1687 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1688 # We will only process correct ratios.
1692 if f.get('vcodec') != 'none':
1693 f['stretched_ratio'] = ratio
1695 self._sort_formats(formats)
1697 self.mark_watched(video_id, video_info)
1701 'uploader': video_uploader,
1702 'uploader_id': video_uploader_id,
1703 'uploader_url': video_uploader_url,
1704 'upload_date': upload_date,
1705 'license': video_license,
1706 'creator': video_creator,
1707 'title': video_title,
1708 'alt_title': video_alt_title,
1709 'thumbnail': video_thumbnail,
1710 'description': video_description,
1711 'categories': video_categories,
1713 'subtitles': video_subtitles,
1714 'automatic_captions': automatic_captions,
1715 'duration': video_duration,
1716 'age_limit': 18 if age_gate else 0,
1717 'annotations': video_annotations,
1718 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1719 'view_count': view_count,
1720 'like_count': like_count,
1721 'dislike_count': dislike_count,
1722 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1725 'start_time': start_time,
1726 'end_time': end_time,
1730 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1731 IE_DESC = 'YouTube.com playlists'
1732 _VALID_URL = r"""(?x)(?:
1737 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1738 \? (?:.*?[&;])*? (?:p|a|list)=
1742 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1743 # Top tracks, they can also include dots
1748 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1750 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1751 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1752 IE_NAME = 'youtube:playlist'
1754 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1756 'title': 'ytdl test PL',
1757 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1759 'playlist_count': 3,
1761 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1763 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1764 'title': 'YDL_Empty_List',
1766 'playlist_count': 0,
1768 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1769 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1771 'title': '29C3: Not my department',
1772 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1774 'playlist_count': 95,
1776 'note': 'issue #673',
1777 'url': 'PLBB231211A4F62143',
1779 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1780 'id': 'PLBB231211A4F62143',
1782 'playlist_mincount': 26,
1784 'note': 'Large playlist',
1785 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1787 'title': 'Uploads from Cauchemar',
1788 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1790 'playlist_mincount': 799,
1792 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1794 'title': 'YDL_safe_search',
1795 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1797 'playlist_count': 2,
1800 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1801 'playlist_count': 4,
1804 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1807 'note': 'Embedded SWF player',
1808 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1809 'playlist_count': 4,
1812 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1815 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1816 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1818 'title': 'Uploads from Interstellar Movie',
1819 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1821 'playlist_mincout': 21,
1824 def _real_initialize(self):
1827 def _extract_mix(self, playlist_id):
1828 # The mixes are generated from a single video
1829 # the id of the playlist is just 'RD' + video_id
1831 last_id = playlist_id[-11:]
1832 for n in itertools.count(1):
1833 url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
1834 webpage = self._download_webpage(
1835 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
1836 new_ids = orderedSet(re.findall(
1837 r'''(?xs)data-video-username=".*?".*?
1838 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1840 # Fetch new pages until all the videos are repeated, it seems that
1841 # there are always 51 unique videos.
1842 new_ids = [_id for _id in new_ids if _id not in ids]
1848 url_results = self._ids_to_results(ids)
1850 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1852 search_title('playlist-title') or
1853 search_title('title long-title') or
1854 search_title('title'))
1855 title = clean_html(title_span)
1857 return self.playlist_result(url_results, playlist_id, title)
1859 def _extract_playlist(self, playlist_id):
1860 url = self._TEMPLATE_URL % playlist_id
1861 page = self._download_webpage(url, playlist_id)
1863 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1864 match = match.strip()
1865 # Check if the playlist exists or is private
1866 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1867 raise ExtractorError(
1868 'The playlist doesn\'t exist or is private, use --username or '
1869 '--netrc to access it.',
1871 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1872 raise ExtractorError(
1873 'Invalid parameters. Maybe URL is incorrect.',
1875 elif re.match(r'[^<]*Choose your language[^<]*', match):
1878 self.report_warning('Youtube gives an alert message: ' + match)
1880 playlist_title = self._html_search_regex(
1881 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1884 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1886 def _check_download_just_video(self, url, playlist_id):
1887 # Check if it's a video-specific URL
1888 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1889 if 'v' in query_dict:
1890 video_id = query_dict['v'][0]
1891 if self._downloader.params.get('noplaylist'):
1892 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1893 return self.url_result(video_id, 'Youtube', video_id=video_id)
1895 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1897 def _real_extract(self, url):
1898 # Extract playlist id
1899 mobj = re.match(self._VALID_URL, url)
1901 raise ExtractorError('Invalid URL: %s' % url)
1902 playlist_id = mobj.group(1) or mobj.group(2)
1904 video = self._check_download_just_video(url, playlist_id)
1908 if playlist_id.startswith(('RD', 'UL', 'PU')):
1909 # Mixes require a custom extraction process
1910 return self._extract_mix(playlist_id)
1912 return self._extract_playlist(playlist_id)
1915 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1916 IE_DESC = 'YouTube.com channels'
1917 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1918 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1919 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1920 IE_NAME = 'youtube:channel'
1922 'note': 'paginated channel',
1923 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1924 'playlist_mincount': 91,
1926 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1927 'title': 'Uploads from lex will',
1930 'note': 'Age restricted channel',
1931 # from https://www.youtube.com/user/DeusExOfficial
1932 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1933 'playlist_mincount': 64,
1935 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1936 'title': 'Uploads from Deus Ex',
1941 def suitable(cls, url):
1942 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1943 else super(YoutubeChannelIE, cls).suitable(url))
1945 def _real_extract(self, url):
1946 channel_id = self._match_id(url)
1948 url = self._TEMPLATE_URL % channel_id
1950 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1951 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1952 # otherwise fallback on channel by page extraction
1953 channel_page = self._download_webpage(
1954 url + '?view=57', channel_id,
1955 'Downloading channel page', fatal=False)
1956 if channel_page is False:
1957 channel_playlist_id = False
1959 channel_playlist_id = self._html_search_meta(
1960 'channelId', channel_page, 'channel id', default=None)
1961 if not channel_playlist_id:
1962 channel_playlist_id = self._search_regex(
1963 r'data-(?:channel-external-|yt)id="([^"]+)"',
1964 channel_page, 'channel id', default=None)
1965 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1966 playlist_id = 'UU' + channel_playlist_id[2:]
1967 return self.url_result(
1968 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1970 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1971 autogenerated = re.search(r'''(?x)
1973 channel-header-autogenerated-label|
1974 yt-channel-title-autogenerated
1975 )[^"]*"''', channel_page) is not None
1978 # The videos are contained in a single page
1979 # the ajax pages can't be used, they are empty
1982 video_id, 'Youtube', video_id=video_id,
1983 video_title=video_title)
1984 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1985 return self.playlist_result(entries, channel_id)
1987 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1990 class YoutubeUserIE(YoutubeChannelIE):
1991 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1992 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/|c/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1993 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1994 IE_NAME = 'youtube:user'
1997 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1998 'playlist_mincount': 320,
2000 'title': 'TheLinuxFoundation',
2003 'url': 'ytuser:phihag',
2004 'only_matching': True,
2006 'url': 'https://www.youtube.com/c/gametrailers',
2007 'only_matching': True,
2011 def suitable(cls, url):
2012 # Don't return True if the url can be extracted with other youtube
2013 # extractor, the regex would is too permissive and it would match.
2014 other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2015 if any(ie.suitable(url) for ie in other_yt_ies):
2018 return super(YoutubeUserIE, cls).suitable(url)
2021 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2022 IE_DESC = 'YouTube.com live streams'
2023 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2024 IE_NAME = 'youtube:live'
2027 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2029 'id': 'a48o2S1cPoo',
2031 'title': 'The Young Turks - Live Main Show',
2032 'uploader': 'The Young Turks',
2033 'uploader_id': 'TheYoungTurks',
2034 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2035 'upload_date': '20150715',
2036 'license': 'Standard YouTube License',
2037 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2038 'categories': ['News & Politics'],
2039 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2041 'dislike_count': int,
2044 'skip_download': True,
2047 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2048 'only_matching': True,
2051 def _real_extract(self, url):
2052 mobj = re.match(self._VALID_URL, url)
2053 channel_id = mobj.group('id')
2054 base_url = mobj.group('base_url')
2055 webpage = self._download_webpage(url, channel_id, fatal=False)
2057 page_type = self._og_search_property(
2058 'type', webpage, 'page type', default=None)
2059 video_id = self._html_search_meta(
2060 'videoId', webpage, 'video id', default=None)
2061 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2062 return self.url_result(video_id, YoutubeIE.ie_key())
2063 return self.url_result(base_url)
2066 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2067 IE_DESC = 'YouTube.com user/channel playlists'
2068 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2069 IE_NAME = 'youtube:playlists'
2072 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2073 'playlist_mincount': 4,
2075 'id': 'ThirstForScience',
2076 'title': 'Thirst for Science',
2079 # with "Load more" button
2080 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2081 'playlist_mincount': 70,
2084 'title': 'Игорь Клейнер',
2087 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2088 'playlist_mincount': 17,
2090 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2091 'title': 'Chem Player',
2096 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2097 IE_DESC = 'YouTube.com searches'
2098 # there doesn't appear to be a real limit, for example if you search for
2099 # 'python' you get more than 8.000.000 results
2100 _MAX_RESULTS = float('inf')
2101 IE_NAME = 'youtube:search'
2102 _SEARCH_KEY = 'ytsearch'
2103 _EXTRA_QUERY_ARGS = {}
2106 def _get_n_results(self, query, n):
2107 """Get a specified number of results for a query"""
2112 for pagenum in itertools.count(1):
2114 'search_query': query.encode('utf-8'),
2118 url_query.update(self._EXTRA_QUERY_ARGS)
2119 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2120 data = self._download_json(
2121 result_url, video_id='query "%s"' % query,
2122 note='Downloading page %s' % pagenum,
2123 errnote='Unable to download API page')
2124 html_content = data[1]['body']['content']
2126 if 'class="search-message' in html_content:
2127 raise ExtractorError(
2128 '[youtube] No video results', expected=True)
2130 new_videos = self._ids_to_results(orderedSet(re.findall(
2131 r'href="/watch\?v=(.{11})', html_content)))
2132 videos += new_videos
2133 if not new_videos or len(videos) > limit:
2138 return self.playlist_result(videos, query)
2141 class YoutubeSearchDateIE(YoutubeSearchIE):
2142 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2143 _SEARCH_KEY = 'ytsearchdate'
2144 IE_DESC = 'YouTube.com searches, newest videos first'
2145 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2148 class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
2149 IE_DESC = 'YouTube.com search URLs'
2150 IE_NAME = 'youtube:search_url'
2151 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2152 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2154 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2155 'playlist_mincount': 5,
2157 'title': 'youtube-dl test video',
2160 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2161 'only_matching': True,
2164 def _real_extract(self, url):
2165 mobj = re.match(self._VALID_URL, url)
2166 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2167 webpage = self._download_webpage(url, query)
2168 return self.playlist_result(self._process_page(webpage), playlist_title=query)
2171 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2172 IE_DESC = 'YouTube.com (multi-season) shows'
2173 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2174 IE_NAME = 'youtube:show'
2176 'url': 'https://www.youtube.com/show/airdisasters',
2177 'playlist_mincount': 5,
2179 'id': 'airdisasters',
2180 'title': 'Air Disasters',
2184 def _real_extract(self, url):
2185 playlist_id = self._match_id(url)
2186 return super(YoutubeShowIE, self)._real_extract(
2187 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2190 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2192 Base class for feed extractors
2193 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2195 _LOGIN_REQUIRED = True
2199 return 'youtube:%s' % self._FEED_NAME
2201 def _real_initialize(self):
2204 def _real_extract(self, url):
2205 page = self._download_webpage(
2206 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2208 # The extraction process is the same as for playlists, but the regex
2209 # for the video ids doesn't contain an index
2211 more_widget_html = content_html = page
2212 for page_num in itertools.count(1):
2213 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2215 # 'recommended' feed has infinite 'load more' and each new portion spins
2216 # the same videos in (sometimes) slightly different order, so we'll check
2217 # for unicity and break when portion has no new videos
2218 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2224 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2228 more = self._download_json(
2229 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2230 'Downloading page #%s' % page_num,
2231 transform_source=uppercase_escape)
2232 content_html = more['content_html']
2233 more_widget_html = more['load_more_widget_html']
2235 return self.playlist_result(
2236 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2239 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2240 IE_NAME = 'youtube:watchlater'
2241 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2242 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2245 'url': 'https://www.youtube.com/playlist?list=WL',
2246 'only_matching': True,
2248 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2249 'only_matching': True,
2252 def _real_extract(self, url):
2253 video = self._check_download_just_video(url, 'WL')
2256 return self._extract_playlist('WL')
2259 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2260 IE_NAME = 'youtube:favorites'
2261 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2262 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2263 _LOGIN_REQUIRED = True
2265 def _real_extract(self, url):
2266 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2267 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2268 return self.url_result(playlist_id, 'YoutubePlaylist')
2271 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2272 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2273 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2274 _FEED_NAME = 'recommended'
2275 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2278 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2279 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2280 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2281 _FEED_NAME = 'subscriptions'
2282 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2285 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2286 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2287 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2288 _FEED_NAME = 'history'
2289 _PLAYLIST_TITLE = 'Youtube History'
2292 class YoutubeTruncatedURLIE(InfoExtractor):
2293 IE_NAME = 'youtube:truncated_url'
2294 IE_DESC = False # Do not list
2295 _VALID_URL = r'''(?x)
2297 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2300 annotation_id=annotation_[^&]+|
2306 attribution_link\?a=[^&]+
2312 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2313 'only_matching': True,
2315 'url': 'http://www.youtube.com/watch?',
2316 'only_matching': True,
2318 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2319 'only_matching': True,
2321 'url': 'https://www.youtube.com/watch?feature=foo',
2322 'only_matching': True,
2324 'url': 'https://www.youtube.com/watch?hl=en-GB',
2325 'only_matching': True,
2327 'url': 'https://www.youtube.com/watch?t=2372',
2328 'only_matching': True,
2331 def _real_extract(self, url):
2332 raise ExtractorError(
2333 'Did you forget to quote the URL? Remember that & is a meta '
2334 'character in most shells, so you want to put the URL in quotes, '
2336 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2337 ' or simply youtube-dl BaW_jenozKc .',
2341 class YoutubeTruncatedIDIE(InfoExtractor):
2342 IE_NAME = 'youtube:truncated_id'
2343 IE_DESC = False # Do not list
2344 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2347 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2348 'only_matching': True,
2351 def _real_extract(self, url):
2352 video_id = self._match_id(url)
2353 raise ExtractorError(
2354 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),