3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
32 get_element_by_attribute,
51 class YoutubeBaseInfoExtractor(InfoExtractor):
52 """Provide base functions for Youtube extractors"""
53 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
54 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
55 _NETRC_MACHINE = 'youtube'
56 # If True it will raise an error if no login info is provided
57 _LOGIN_REQUIRED = False
59 def _set_language(self):
61 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
62 # YouTube sets the expire time to about two months
63 expire_time=time.time() + 2 * 30 * 24 * 3600)
65 def _ids_to_results(self, ids):
67 self.url_result(vid_id, 'Youtube', video_id=vid_id)
72 Attempt to log in to YouTube.
73 True is returned if successful or skipped.
74 False is returned if login failed.
76 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
78 (username, password) = self._get_login_info()
79 # No authentication to be performed
81 if self._LOGIN_REQUIRED:
82 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
85 login_page = self._download_webpage(
86 self._LOGIN_URL, None,
87 note='Downloading login page',
88 errnote='unable to fetch login page', fatal=False)
89 if login_page is False:
92 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
93 login_page, 'Login GALX parameter')
97 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
102 'PersistentCookie': 'yes',
104 'bgresponse': 'js_disabled',
105 'checkConnection': '',
106 'checkedDomains': 'youtube',
113 'service': 'youtube',
118 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
120 req = sanitized_Request(self._LOGIN_URL, login_data)
121 login_results = self._download_webpage(
123 note='Logging in', errnote='unable to log in', fatal=False)
124 if login_results is False:
127 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
128 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
131 # TODO add SMS and phone call support - these require making a request and then prompting the user
133 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
134 tfa_code = self._get_tfa_info('2-step verification code')
137 self._downloader.report_warning(
138 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
139 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
142 tfa_code = remove_start(tfa_code, 'G-')
144 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
146 tfa_form_strs.update({
151 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
153 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
154 tfa_results = self._download_webpage(
156 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
158 if tfa_results is False:
161 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
162 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
164 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
165 self._downloader.report_warning('unable to log in - did the page structure change?')
167 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
168 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
171 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
172 self._downloader.report_warning('unable to log in: bad username or password')
176 def _real_initialize(self):
177 if self._downloader is None:
180 if not self._login():
184 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
185 # Extract entries from page with "Load more" button
186 def _entries(self, page, playlist_id):
187 more_widget_html = content_html = page
188 for page_num in itertools.count(1):
189 for entry in self._process_page(content_html):
192 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
196 more = self._download_json(
197 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
198 'Downloading page #%s' % page_num,
199 transform_source=uppercase_escape)
200 content_html = more['content_html']
201 if not content_html.strip():
202 # Some webpages show a "Load more" button but they don't
205 more_widget_html = more['load_more_widget_html']
208 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
209 def _process_page(self, content):
210 for video_id, video_title in self.extract_videos_from_page(content):
211 yield self.url_result(video_id, 'Youtube', video_id, video_title)
213 def extract_videos_from_page(self, page):
216 for mobj in re.finditer(self._VIDEO_RE, page):
217 # The link with index 0 is not the first video of the playlist (not sure if still actual)
218 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
220 video_id = mobj.group('id')
221 video_title = unescapeHTML(mobj.group('title'))
223 video_title = video_title.strip()
225 idx = ids_in_page.index(video_id)
226 if video_title and not titles_in_page[idx]:
227 titles_in_page[idx] = video_title
229 ids_in_page.append(video_id)
230 titles_in_page.append(video_title)
231 return zip(ids_in_page, titles_in_page)
234 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
235 def _process_page(self, content):
236 for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
237 yield self.url_result(
238 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
240 def _real_extract(self, url):
241 playlist_id = self._match_id(url)
242 webpage = self._download_webpage(url, playlist_id)
243 title = self._og_search_title(webpage, fatal=False)
244 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
247 class YoutubeIE(YoutubeBaseInfoExtractor):
248 IE_DESC = 'YouTube.com'
249 _VALID_URL = r"""(?x)^
251 (?:https?://|//) # http(s):// or protocol-independent URL
252 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
253 (?:www\.)?deturl\.com/www\.youtube\.com/|
254 (?:www\.)?pwnyoutube\.com/|
255 (?:www\.)?yourepeat\.com/|
256 tube\.majestyc\.net/|
257 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
258 (?:.*?\#/)? # handle anchor (#/) redirect urls
259 (?: # the various things that can precede the ID:
260 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
261 |(?: # or the v= param in all its forms
262 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
263 (?:\?|\#!?) # the params delimiter ? or # or #!
264 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
269 youtu\.be| # just youtu.be/xxxx
270 vid\.plus # or vid.plus/xxxx
272 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
274 )? # all until now is optional -> you can pass the naked ID
275 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
276 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
277 (?(1).+)? # if we found the ID, everything can follow
279 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
281 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
282 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
283 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
284 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
285 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
286 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
287 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
288 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '36': {'ext': '3gp', 'width': 320, 'height': 240, 'acodec': 'aac', 'abr': 32, 'vcodec': 'mp4v'},
290 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
291 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
292 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
293 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
294 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
295 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
296 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
297 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
301 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
302 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
303 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
304 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
305 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
306 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
307 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
309 # Apple HTTP Live Streaming
310 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
311 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
312 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
313 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
314 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
315 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
316 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
319 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
320 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
321 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
322 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
323 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
324 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
325 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
326 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
327 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
328 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
329 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'h264', 'preference': -40},
332 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
333 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
334 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
337 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
339 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
344 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
345 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
346 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
347 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
348 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
349 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
350 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
351 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
352 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
353 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
354 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
355 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
356 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
357 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'preference': -40},
358 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
361 '171': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
362 '172': {'ext': 'webm', 'acodec': 'vorbis', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
364 # Dash webm audio with opus inside
365 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
366 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
367 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
370 '_rtmp': {'protocol': 'rtmp'},
376 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
380 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
381 'uploader': 'Philipp Hagemeister',
382 'uploader_id': 'phihag',
383 'upload_date': '20121002',
384 '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 .',
385 'categories': ['Science & Technology'],
386 'tags': ['youtube-dl'],
388 'dislike_count': int,
394 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
395 'note': 'Test generic use_cipher_signature video (#897)',
399 'upload_date': '20120506',
400 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
401 'alt_title': 'I Love It (feat. Charli XCX)',
402 'description': 'md5:782e8651347686cba06e58f71ab51773',
403 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
404 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
405 'iconic ep', 'iconic', 'love', 'it'],
406 'uploader': 'Icona Pop',
407 'uploader_id': 'IconaPop',
408 'creator': 'Icona Pop',
412 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
413 'note': 'Test VEVO video with age protection (#956)',
417 'upload_date': '20130703',
418 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
419 'alt_title': 'Tunnel Vision',
420 'description': 'md5:64249768eec3bc4276236606ea996373',
421 'uploader': 'justintimberlakeVEVO',
422 'uploader_id': 'justintimberlakeVEVO',
423 'creator': 'Justin Timberlake',
428 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
429 'note': 'Embed-only video (#1746)',
433 'upload_date': '20120608',
434 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
435 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
436 'uploader': 'SET India',
437 'uploader_id': 'setindia',
442 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
443 'note': 'Use the first video ID in the URL',
447 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
448 'uploader': 'Philipp Hagemeister',
449 'uploader_id': 'phihag',
450 'upload_date': '20121002',
451 '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 .',
452 'categories': ['Science & Technology'],
453 'tags': ['youtube-dl'],
455 'dislike_count': int,
458 'skip_download': True,
462 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
463 'note': '256k DASH audio (format 141) via DASH manifest',
467 'upload_date': '20121002',
468 'uploader_id': '8KVIDEO',
470 'uploader': '8KVIDEO',
471 'title': 'UHDTV TEST 8K VIDEO.mp4'
474 'youtube_include_dash_manifest': True,
478 # DASH manifest with encrypted signature
480 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
484 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
485 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
486 'uploader': 'AfrojackVEVO',
487 'uploader_id': 'AfrojackVEVO',
488 'upload_date': '20131011',
491 'youtube_include_dash_manifest': True,
495 # JS player signature function name containing $
497 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
501 'title': 'Taylor Swift - Shake It Off',
502 'alt_title': 'Shake It Off',
503 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
504 'uploader': 'TaylorSwiftVEVO',
505 'uploader_id': 'TaylorSwiftVEVO',
506 'upload_date': '20140818',
507 'creator': 'Taylor Swift',
510 'youtube_include_dash_manifest': True,
516 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
520 'upload_date': '20100909',
521 'uploader': 'The Amazing Atheist',
522 'uploader_id': 'TheAmazingAtheist',
523 'title': 'Burning Everyone\'s Koran',
524 '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',
527 # Normal age-gate video (No vevo, embed allowed)
529 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
533 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
534 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
535 'uploader': 'The Witcher',
536 'uploader_id': 'WitcherGame',
537 'upload_date': '20140605',
541 # Age-gate video with encrypted signature
543 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
547 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
548 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
549 'uploader': 'LloydVEVO',
550 'uploader_id': 'LloydVEVO',
551 'upload_date': '20110629',
555 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
557 'url': '__2ABJjxzNo',
561 'upload_date': '20100430',
562 'uploader_id': 'deadmau5',
563 'creator': 'deadmau5',
564 'description': 'md5:12c56784b8032162bb936a5f76d55360',
565 'uploader': 'deadmau5',
566 'title': 'Deadmau5 - Some Chords (HD)',
567 'alt_title': 'Some Chords',
569 'expected_warnings': [
570 'DASH manifest missing',
573 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
575 'url': 'lqQg6PlCWgI',
579 'upload_date': '20150827',
580 'uploader_id': 'olympic',
581 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
582 'uploader': 'Olympics',
583 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
586 'skip_download': 'requires avconv',
591 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
595 'stretched_ratio': 16 / 9.,
596 'upload_date': '20110310',
597 'uploader_id': 'AllenMeow',
598 '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',
600 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
603 # url_encoded_fmt_stream_map is empty string
605 'url': 'qEJwOuvDf7I',
609 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
611 'upload_date': '20150404',
612 'uploader_id': 'spbelect',
613 'uploader': 'Наблюдатели Петербурга',
616 'skip_download': 'requires avconv',
618 'skip': 'This live event has ended.',
620 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
622 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
626 'title': 'md5:7b81415841e02ecd4313668cde88737a',
627 'description': 'md5:116377fd2963b81ec4ce64b542173306',
628 'upload_date': '20150625',
629 'uploader_id': 'dorappi2000',
630 'uploader': 'dorappi2000',
631 'formats': 'mincount:33',
634 # DASH manifest with segment_list
636 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
637 'md5': '8ce563a1d667b599d21064e982ab9e31',
641 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
642 'uploader': 'Airtek',
643 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
644 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
645 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
648 'youtube_include_dash_manifest': True,
649 'format': '135', # bestvideo
653 # Multifeed videos (multiple cameras), URL is for Main Camera
654 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
657 'title': 'teamPGP: Rocket League Noob Stream',
658 'description': 'md5:dc7872fb300e143831327f1bae3af010',
664 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
665 'description': 'md5:dc7872fb300e143831327f1bae3af010',
666 'upload_date': '20150721',
667 'uploader': 'Beer Games Beer',
668 'uploader_id': 'beergamesbeer',
674 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
675 'description': 'md5:dc7872fb300e143831327f1bae3af010',
676 'upload_date': '20150721',
677 'uploader': 'Beer Games Beer',
678 'uploader_id': 'beergamesbeer',
684 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
685 'description': 'md5:dc7872fb300e143831327f1bae3af010',
686 'upload_date': '20150721',
687 'uploader': 'Beer Games Beer',
688 'uploader_id': 'beergamesbeer',
694 'title': 'teamPGP: Rocket League Noob Stream (zim)',
695 'description': 'md5:dc7872fb300e143831327f1bae3af010',
696 'upload_date': '20150721',
697 'uploader': 'Beer Games Beer',
698 'uploader_id': 'beergamesbeer',
702 'skip_download': True,
706 'url': 'http://vid.plus/FlRa-iH7PGw',
707 'only_matching': True,
710 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
711 # Also tests cut-off URL expansion in video description (see
712 # https://github.com/rg3/youtube-dl/issues/1892,
713 # https://github.com/rg3/youtube-dl/issues/8164)
714 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
718 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
719 'alt_title': 'Dark Walk',
720 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
721 'upload_date': '20151119',
722 'uploader_id': 'IronSoulElf',
723 'uploader': 'IronSoulElf',
724 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
727 'skip_download': True,
731 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
732 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
733 'only_matching': True,
736 # Video with yt:stretch=17:0
737 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
741 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
742 'description': 'md5:ee18a25c350637c8faff806845bddee9',
743 'upload_date': '20151107',
744 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
745 'uploader': 'CH GAMER DROID',
748 'skip_download': True,
752 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
753 'only_matching': True,
757 def __init__(self, *args, **kwargs):
758 super(YoutubeIE, self).__init__(*args, **kwargs)
759 self._player_cache = {}
761 def report_video_info_webpage_download(self, video_id):
762 """Report attempt to download video info webpage."""
763 self.to_screen('%s: Downloading video info webpage' % video_id)
765 def report_information_extraction(self, video_id):
766 """Report attempt to extract video information."""
767 self.to_screen('%s: Extracting video information' % video_id)
769 def report_unavailable_format(self, video_id, format):
770 """Report extracted video URL."""
771 self.to_screen('%s: Format %s not available' % (video_id, format))
773 def report_rtmp_download(self):
774 """Indicate the download will use the RTMP protocol."""
775 self.to_screen('RTMP download detected')
777 def _signature_cache_id(self, example_sig):
778 """ Return a string representation of a signature """
779 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
781 def _extract_signature_function(self, video_id, player_url, example_sig):
783 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
786 raise ExtractorError('Cannot identify player %r' % player_url)
787 player_type = id_m.group('ext')
788 player_id = id_m.group('id')
790 # Read from filesystem cache
791 func_id = '%s_%s_%s' % (
792 player_type, player_id, self._signature_cache_id(example_sig))
793 assert os.path.basename(func_id) == func_id
795 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
796 if cache_spec is not None:
797 return lambda s: ''.join(s[i] for i in cache_spec)
800 'Downloading player %s' % player_url
801 if self._downloader.params.get('verbose') else
802 'Downloading %s player %s' % (player_type, player_id)
804 if player_type == 'js':
805 code = self._download_webpage(
806 player_url, video_id,
808 errnote='Download of %s failed' % player_url)
809 res = self._parse_sig_js(code)
810 elif player_type == 'swf':
811 urlh = self._request_webpage(
812 player_url, video_id,
814 errnote='Download of %s failed' % player_url)
816 res = self._parse_sig_swf(code)
818 assert False, 'Invalid player type %r' % player_type
820 test_string = ''.join(map(compat_chr, range(len(example_sig))))
821 cache_res = res(test_string)
822 cache_spec = [ord(c) for c in cache_res]
824 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
827 def _print_sig_code(self, func, example_sig):
828 def gen_sig_code(idxs):
829 def _genslice(start, end, step):
830 starts = '' if start == 0 else str(start)
831 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
832 steps = '' if step == 1 else (':%d' % step)
833 return 's[%s%s%s]' % (starts, ends, steps)
836 # Quelch pyflakes warnings - start will be set when step is set
837 start = '(Never used)'
838 for i, prev in zip(idxs[1:], idxs[:-1]):
842 yield _genslice(start, prev, step)
845 if i - prev in [-1, 1]:
854 yield _genslice(start, i, step)
856 test_string = ''.join(map(compat_chr, range(len(example_sig))))
857 cache_res = func(test_string)
858 cache_spec = [ord(c) for c in cache_res]
859 expr_code = ' + '.join(gen_sig_code(cache_spec))
860 signature_id_tuple = '(%s)' % (
861 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
862 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
863 ' return %s\n') % (signature_id_tuple, expr_code)
864 self.to_screen('Extracted signature function:\n' + code)
866 def _parse_sig_js(self, jscode):
867 funcname = self._search_regex(
868 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
869 'Initial JS player signature function name')
871 jsi = JSInterpreter(jscode)
872 initial_function = jsi.extract_function(funcname)
873 return lambda s: initial_function([s])
875 def _parse_sig_swf(self, file_contents):
876 swfi = SWFInterpreter(file_contents)
877 TARGET_CLASSNAME = 'SignatureDecipher'
878 searched_class = swfi.extract_class(TARGET_CLASSNAME)
879 initial_function = swfi.extract_function(searched_class, 'decipher')
880 return lambda s: initial_function([s])
882 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
883 """Turn the encrypted s field into a working signature"""
885 if player_url is None:
886 raise ExtractorError('Cannot decrypt signature without player_url')
888 if player_url.startswith('//'):
889 player_url = 'https:' + player_url
891 player_id = (player_url, self._signature_cache_id(s))
892 if player_id not in self._player_cache:
893 func = self._extract_signature_function(
894 video_id, player_url, s
896 self._player_cache[player_id] = func
897 func = self._player_cache[player_id]
898 if self._downloader.params.get('youtube_print_sig_code'):
899 self._print_sig_code(func, s)
901 except Exception as e:
902 tb = traceback.format_exc()
903 raise ExtractorError(
904 'Signature extraction failed: ' + tb, cause=e)
906 def _get_subtitles(self, video_id, webpage):
908 subs_doc = self._download_xml(
909 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
910 video_id, note=False)
911 except ExtractorError as err:
912 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
916 for track in subs_doc.findall('track'):
917 lang = track.attrib['lang_code']
918 if lang in sub_lang_list:
921 for ext in ['sbv', 'vtt', 'srt']:
922 params = compat_urllib_parse.urlencode({
926 'name': track.attrib['name'].encode('utf-8'),
929 'url': 'https://www.youtube.com/api/timedtext?' + params,
932 sub_lang_list[lang] = sub_formats
933 if not sub_lang_list:
934 self._downloader.report_warning('video doesn\'t have subtitles')
938 def _get_ytplayer_config(self, video_id, webpage):
940 # User data may contain arbitrary character sequences that may affect
941 # JSON extraction with regex, e.g. when '};' is contained the second
942 # regex won't capture the whole JSON. Yet working around by trying more
943 # concrete regex first keeping in mind proper quoted string handling
944 # to be implemented in future that will replace this workaround (see
945 # https://github.com/rg3/youtube-dl/issues/7468,
946 # https://github.com/rg3/youtube-dl/pull/7599)
947 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
948 r';ytplayer\.config\s*=\s*({.+?});',
950 config = self._search_regex(
951 patterns, webpage, 'ytplayer.config', default=None)
953 return self._parse_json(
954 uppercase_escape(config), video_id, fatal=False)
956 def _get_automatic_captions(self, video_id, webpage):
957 """We need the webpage for getting the captions url, pass it as an
958 argument to speed up the process."""
959 self.to_screen('%s: Looking for automatic captions' % video_id)
960 player_config = self._get_ytplayer_config(video_id, webpage)
961 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
962 if not player_config:
963 self._downloader.report_warning(err_msg)
966 args = player_config['args']
967 caption_url = args['ttsurl']
969 self._downloader.report_warning(err_msg)
971 timestamp = args['timestamp']
972 # We get the available subtitles
973 list_params = compat_urllib_parse.urlencode({
978 list_url = caption_url + '&' + list_params
979 caption_list = self._download_xml(list_url, video_id)
980 original_lang_node = caption_list.find('track')
981 if original_lang_node is None:
982 self._downloader.report_warning('Video doesn\'t have automatic captions')
984 original_lang = original_lang_node.attrib['lang_code']
985 caption_kind = original_lang_node.attrib.get('kind', '')
988 for lang_node in caption_list.findall('target'):
989 sub_lang = lang_node.attrib['lang_code']
991 for ext in ['sbv', 'vtt', 'srt']:
992 params = compat_urllib_parse.urlencode({
993 'lang': original_lang,
997 'kind': caption_kind,
1000 'url': caption_url + '&' + params,
1003 sub_lang_list[sub_lang] = sub_formats
1004 return sub_lang_list
1005 # An extractor error can be raise by the download process if there are
1006 # no automatic captions but there are subtitles
1007 except (KeyError, ExtractorError):
1008 self._downloader.report_warning(err_msg)
1012 def extract_id(cls, url):
1013 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1015 raise ExtractorError('Invalid URL: %s' % url)
1016 video_id = mobj.group(2)
1019 def _extract_from_m3u8(self, manifest_url, video_id):
1022 def _get_urls(_manifest):
1023 lines = _manifest.split('\n')
1024 urls = filter(lambda l: l and not l.startswith('#'),
1027 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1028 formats_urls = _get_urls(manifest)
1029 for format_url in formats_urls:
1030 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1031 url_map[itag] = format_url
1034 def _extract_annotations(self, video_id):
1035 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1036 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1038 def _parse_dash_manifest(
1039 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
1040 def decrypt_sig(mobj):
1042 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1043 return '/signature/%s' % dec_s
1044 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
1045 dash_doc = self._download_xml(
1046 dash_manifest_url, video_id,
1047 note='Downloading DASH manifest',
1048 errnote='Could not download DASH manifest',
1051 if dash_doc is False:
1055 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1056 mime_type = a.attrib.get('mimeType')
1057 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1058 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1061 if mime_type == 'text/vtt':
1062 # TODO implement WebVTT downloading
1064 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
1065 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1066 format_id = r.attrib['id']
1067 video_url = url_el.text
1068 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1070 'format_id': format_id,
1072 'width': int_or_none(r.attrib.get('width')),
1073 'height': int_or_none(r.attrib.get('height')),
1074 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
1075 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
1076 'filesize': filesize,
1077 'fps': int_or_none(r.attrib.get('frameRate')),
1079 if segment_list is not None:
1081 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
1082 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1083 'protocol': 'http_dash_segments',
1086 existing_format = next(
1087 fo for fo in formats
1088 if fo['format_id'] == format_id)
1089 except StopIteration:
1090 full_info = self._formats.get(format_id, {}).copy()
1092 codecs = r.attrib.get('codecs')
1094 if full_info.get('acodec') == 'none':
1095 full_info['vcodec'] = codecs
1096 elif full_info.get('vcodec') == 'none':
1097 full_info['acodec'] = codecs
1098 formats.append(full_info)
1100 existing_format.update(f)
1102 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
1105 def _real_extract(self, url):
1106 url, smuggled_data = unsmuggle_url(url, {})
1109 'http' if self._downloader.params.get('prefer_insecure', False)
1114 parsed_url = compat_urllib_parse_urlparse(url)
1115 for component in [parsed_url.fragment, parsed_url.query]:
1116 query = compat_parse_qs(component)
1117 if start_time is None and 't' in query:
1118 start_time = parse_duration(query['t'][0])
1119 if start_time is None and 'start' in query:
1120 start_time = parse_duration(query['start'][0])
1121 if end_time is None and 'end' in query:
1122 end_time = parse_duration(query['end'][0])
1124 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1125 mobj = re.search(self._NEXT_URL_RE, url)
1127 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1128 video_id = self.extract_id(url)
1131 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1132 video_webpage = self._download_webpage(url, video_id)
1134 # Attempt to extract SWF player URL
1135 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1136 if mobj is not None:
1137 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1143 def add_dash_mpd(video_info):
1144 dash_mpd = video_info.get('dashmpd')
1145 if dash_mpd and dash_mpd[0] not in dash_mpds:
1146 dash_mpds.append(dash_mpd[0])
1149 embed_webpage = None
1151 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1153 # We simulate the access to the video from www.youtube.com/v/{video_id}
1154 # this can be viewed without login into Youtube
1155 url = proto + '://www.youtube.com/embed/%s' % video_id
1156 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1157 data = compat_urllib_parse.urlencode({
1158 'video_id': video_id,
1159 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1160 'sts': self._search_regex(
1161 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1163 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1164 video_info_webpage = self._download_webpage(
1165 video_info_url, video_id,
1166 note='Refetching age-gated info webpage',
1167 errnote='unable to download video info webpage')
1168 video_info = compat_parse_qs(video_info_webpage)
1169 add_dash_mpd(video_info)
1173 # Try looking directly into the video webpage
1174 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1176 args = ytplayer_config['args']
1177 if args.get('url_encoded_fmt_stream_map'):
1178 # Convert to the same format returned by compat_parse_qs
1179 video_info = dict((k, [v]) for k, v in args.items())
1180 add_dash_mpd(video_info)
1181 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1183 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1184 # We also try looking in get_video_info since it may contain different dashmpd
1185 # URL that points to a DASH manifest with possibly different itag set (some itags
1186 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1187 # manifest pointed by get_video_info's dashmpd).
1188 # The general idea is to take a union of itags of both DASH manifests (for example
1189 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1190 self.report_video_info_webpage_download(video_id)
1191 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1193 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1194 % (proto, video_id, el_type))
1195 video_info_webpage = self._download_webpage(
1197 video_id, note=False,
1198 errnote='unable to download video info webpage')
1199 get_video_info = compat_parse_qs(video_info_webpage)
1200 if get_video_info.get('use_cipher_signature') != ['True']:
1201 add_dash_mpd(get_video_info)
1203 video_info = get_video_info
1204 if 'token' in get_video_info:
1205 # Different get_video_info requests may report different results, e.g.
1206 # some may report video unavailability, but some may serve it without
1207 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1208 # the original webpage as well as el=info and el=embedded get_video_info
1209 # requests report video unavailability due to geo restriction while
1210 # el=detailpage succeeds and returns valid data). This is probably
1211 # due to YouTube measures against IP ranges of hosting providers.
1212 # Working around by preferring the first succeeded video_info containing
1213 # the token if no such video_info yet was found.
1214 if 'token' not in video_info:
1215 video_info = get_video_info
1217 if 'token' not in video_info:
1218 if 'reason' in video_info:
1219 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1220 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1222 raise ExtractorError('YouTube said: This video is available in %s only' % (
1223 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1225 raise ExtractorError(
1226 'YouTube said: %s' % video_info['reason'][0],
1227 expected=True, video_id=video_id)
1229 raise ExtractorError(
1230 '"token" parameter not in video info for unknown reason',
1234 if 'title' in video_info:
1235 video_title = video_info['title'][0]
1237 self._downloader.report_warning('Unable to extract video title')
1241 video_description = get_element_by_id("eow-description", video_webpage)
1242 if video_description:
1243 video_description = re.sub(r'''(?x)
1245 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1246 (?:title|href)="([^"]+)"\s+
1247 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1248 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1251 ''', r'\1', video_description)
1252 video_description = clean_html(video_description)
1254 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1256 video_description = unescapeHTML(fd_mobj.group(1))
1258 video_description = ''
1260 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1261 if not self._downloader.params.get('noplaylist'):
1264 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1265 for feed in multifeed_metadata_list.split(','):
1266 feed_data = compat_parse_qs(feed)
1268 '_type': 'url_transparent',
1269 'ie_key': 'Youtube',
1271 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1272 {'force_singlefeed': True}),
1273 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1275 feed_ids.append(feed_data['id'][0])
1277 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1278 % (', '.join(feed_ids), video_id))
1279 return self.playlist_result(entries, video_id, video_title, video_description)
1280 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1282 if 'view_count' in video_info:
1283 view_count = int(video_info['view_count'][0])
1287 # Check for "rental" videos
1288 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1289 raise ExtractorError('"rental" videos not supported')
1291 # Start extracting information
1292 self.report_information_extraction(video_id)
1295 if 'author' not in video_info:
1296 raise ExtractorError('Unable to extract uploader name')
1297 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1300 video_uploader_id = None
1301 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1302 if mobj is not None:
1303 video_uploader_id = mobj.group(1)
1305 self._downloader.report_warning('unable to extract uploader nickname')
1308 # We try first to get a high quality image:
1309 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1310 video_webpage, re.DOTALL)
1311 if m_thumb is not None:
1312 video_thumbnail = m_thumb.group(1)
1313 elif 'thumbnail_url' not in video_info:
1314 self._downloader.report_warning('unable to extract video thumbnail')
1315 video_thumbnail = None
1316 else: # don't panic if we can't find it
1317 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1320 upload_date = self._html_search_meta(
1321 'datePublished', video_webpage, 'upload date', default=None)
1323 upload_date = self._search_regex(
1324 [r'(?s)id="eow-date.*?>(.*?)</span>',
1325 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1326 video_webpage, 'upload date', default=None)
1328 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1329 upload_date = unified_strdate(upload_date)
1331 m_music = re.search(
1332 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1335 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1336 video_creator = clean_html(m_music.group('creator'))
1338 video_alt_title = video_creator = None
1340 m_cat_container = self._search_regex(
1341 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1342 video_webpage, 'categories', default=None)
1344 category = self._html_search_regex(
1345 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1347 video_categories = None if category is None else [category]
1349 video_categories = None
1352 unescapeHTML(m.group('content'))
1353 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1355 def _extract_count(count_name):
1356 return str_to_int(self._search_regex(
1357 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1358 % re.escape(count_name),
1359 video_webpage, count_name, default=None))
1361 like_count = _extract_count('like')
1362 dislike_count = _extract_count('dislike')
1365 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1366 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1368 if 'length_seconds' not in video_info:
1369 self._downloader.report_warning('unable to extract video duration')
1370 video_duration = None
1372 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1375 video_annotations = None
1376 if self._downloader.params.get('writeannotations', False):
1377 video_annotations = self._extract_annotations(video_id)
1379 def _map_to_format_list(urlmap):
1381 for itag, video_real_url in urlmap.items():
1384 'url': video_real_url,
1385 'player_url': player_url,
1387 if itag in self._formats:
1388 dct.update(self._formats[itag])
1392 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1393 self.report_rtmp_download()
1395 'format_id': '_rtmp',
1397 'url': video_info['conn'][0],
1398 'player_url': player_url,
1400 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1401 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1402 if 'rtmpe%3Dyes' in encoded_url_map:
1403 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1405 for url_data_str in encoded_url_map.split(','):
1406 url_data = compat_parse_qs(url_data_str)
1407 if 'itag' not in url_data or 'url' not in url_data:
1409 format_id = url_data['itag'][0]
1410 url = url_data['url'][0]
1412 if 'sig' in url_data:
1413 url += '&signature=' + url_data['sig'][0]
1414 elif 's' in url_data:
1415 encrypted_sig = url_data['s'][0]
1416 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1418 jsplayer_url_json = self._search_regex(
1420 embed_webpage if age_gate else video_webpage,
1421 'JS player URL (1)', default=None)
1422 if not jsplayer_url_json and not age_gate:
1423 # We need the embed website after all
1424 if embed_webpage is None:
1425 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1426 embed_webpage = self._download_webpage(
1427 embed_url, video_id, 'Downloading embed webpage')
1428 jsplayer_url_json = self._search_regex(
1429 ASSETS_RE, embed_webpage, 'JS player URL')
1431 player_url = json.loads(jsplayer_url_json)
1432 if player_url is None:
1433 player_url_json = self._search_regex(
1434 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1435 video_webpage, 'age gate player URL')
1436 player_url = json.loads(player_url_json)
1438 if self._downloader.params.get('verbose'):
1439 if player_url is None:
1440 player_version = 'unknown'
1441 player_desc = 'unknown'
1443 if player_url.endswith('swf'):
1444 player_version = self._search_regex(
1445 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1446 'flash player', fatal=False)
1447 player_desc = 'flash player %s' % player_version
1449 player_version = self._search_regex(
1450 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1452 'html5 player', fatal=False)
1453 player_desc = 'html5 player %s' % player_version
1455 parts_sizes = self._signature_cache_id(encrypted_sig)
1456 self.to_screen('{%s} signature length %s, %s' %
1457 (format_id, parts_sizes, player_desc))
1459 signature = self._decrypt_signature(
1460 encrypted_sig, video_id, player_url, age_gate)
1461 url += '&signature=' + signature
1462 if 'ratebypass' not in url:
1463 url += '&ratebypass=yes'
1466 'format_id': format_id,
1468 'player_url': player_url,
1470 if format_id in self._formats:
1471 dct.update(self._formats[format_id])
1473 # Some itags are not included in DASH manifest thus corresponding formats will
1474 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1475 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1476 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1477 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1480 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1481 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1484 'fps': int_or_none(url_data.get('fps', [None])[0]),
1485 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1487 for key, value in more_fields.items():
1490 type_ = url_data.get('type', [None])[0]
1492 type_split = type_.split(';')
1493 kind_ext = type_split[0].split('/')
1494 if len(kind_ext) == 2:
1496 dct['ext'] = mimetype2ext(type_split[0])
1497 if kind in ('audio', 'video'):
1499 for mobj in re.finditer(
1500 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1501 if mobj.group('key') == 'codecs':
1502 codecs = mobj.group('val')
1505 codecs = codecs.split(',')
1506 if len(codecs) == 2:
1507 acodec, vcodec = codecs[1], codecs[0]
1509 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1515 elif video_info.get('hlsvp'):
1516 manifest_url = video_info['hlsvp'][0]
1517 url_map = self._extract_from_m3u8(manifest_url, video_id)
1518 formats = _map_to_format_list(url_map)
1519 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1520 for a_format in formats:
1521 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1523 unavailable_message = self._html_search_regex(
1524 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1525 video_webpage, 'unavailable message', default=None)
1526 if unavailable_message:
1527 raise ExtractorError(unavailable_message, expected=True)
1528 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1530 # Look for the DASH manifest
1531 if self._downloader.params.get('youtube_include_dash_manifest', True):
1532 dash_mpd_fatal = True
1533 for dash_manifest_url in dash_mpds:
1536 for df in self._parse_dash_manifest(
1537 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1538 # Do not overwrite DASH format found in some previous DASH manifest
1539 if df['format_id'] not in dash_formats:
1540 dash_formats[df['format_id']] = df
1541 # Additional DASH manifests may end up in HTTP Error 403 therefore
1542 # allow them to fail without bug report message if we already have
1543 # some DASH manifest succeeded. This is temporary workaround to reduce
1544 # burst of bug reports until we figure out the reason and whether it
1545 # can be fixed at all.
1546 dash_mpd_fatal = False
1547 except (ExtractorError, KeyError) as e:
1548 self.report_warning(
1549 'Skipping DASH manifest: %r' % e, video_id)
1551 # Remove the formats we found through non-DASH, they
1552 # contain less info and it can be wrong, because we use
1553 # fixed values (for example the resolution). See
1554 # https://github.com/rg3/youtube-dl/issues/5774 for an
1556 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1557 formats.extend(dash_formats.values())
1559 # Check for malformed aspect ratio
1560 stretched_m = re.search(
1561 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1564 w = float(stretched_m.group('w'))
1565 h = float(stretched_m.group('h'))
1566 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1567 # We will only process correct ratios.
1571 if f.get('vcodec') != 'none':
1572 f['stretched_ratio'] = ratio
1574 self._sort_formats(formats)
1578 'uploader': video_uploader,
1579 'uploader_id': video_uploader_id,
1580 'upload_date': upload_date,
1581 'creator': video_creator,
1582 'title': video_title,
1583 'alt_title': video_alt_title,
1584 'thumbnail': video_thumbnail,
1585 'description': video_description,
1586 'categories': video_categories,
1588 'subtitles': video_subtitles,
1589 'automatic_captions': automatic_captions,
1590 'duration': video_duration,
1591 'age_limit': 18 if age_gate else 0,
1592 'annotations': video_annotations,
1593 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1594 'view_count': view_count,
1595 'like_count': like_count,
1596 'dislike_count': dislike_count,
1597 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1600 'start_time': start_time,
1601 'end_time': end_time,
1605 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1606 IE_DESC = 'YouTube.com playlists'
1607 _VALID_URL = r"""(?x)(?:
1612 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1613 \? (?:.*?[&;])*? (?:p|a|list)=
1617 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1618 # Top tracks, they can also include dots
1623 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1625 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1626 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1627 IE_NAME = 'youtube:playlist'
1629 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1631 'title': 'ytdl test PL',
1632 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1634 'playlist_count': 3,
1636 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1638 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1639 'title': 'YDL_Empty_List',
1641 'playlist_count': 0,
1643 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1644 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1646 'title': '29C3: Not my department',
1647 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1649 'playlist_count': 95,
1651 'note': 'issue #673',
1652 'url': 'PLBB231211A4F62143',
1654 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1655 'id': 'PLBB231211A4F62143',
1657 'playlist_mincount': 26,
1659 'note': 'Large playlist',
1660 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1662 'title': 'Uploads from Cauchemar',
1663 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1665 'playlist_mincount': 799,
1667 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1669 'title': 'YDL_safe_search',
1670 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1672 'playlist_count': 2,
1675 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1676 'playlist_count': 4,
1679 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1682 'note': 'Embedded SWF player',
1683 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1684 'playlist_count': 4,
1687 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1690 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1691 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1693 'title': 'Uploads from Interstellar Movie',
1694 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1696 'playlist_mincout': 21,
1699 def _real_initialize(self):
1702 def _extract_mix(self, playlist_id):
1703 # The mixes are generated from a single video
1704 # the id of the playlist is just 'RD' + video_id
1705 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1706 webpage = self._download_webpage(
1707 url, playlist_id, 'Downloading Youtube mix')
1708 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1710 search_title('playlist-title') or
1711 search_title('title long-title') or
1712 search_title('title'))
1713 title = clean_html(title_span)
1714 ids = orderedSet(re.findall(
1715 r'''(?xs)data-video-username=".*?".*?
1716 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1718 url_results = self._ids_to_results(ids)
1720 return self.playlist_result(url_results, playlist_id, title)
1722 def _extract_playlist(self, playlist_id):
1723 url = self._TEMPLATE_URL % playlist_id
1724 page = self._download_webpage(url, playlist_id)
1726 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1727 match = match.strip()
1728 # Check if the playlist exists or is private
1729 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1730 raise ExtractorError(
1731 'The playlist doesn\'t exist or is private, use --username or '
1732 '--netrc to access it.',
1734 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1735 raise ExtractorError(
1736 'Invalid parameters. Maybe URL is incorrect.',
1738 elif re.match(r'[^<]*Choose your language[^<]*', match):
1741 self.report_warning('Youtube gives an alert message: ' + match)
1743 playlist_title = self._html_search_regex(
1744 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1747 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1749 def _real_extract(self, url):
1750 # Extract playlist id
1751 mobj = re.match(self._VALID_URL, url)
1753 raise ExtractorError('Invalid URL: %s' % url)
1754 playlist_id = mobj.group(1) or mobj.group(2)
1756 # Check if it's a video-specific URL
1757 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1758 if 'v' in query_dict:
1759 video_id = query_dict['v'][0]
1760 if self._downloader.params.get('noplaylist'):
1761 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1762 return self.url_result(video_id, 'Youtube', video_id=video_id)
1764 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1766 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1767 # Mixes require a custom extraction process
1768 return self._extract_mix(playlist_id)
1770 return self._extract_playlist(playlist_id)
1773 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1774 IE_DESC = 'YouTube.com channels'
1775 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1776 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1777 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1778 IE_NAME = 'youtube:channel'
1780 'note': 'paginated channel',
1781 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1782 'playlist_mincount': 91,
1784 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1785 'title': 'Uploads from lex will',
1788 'note': 'Age restricted channel',
1789 # from https://www.youtube.com/user/DeusExOfficial
1790 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1791 'playlist_mincount': 64,
1793 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1794 'title': 'Uploads from Deus Ex',
1799 def suitable(cls, url):
1800 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1802 def _real_extract(self, url):
1803 channel_id = self._match_id(url)
1805 url = self._TEMPLATE_URL % channel_id
1807 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1808 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1809 # otherwise fallback on channel by page extraction
1810 channel_page = self._download_webpage(
1811 url + '?view=57', channel_id,
1812 'Downloading channel page', fatal=False)
1813 if channel_page is False:
1814 channel_playlist_id = False
1816 channel_playlist_id = self._html_search_meta(
1817 'channelId', channel_page, 'channel id', default=None)
1818 if not channel_playlist_id:
1819 channel_playlist_id = self._search_regex(
1820 r'data-(?:channel-external-|yt)id="([^"]+)"',
1821 channel_page, 'channel id', default=None)
1822 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1823 playlist_id = 'UU' + channel_playlist_id[2:]
1824 return self.url_result(
1825 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1827 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1828 autogenerated = re.search(r'''(?x)
1830 channel-header-autogenerated-label|
1831 yt-channel-title-autogenerated
1832 )[^"]*"''', channel_page) is not None
1835 # The videos are contained in a single page
1836 # the ajax pages can't be used, they are empty
1839 video_id, 'Youtube', video_id=video_id,
1840 video_title=video_title)
1841 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1842 return self.playlist_result(entries, channel_id)
1844 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1847 class YoutubeUserIE(YoutubeChannelIE):
1848 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1849 _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1850 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1851 IE_NAME = 'youtube:user'
1854 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1855 'playlist_mincount': 320,
1857 'title': 'TheLinuxFoundation',
1860 'url': 'ytuser:phihag',
1861 'only_matching': True,
1865 def suitable(cls, url):
1866 # Don't return True if the url can be extracted with other youtube
1867 # extractor, the regex would is too permissive and it would match.
1868 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1869 if any(ie.suitable(url) for ie in other_ies):
1872 return super(YoutubeUserIE, cls).suitable(url)
1875 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1876 IE_DESC = 'YouTube.com user/channel playlists'
1877 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1878 IE_NAME = 'youtube:playlists'
1881 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1882 'playlist_mincount': 4,
1884 'id': 'ThirstForScience',
1885 'title': 'Thirst for Science',
1888 # with "Load more" button
1889 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1890 'playlist_mincount': 70,
1893 'title': 'Игорь Клейнер',
1896 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1897 'playlist_mincount': 17,
1899 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1900 'title': 'Chem Player',
1905 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1906 IE_DESC = 'YouTube.com searches'
1907 # there doesn't appear to be a real limit, for example if you search for
1908 # 'python' you get more than 8.000.000 results
1909 _MAX_RESULTS = float('inf')
1910 IE_NAME = 'youtube:search'
1911 _SEARCH_KEY = 'ytsearch'
1912 _EXTRA_QUERY_ARGS = {}
1915 def _get_n_results(self, query, n):
1916 """Get a specified number of results for a query"""
1921 for pagenum in itertools.count(1):
1923 'search_query': query.encode('utf-8'),
1927 url_query.update(self._EXTRA_QUERY_ARGS)
1928 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1929 data = self._download_json(
1930 result_url, video_id='query "%s"' % query,
1931 note='Downloading page %s' % pagenum,
1932 errnote='Unable to download API page')
1933 html_content = data[1]['body']['content']
1935 if 'class="search-message' in html_content:
1936 raise ExtractorError(
1937 '[youtube] No video results', expected=True)
1939 new_videos = self._ids_to_results(orderedSet(re.findall(
1940 r'href="/watch\?v=(.{11})', html_content)))
1941 videos += new_videos
1942 if not new_videos or len(videos) > limit:
1947 return self.playlist_result(videos, query)
1950 class YoutubeSearchDateIE(YoutubeSearchIE):
1951 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1952 _SEARCH_KEY = 'ytsearchdate'
1953 IE_DESC = 'YouTube.com searches, newest videos first'
1954 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1957 class YoutubeSearchURLIE(InfoExtractor):
1958 IE_DESC = 'YouTube.com search URLs'
1959 IE_NAME = 'youtube:search_url'
1960 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1962 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1963 'playlist_mincount': 5,
1965 'title': 'youtube-dl test video',
1969 def _real_extract(self, url):
1970 mobj = re.match(self._VALID_URL, url)
1971 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1973 webpage = self._download_webpage(url, query)
1974 result_code = self._search_regex(
1975 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1977 part_codes = re.findall(
1978 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1980 for part_code in part_codes:
1981 part_title = self._html_search_regex(
1982 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1983 part_url_snippet = self._html_search_regex(
1984 r'(?s)href="([^"]+)"', part_code, 'item URL')
1985 part_url = compat_urlparse.urljoin(
1986 'https://www.youtube.com/', part_url_snippet)
1990 'title': part_title,
1994 '_type': 'playlist',
2000 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2001 IE_DESC = 'YouTube.com (multi-season) shows'
2002 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2003 IE_NAME = 'youtube:show'
2005 'url': 'https://www.youtube.com/show/airdisasters',
2006 'playlist_mincount': 5,
2008 'id': 'airdisasters',
2009 'title': 'Air Disasters',
2013 def _real_extract(self, url):
2014 playlist_id = self._match_id(url)
2015 return super(YoutubeShowIE, self)._real_extract(
2016 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2019 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2021 Base class for feed extractors
2022 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2024 _LOGIN_REQUIRED = True
2028 return 'youtube:%s' % self._FEED_NAME
2030 def _real_initialize(self):
2033 def _real_extract(self, url):
2034 page = self._download_webpage(
2035 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2037 # The extraction process is the same as for playlists, but the regex
2038 # for the video ids doesn't contain an index
2040 more_widget_html = content_html = page
2041 for page_num in itertools.count(1):
2042 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2044 # 'recommended' feed has infinite 'load more' and each new portion spins
2045 # the same videos in (sometimes) slightly different order, so we'll check
2046 # for unicity and break when portion has no new videos
2047 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2053 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2057 more = self._download_json(
2058 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2059 'Downloading page #%s' % page_num,
2060 transform_source=uppercase_escape)
2061 content_html = more['content_html']
2062 more_widget_html = more['load_more_widget_html']
2064 return self.playlist_result(
2065 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2068 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2069 IE_NAME = 'youtube:watchlater'
2070 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2071 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2073 _TESTS = [] # override PlaylistIE tests
2075 def _real_extract(self, url):
2076 return self._extract_playlist('WL')
2079 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2080 IE_NAME = 'youtube:favorites'
2081 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2082 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2083 _LOGIN_REQUIRED = True
2085 def _real_extract(self, url):
2086 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2087 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2088 return self.url_result(playlist_id, 'YoutubePlaylist')
2091 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2092 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2093 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2094 _FEED_NAME = 'recommended'
2095 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2098 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2099 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2100 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2101 _FEED_NAME = 'subscriptions'
2102 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2105 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2106 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2107 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2108 _FEED_NAME = 'history'
2109 _PLAYLIST_TITLE = 'Youtube History'
2112 class YoutubeTruncatedURLIE(InfoExtractor):
2113 IE_NAME = 'youtube:truncated_url'
2114 IE_DESC = False # Do not list
2115 _VALID_URL = r'''(?x)
2117 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2120 annotation_id=annotation_[^&]+|
2126 attribution_link\?a=[^&]+
2132 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2133 'only_matching': True,
2135 'url': 'http://www.youtube.com/watch?',
2136 'only_matching': True,
2138 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2139 'only_matching': True,
2141 'url': 'https://www.youtube.com/watch?feature=foo',
2142 'only_matching': True,
2144 'url': 'https://www.youtube.com/watch?hl=en-GB',
2145 'only_matching': True,
2147 'url': 'https://www.youtube.com/watch?t=2372',
2148 'only_matching': True,
2151 def _real_extract(self, url):
2152 raise ExtractorError(
2153 'Did you forget to quote the URL? Remember that & is a meta '
2154 'character in most shells, so you want to put the URL in quotes, '
2156 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2157 ' or simply youtube-dl BaW_jenozKc .',
2161 class YoutubeTruncatedIDIE(InfoExtractor):
2162 IE_NAME = 'youtube:truncated_id'
2163 IE_DESC = False # Do not list
2164 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2167 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2168 'only_matching': True,
2171 def _real_extract(self, url):
2172 video_id = self._match_id(url)
2173 raise ExtractorError(
2174 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),